Showing 1–50 of 167 results for author: Chiaberge, M

Learning Social Cost Functions for Human-Aware Path Planning

Authors: Andrea Eirale, Matteo Leonetti, Marcello Chiaberge

Abstract: Achieving social acceptance is one of the main goals of Social Robotic Navigation. Despite this topic has received increasing interest in recent years, most of the research has focused on driving the robotic agent along obstacle-free trajectories, planning around estimates of future human motion to respect personal distances and optimize navigation. However, social interactions in everyday life ar… ▽ More Achieving social acceptance is one of the main goals of Social Robotic Navigation. Despite this topic has received increasing interest in recent years, most of the research has focused on driving the robotic agent along obstacle-free trajectories, planning around estimates of future human motion to respect personal distances and optimize navigation. However, social interactions in everyday life are also dictated by norms that do not strictly depend on movement, such as when standing at the end of a queue rather than cutting it. In this paper, we propose a novel method to recognize common social scenarios and modify a traditional planner's cost function to adapt to them. This solution enables the robot to carry out different social navigation behaviors that would not arise otherwise, maintaining the robustness of traditional navigation. Our approach allows the robot to learn different social norms with a single learned model, rather than having different modules for each task. As a proof of concept, we consider the tasks of queuing and respect interaction spaces of groups of people talking to one another, but the method can be extended to other human activities that do not involve motion. △ Less

Submitted 15 July, 2024; originally announced July 2024.

A Self-Supervised Task for Fault Detection in Satellite Multivariate Time Series

Authors: Carlo Cena, Silvia Bucci, Alessandro Balossino, Marcello Chiaberge

Abstract: In the space sector, due to environmental conditions and restricted accessibility, robust fault detection methods are imperative for ensuring mission success and safeguarding valuable assets. This work proposes a novel approach leveraging Physics-Informed Real NVP neural networks, renowned for their ability to model complex and high-dimensional distributions, augmented with a self-supervised task… ▽ More In the space sector, due to environmental conditions and restricted accessibility, robust fault detection methods are imperative for ensuring mission success and safeguarding valuable assets. This work proposes a novel approach leveraging Physics-Informed Real NVP neural networks, renowned for their ability to model complex and high-dimensional distributions, augmented with a self-supervised task based on sensors' data permutation. It focuses on enhancing fault detection within the satellite multivariate time series. The experiments involve various configurations, including pre-training with self-supervision, multi-task learning, and standalone self-supervised training. Results indicate significant performance improvements across all settings. In particular, employing only the self-supervised loss yields the best overall results, suggesting its efficacy in guiding the network to extract relevant features for fault detection. This study presents a promising direction for improving fault detection in space systems and warrants further exploration in other datasets and applications. △ Less

Submitted 3 July, 2024; originally announced July 2024.

Comments: SPAICE: AI in and for Space, 2024

Physics-Informed Real NVP for Satellite Power System Fault Detection

Authors: Carlo Cena, Umberto Albertin, Mauro Martini, Silvia Bucci, Marcello Chiaberge

Abstract: The unique challenges posed by the space environment, characterized by extreme conditions and limited accessibility, raise the need for robust and reliable techniques to identify and prevent satellite faults. Fault detection methods in the space sector are required to ensure mission success and to protect valuable assets. In this context, this paper proposes an Artificial Intelligence (AI) based f… ▽ More The unique challenges posed by the space environment, characterized by extreme conditions and limited accessibility, raise the need for robust and reliable techniques to identify and prevent satellite faults. Fault detection methods in the space sector are required to ensure mission success and to protect valuable assets. In this context, this paper proposes an Artificial Intelligence (AI) based fault detection methodology and evaluates its performance on ADAPT (Advanced Diagnostics and Prognostics Testbed), an Electrical Power System (EPS) dataset, crafted in laboratory by NASA. Our study focuses on the application of a physics-informed (PI) real-valued non-volume preserving (Real NVP) model for fault detection in space systems. The efficacy of this method is systematically compared against other AI approaches such as Gated Recurrent Unit (GRU) and Autoencoder-based techniques. Results show that our physics-informed approach outperforms existing methods of fault detection, demonstrating its suitability for addressing the unique challenges of satellite EPS sub-system faults. Furthermore, we unveil the competitive advantage of physics-informed loss in AI models to address specific space needs, namely robustness, reliability, and power constraints, crucial for space exploration and satellite missions. △ Less

Submitted 27 May, 2024; originally announced May 2024.

Comments: Accepted at International Conference on Advanced Intelligent Mechatronics (AIM) 2024

Adaptive Social Force Window Planner with Reinforcement Learning

Authors: Mauro Martini, Noé Pérez-Higueras, Andrea Ostuni, Marcello Chiaberge, Fernando Caballero, Luis Merino

Abstract: Human-aware navigation is a complex task for mobile robots, requiring an autonomous navigation system capable of achieving efficient path planning together with socially compliant behaviors. Social planners usually add costs or constraints to the objective function, leading to intricate tuning processes or tailoring the solution to the specific social scenario. Machine Learning can enhance planner… ▽ More Human-aware navigation is a complex task for mobile robots, requiring an autonomous navigation system capable of achieving efficient path planning together with socially compliant behaviors. Social planners usually add costs or constraints to the objective function, leading to intricate tuning processes or tailoring the solution to the specific social scenario. Machine Learning can enhance planners' versatility and help them learn complex social behaviors from data. This work proposes an adaptive social planner, using a Deep Reinforcement Learning agent to dynamically adjust the weighting parameters of the cost function used to evaluate trajectories. The resulting planner combines the robustness of the classic Dynamic Window Approach, integrated with a social cost based on the Social Force Model, and the flexibility of learning methods to boost the overall performance on social navigation tasks. Our extensive experimentation on different environments demonstrates the general advantage of the proposed method over static cost planners. △ Less

Submitted 21 April, 2024; originally announced April 2024.

Semi-Supervised Novelty Detection for Precise Ultra-Wideband Error Signal Prediction

Authors: Umberto Albertin, Alessandro Navone, Mauro Martini, Marcello Chiaberge

Abstract: Ultra-Wideband (UWB) technology is an emerging low-cost solution for localization in a generic environment. However, UWB signal can be affected by signal reflections and non-line-of-sight (NLoS) conditions between anchors; hence, in a broader sense, the specific geometry of the environment and the disposition of obstructing elements in the map may drastically hinder the reliability of UWB for prec… ▽ More Ultra-Wideband (UWB) technology is an emerging low-cost solution for localization in a generic environment. However, UWB signal can be affected by signal reflections and non-line-of-sight (NLoS) conditions between anchors; hence, in a broader sense, the specific geometry of the environment and the disposition of obstructing elements in the map may drastically hinder the reliability of UWB for precise robot localization. This work aims to mitigate this problem by learning a map-specific characterization of the UWB quality signal with a fingerprint semi-supervised novelty detection methodology. An unsupervised autoencoder neural network is trained on nominal UWB map conditions, and then it is used to predict errors derived from the introduction of perturbing novelties in the environment. This work poses a step change in the understanding of UWB localization and its reliability in evolving environmental conditions. The resulting performance of the proposed method is proved by fine-grained experiments obtained with a visual tracking ground truth. △ Less

Submitted 8 April, 2024; originally announced April 2024.

Non-linear Model Predictive Control for Multi-task GPS-free Autonomous Navigation in Vineyards

Authors: Matteo Sperti, Marco Ambrosio, Mauro Martini, Alessandro Navone, Andrea Ostuni, Marcello Chiaberge

Abstract: Autonomous navigation is the foundation of agricultural robots. This paper focuses on developing an advanced autonomous navigation system for a rover operating within row-based crops. A position-agnostic system is proposed to address the challenging situation when standard localization methods, like GPS, fail due to unfavorable weather or obstructed signals. This breakthrough is especially vital i… ▽ More Autonomous navigation is the foundation of agricultural robots. This paper focuses on developing an advanced autonomous navigation system for a rover operating within row-based crops. A position-agnostic system is proposed to address the challenging situation when standard localization methods, like GPS, fail due to unfavorable weather or obstructed signals. This breakthrough is especially vital in densely vegetated regions, including areas covered by thick tree canopies or pergola vineyards. This work proposed a novel system that leverages a single RGB-D camera and a Non-linear Model Predictive Control strategy to navigate through entire rows, adapting to various crop spacing. The presented solution demonstrates versatility in handling diverse crop densities, environmental factors, and multiple navigation tasks to support agricultural activities at an extremely cost-effective implementation. Experimental validation in simulated and real vineyards underscores the system's robustness and competitiveness in both standard row traversal and target objects approach. △ Less

Submitted 8 April, 2024; originally announced April 2024.

GPS-free Autonomous Navigation in Cluttered Tree Rows with Deep Semantic Segmentation

Authors: Alessandro Navone, Mauro Martini, Marco Ambrosio, Andrea Ostuni, Simone Angarano, Marcello Chiaberge

Abstract: Segmentation-based autonomous navigation has recently been presented as an appealing approach to guiding robotic platforms through crop rows without requiring perfect GPS localization. Nevertheless, current techniques are restricted to situations where the distinct separation between the plants and the sky allows for the identification of the row's center. However, tall, dense vegetation, such as… ▽ More Segmentation-based autonomous navigation has recently been presented as an appealing approach to guiding robotic platforms through crop rows without requiring perfect GPS localization. Nevertheless, current techniques are restricted to situations where the distinct separation between the plants and the sky allows for the identification of the row's center. However, tall, dense vegetation, such as high tree rows and orchards, is the primary cause of GPS signal blockage. In this study, we increase the overall robustness and adaptability of the control algorithm by extending the segmentation-based robotic guiding to those cases where canopies and branches occlude the sky and prevent the utilization of GPS and earlier approaches. An efficient Deep Neural Network architecture has been used to address semantic segmentation, performing the training with synthetic data only. Numerous vineyards and tree fields have undergone extensive testing in both simulation and real-world to show the solution's competitive benefits. △ Less

Submitted 8 April, 2024; originally announced April 2024.

Comments: arXiv admin note: text overlap with arXiv:2304.08988

Imaging Spectropolarimetry -- A New Observing Mode on the Hubble Space Telescope's Advanced Camera for Surveys

Authors: Nimish P. Hathi, Dean C. Hines, Yotam Cohen, Norman A. Grogin, Marco Chiaberge

Abstract: Imaging spectropolarimetry is a new observing mode on the Advanced Camera for Surveys (ACS) aboard the Hubble Space Telescope (HST) that was commissioned in Cycle 30 and is available to HST observers starting in Cycle 31 (i.e., from 2023). It is a technique that is accessible from ground-based observatories, but the superb spatial resolution afforded by HST/ACS combined with the slitless nature of… ▽ More Imaging spectropolarimetry is a new observing mode on the Advanced Camera for Surveys (ACS) aboard the Hubble Space Telescope (HST) that was commissioned in Cycle 30 and is available to HST observers starting in Cycle 31 (i.e., from 2023). It is a technique that is accessible from ground-based observatories, but the superb spatial resolution afforded by HST/ACS combined with the slitless nature of HST/ACS grism spectroscopy opens up the possibility of studying polarized extended emission in a way that is not currently possible even with Adaptive Optics facilities on the ground. This mode could help to study interesting targets including (but not limited to) QSOs, AGN and Radio Galaxies, ISM Dust Properties, Pre-Planetary Nebulae, Proto-Planetary and Debris Disks, Supernovae/Supernova Remnants, and Solar System objects. This research note presents the preliminary results from the calibration programs used to calibrate imaging spectropolarimetry on HST/ACS. △ Less

Submitted 26 February, 2024; originally announced February 2024.

Comments: To be published in Research Notes of the AAS (RNAAS), 6 pages, 1 figure

Heavily Obscured AGN detection: a Radio vs X-ray challenge

Authors: Giovanni Mazzolari, Roberto Gilli, Marcella Brusa, Marco Mignoli, Fabio Vito, Isabella Prandoni, Stefano Marchesi, Marco Chiaberge, Giorgio Lanzuisi, Quirino D'Amato, Andrea Comastri, Cristian Vignali, Kazushi Iwasawa, Colin Norman

Abstract: In this work, we study the AGN radio detection effectiveness in the major deep extragalactic surveys, considering different AGN obscuration levels, redshift, and AGN bolometric luminosities. We particularly focus on comparing their radio and X-ray detectability, making predictions for present and future radio surveys. We extrapolate the predictions of AGN population synthesis model of cosmic X-ray… ▽ More In this work, we study the AGN radio detection effectiveness in the major deep extragalactic surveys, considering different AGN obscuration levels, redshift, and AGN bolometric luminosities. We particularly focus on comparing their radio and X-ray detectability, making predictions for present and future radio surveys. We extrapolate the predictions of AGN population synthesis model of cosmic X-ray background (CXB) to the radio band, by deriving the 1.4 GHz luminosity functions of unobscured (i.e. with hydrogen column densities $\log N_{H} <22$), obscured ($22<\log N_{H}<24$) and Compton-thick (CTK, $\log N_{H} >24$) AGN. We then use these functions to forecast the number of detectable AGN based on the area, flux limit, and completeness of a given radio survey and compare it with the AGN number resulting from X-ray predictions. When applied to deep extragalactic fields covered both by radio and X-ray observations, we show that, while X-ray selection is generally more effective in detecting unobscured AGN, the surface density of CTK AGN radio detected is on average $\sim 10$ times larger than the X-ray one, and even greater at high redshifts, considering the current surveys and facilities. Our results suggest that thousands of CTK AGN are already present in current radio catalogs, but most of them escaped any detection in the corresponding X-ray observations. We also present expectations for the number of AGN to be detected by the Square Kilometer Array Observatory (SKAO) in its future deep and wide radio continuum surveys, finding that it will be able to detect more than 2000 AGN at $z>6$ and some tens at $z>10$, more than half of which are expected to be CTK. △ Less

Submitted 31 January, 2024; originally announced February 2024.

Comments: 17 pages, 11 figures, accepted for publication on A&A

Journal ref: A&A 687, A120 (2024)

Constraints on the narrow-line region of the X-ray quasi-periodic eruption source GSN 069

Authors: Kishore C. Patra, Wenbin Lu, Yilun Ma, Eliot Quataert, Giovanni Miniutti, Marco Chiaberge, Alexei V. Filippenko

Abstract: The origins of quasi-periodic eruptions (QPEs) are poorly understood, although most theoretical explanations invoke an accretion disk around a supermassive black hole. The gas and stellar environments in the galactic nuclei of these sources are also poorly constrained. In this paper, we present an analysis of archival Hubble Space Telescope (HST) images to study the narrow-line [O III] emission in… ▽ More The origins of quasi-periodic eruptions (QPEs) are poorly understood, although most theoretical explanations invoke an accretion disk around a supermassive black hole. The gas and stellar environments in the galactic nuclei of these sources are also poorly constrained. In this paper, we present an analysis of archival Hubble Space Telescope (HST) images to study the narrow-line [O III] emission in the QPE source GSN 069. We find strong evidence for a compact nuclear [O III] emission region of size $\lesssim 35$ pc, overlaid on top of extended [O III] emission up to 2 kpc away from the nucleus. The age of the accretion system is estimated to be between 10 and 100 yr. The [O III] luminosity of the compact region was measured to be $(2.1 \pm 0.3) \times 10^{40}\,\rm erg\,s^{-1}$. Based on CLOUDY simulations, we constrain that the [O III] emitting gas has a hydrogen number density in the range $5 \times 10^{3} < n_{\rm H} \lesssim 10^{8}\,\rm cm^{-3}$ and volume filling factor $f_{\rm V} < 2 \times 10^{-3}$. We suggest that the dense gas in the nuclear region of GSN 069 originates from molecular clouds (with total mass $\gtrsim 3 \times 10^{3}\,M_{\odot}$), which are freshly ionised by the soft X-ray photons from the accretion disk. We predict possible evolution of the compact narrow-line region on emission-line diagnostic diagrams, and hence future HST or integral-field unit observations can be used to further pin down the age of this puzzling system. △ Less

Submitted 9 October, 2023; originally announced October 2023.

Comments: 10 pages, 7 figures. Comments welcome

Lavender Autonomous Navigation with Semantic Segmentation at the Edge

Authors: Alessandro Navone, Fabrizio Romanelli, Marco Ambrosio, Mauro Martini, Simone Angarano, Marcello Chiaberge

Abstract: Achieving success in agricultural activities heavily relies on precise navigation in row crop fields. Recently, segmentation-based navigation has emerged as a reliable technique when GPS-based localization is unavailable or higher accuracy is needed due to vegetation or unfavorable weather conditions. It also comes in handy when plants are growing rapidly and require an online adaptation of the na… ▽ More Achieving success in agricultural activities heavily relies on precise navigation in row crop fields. Recently, segmentation-based navigation has emerged as a reliable technique when GPS-based localization is unavailable or higher accuracy is needed due to vegetation or unfavorable weather conditions. It also comes in handy when plants are growing rapidly and require an online adaptation of the navigation algorithm. This work applies a segmentation-based visual agnostic navigation algorithm to lavender fields, considering both simulation and real-world scenarios. The effectiveness of this approach is validated through a wide set of experimental tests, which show the capability of the proposed solution to generalize over different scenarios and provide highly-reliable results. △ Less

Submitted 13 September, 2023; originally announced September 2023.

The puzzling properties of the MACS1149-JD1 galaxy at z=9.11

Authors: Massimo Stiavelli, Takahiro Morishita, Marco Chiaberge, Claudio Grillo, Nicha Leethochawalit, Piero Rosati, Stefan Schuldt, Michele Trenti, Tommaso Treu

Abstract: We analyze new JWST NIRCam and NIRSpec data on the redshift 9.11 galaxy MACS1149-JD1. Our NIRCam imaging data reveal that JD1 comprises three spatially distinct components. Our spectroscopic data indicate that JD1 appears dust-free but is already enriched, $12 + \log {\rm (O/H) } = 7.90^{+0.04}_{-0.05}$. We also find that the Carbon and Neon abundances in JD1 are below the solar abundance ratio. P… ▽ More We analyze new JWST NIRCam and NIRSpec data on the redshift 9.11 galaxy MACS1149-JD1. Our NIRCam imaging data reveal that JD1 comprises three spatially distinct components. Our spectroscopic data indicate that JD1 appears dust-free but is already enriched, $12 + \log {\rm (O/H) } = 7.90^{+0.04}_{-0.05}$. We also find that the Carbon and Neon abundances in JD1 are below the solar abundance ratio. Particularly the Carbon under-abundance is suggestive of recent star formation where Type~II supernovae have already enriched the ISM in Oxygen but intermediate mass stars have not yet enriched the ISM in Carbon. A recent burst of star formation is also revealed by the star formation history derived from NIRCam photometry. Our data do not reveal the presence of a significant amount of old populations, resulting in a factor of $\sim7\times$ smaller stellar mass than previous estimates. Thus, our data support the view that JD1 is a young galaxy. △ Less

Submitted 13 October, 2023; v1 submitted 28 August, 2023; originally announced August 2023.

Comments: Accepted for publication

Uncovering a Massive z~7.7 Galaxy Hosting a Heavily Obscured Radio-Loud QSO Candidate in COSMOS-Web

Authors: Erini Lambrides, Marco Chiaberge, Arianna Long, Daizhong Liu, Hollis B. Akins, Andrew F. Ptak, Irham Taufik Andika, Alessandro Capetti, Caitlin M. Casey, Jaclyn B. Champagne, Katherine Chworowsky, Tracy E. Clarke, Olivia R. Cooper, Xuheng Ding, Dillon Z. Dong, Andreas L. Faisst, Jordan Y. Forman, Maximilien Franco, Steven Gillman, Ghassem Gozaliasl, Kirsten R. Hall, Santosh Harish, Christopher C. Hayward, Michaela Hirschmann, Taylor A. Hutchison , et al. (25 additional authors not shown)

Abstract: In this letter, we report the discovery of the highest redshift, heavily obscured, radio-loud AGN candidate selected using JWST NIRCam/MIRI, mid-IR, sub-mm, and radio imaging in the COSMOS-Web field. Using multi-frequency radio observations and mid-IR photometry, we identify a powerful, radio-loud (RL), growing supermassive black hole (SMBH) with significant spectral steepening of the radio SED (… ▽ More In this letter, we report the discovery of the highest redshift, heavily obscured, radio-loud AGN candidate selected using JWST NIRCam/MIRI, mid-IR, sub-mm, and radio imaging in the COSMOS-Web field. Using multi-frequency radio observations and mid-IR photometry, we identify a powerful, radio-loud (RL), growing supermassive black hole (SMBH) with significant spectral steepening of the radio SED ($f_{1.28 \mathrm{GHz}} \sim 2$ mJy, $q_{24μm} = -1.1$, $α_{1.28-3\mathrm{GHz}}=-1.2$, $Δα= -0.4$). In conjunction with ALMA, deep ground-based observations, ancillary space-based data, and the unprecedented resolution and sensitivity of JWST, we find no evidence of AGN contribution to the UV/optical/NIR data and thus infer heavy amounts of obscuration (N$_{\mathrm{H}} > 10^{23}$ cm$^{-2}$). Using the wealth of deep UV to sub-mm photometric data, we report a singular solution photo-z of $z_\mathrm{phot}$ = 7.7$^{+0.4}_{-0.3}$ and estimate an extremely massive host-galaxy ($\log M_{\star} = 11.4 -12\,\mathrm{M}_{\odot}$) hosting a powerful, growing SMBH (L$_{\mathrm{Bol}} = 4-12 \times 10^{46}$ erg s$^{-1}$). This source represents the furthest known obscured RL AGN candidate, and its level of obscuration aligns with the most representative but observationally scarce population of AGN at these epochs. △ Less

Submitted 15 December, 2023; v1 submitted 24 August, 2023; originally announced August 2023.

Comments: Accepted to ApJL

Enhanced Sub-kpc Scale Star-formation: Results From A JWST Size Analysis of 341 Galaxies At 5<z<14

Authors: Takahiro Morishita, Massimo Stiavelli, Ranga-Ram Chary, Michele Trenti, Pietro Bergamini, Marco Chiaberge, Nicha Leethochawalit, Guido Roberts-Borsani, Xuejian Shen, Tommaso Treu

Abstract: We present a comprehensive search and analysis of high-redshift galaxies in a suite of nine public JWST extragalactic fields taken in Cycle 1, covering a total effective search area of $\sim358{\rm arcmin^2}$. Through conservative ($8σ$) photometric selection, we identify 341 galaxies at $5<z<14$, with 109 having spectroscopic redshift measurements from the literature, including recent JWST NIRSpe… ▽ More We present a comprehensive search and analysis of high-redshift galaxies in a suite of nine public JWST extragalactic fields taken in Cycle 1, covering a total effective search area of $\sim358{\rm arcmin^2}$. Through conservative ($8σ$) photometric selection, we identify 341 galaxies at $5<z<14$, with 109 having spectroscopic redshift measurements from the literature, including recent JWST NIRSpec observations. Our regression analysis reveals that the rest-frame UV size-stellar mass relation follows $R_{\rm eff}\propto M_*^{0.19\pm0.03}$, similar to that of star-forming galaxies at $z\sim3$, but scaled down in size by $\sim0.7$dex. We find a much slower rate for the average size evolution over the redshift range, $R_{\rm eff}\propto(1+z)^{-0.4\pm0.2}$, than that derived in the literature. A fraction ($\sim13\,\%$) of our sample are marginally resolved even in the NIRCam imaging ($<100$pc), located at $>1.5\,σ$ below the derived size-mass slope. These compact sources exhibit a high star formation surface density $Σ_{\rm SFR}>10\,M_\odot\,{\rm yr^{-1}\,kpc^{-2}}$, a range in which only $<0.01\,\%$ of the local star-forming galaxy sample is found. For those with available NIRSpec data, no evidence of ongoing supermassive black hole accretion is observed. A potential explanation for the observed high [OIII]-to-Hbeta ratios could be high shock velocities, likely originating within intense star-forming regions characterized by high $Σ_{\rm SFR}$. Lastly, we find that the rest-frame UV and optical sizes of our sample are comparable. Our results are consistent with these early galaxies building up their structures inside-out and yet to exhibit the strong color gradient seen at lower redshift. △ Less

Submitted 12 December, 2023; v1 submitted 9 August, 2023; originally announced August 2023.

Comments: Accepted for publication in ApJ

Enhancing Navigation Benchmarking and Perception Data Generation for Row-based Crops in Simulation

Authors: Mauro Martini, Andrea Eirale, Brenno Tuberga, Marco Ambrosio, Andrea Ostuni, Francesco Messina, Luigi Mazzara, Marcello Chiaberge

Abstract: Service robotics is recently enhancing precision agriculture enabling many automated processes based on efficient autonomous navigation solutions. However, data generation and infield validation campaigns hinder the progress of large-scale autonomous platforms. Simulated environments and deep visual perception are spreading as successful tools to speed up the development of robust navigation with… ▽ More Service robotics is recently enhancing precision agriculture enabling many automated processes based on efficient autonomous navigation solutions. However, data generation and infield validation campaigns hinder the progress of large-scale autonomous platforms. Simulated environments and deep visual perception are spreading as successful tools to speed up the development of robust navigation with low-cost RGB-D cameras. In this context, the contribution of this work is twofold: a synthetic dataset to train deep semantic segmentation networks together with a collection of virtual scenarios for a fast evaluation of navigation algorithms. Moreover, an automatic parametric approach is developed to explore different field geometries and features. The simulation framework and the dataset have been evaluated by training a deep segmentation network on different crops and benchmarking the resulting navigation. △ Less

Submitted 27 June, 2023; originally announced June 2023.

Comments: Accepted at the 14th European Conference on Precision Agriculture (ECPA) 2023

Variability of extragalactic X-ray jets on kiloparsec scales

Authors: Eileen T. Meyer, Aamil Shaik, Yanbo Tang, Nancy Reid, Karthik Reddy, Peter Breiding, Markos Georganopoulos, Marco Chiaberge, Eric Perlman, Devon Clautice, William Sparks, Nat DeNigris, Max Trevor

Abstract: Unexpectedly strong X-ray emission from extragalactic radio jets on kiloparsec scales has been one of the major discoveries of Chandra, the only X-ray observatory capable of sub-arcsecond-scale imaging. The origin of this X-ray emission, which appears as a second spectral component from that of the radio emission, has been debated for over two decades. The most commonly assumed mechanism is invers… ▽ More Unexpectedly strong X-ray emission from extragalactic radio jets on kiloparsec scales has been one of the major discoveries of Chandra, the only X-ray observatory capable of sub-arcsecond-scale imaging. The origin of this X-ray emission, which appears as a second spectral component from that of the radio emission, has been debated for over two decades. The most commonly assumed mechanism is inverse Compton upscattering of the Cosmic Microwave Background (IC-CMB) by very low-energy electrons in a still highly relativistic jet. Under this mechanism, no variability in the X-ray emission is expected. Here we report the detection of X-ray variability in the large-scale jet population, using a novel statistical analysis of 53 jets with multiple Chandra observations. Taken as a population, we find that the distribution of p-values from a Poisson model is strongly inconsistent with steady emission, with a global p-value of 1.96e-4 under a Kolmogorov-Smirnov test against the expected Uniform (0,1) distribution. These results strongly imply that the dominant mechanism of X-ray production in kpc-scale jets is synchrotron emission by a second population of electrons reaching multi-TeV energies. X-ray variability on the time-scale of months to a few years implies extremely small emitting volumes much smaller than the cross-section of the jet. △ Less

Submitted 10 February, 2024; v1 submitted 30 May, 2023; originally announced May 2023.

Comments: Published in Nature Astronomy 29 May 2023; Supplemental Information and Excel File included in 1st submission; Published article in most recent

Powerful Radio-Loud Quasars are Triggered by Galaxy Mergers in the Cosmic Bright Ages

Authors: Peter Breiding, Marco Chiaberge, Erini Lambrides, Eileen T. Meyer, S. P. Willner, Bryan Hilbert, Martin Haas, George Miley, Eric S. Perlman, Peter Barthel, Christopher P. O'Dea, Alessandro Capetti, Belinda Wilkes, Stefi A. Baum, Duccio F. Macchetto, Grant Tremblay, Colin Norman

Abstract: While supermassive black holes are ubiquitous features of galactic nuclei, only a small minority are observed during episodes of luminous accretion. The physical mechanism(s) driving the onset of fueling and ignition in these active galactic nuclei (AGN) are still largely unknown for many galaxies and AGN-selection criteria. Attention has focused on AGN triggering by means of major galaxy mergers… ▽ More While supermassive black holes are ubiquitous features of galactic nuclei, only a small minority are observed during episodes of luminous accretion. The physical mechanism(s) driving the onset of fueling and ignition in these active galactic nuclei (AGN) are still largely unknown for many galaxies and AGN-selection criteria. Attention has focused on AGN triggering by means of major galaxy mergers gravitationally funneling gas towards the galactic center, with evidence both for and against this scenario. However, several recent studies have found that radio-loud AGN overwhelmingly reside in ongoing or recent major galaxy mergers. In this study, we test the hypothesis that major galaxy mergers are important triggers for radio-loud AGN activity in powerful quasars during cosmic noon (1 < z < 2). To this end, we compare Hubble Space Telescope WFC3/IR observations of the z > 1 3CR radio-loud broad-lined quasars to three matched radio-quiet quasar control samples. We find strong evidence for major-merger activity in nearly all radio-loud AGN, in contrast to the much lower merger fraction in the radio-quiet AGN. These results suggest major galaxy mergers are key ingredients to launching powerful radio jets. Given many of our radio-loud quasars are blue, our results present a possible challenge to the "blow-out" paradigm of galaxy evolution models in which blue quasars are the quiescent end result following a period of red quasar feedback initiated by a galaxy merger. Finally, we find a tight correlation between black hole mass and host galaxy luminosity for these different high-redshift AGN samples inconsistent with those observed for local elliptical galaxies. △ Less

Submitted 1 March, 2024; v1 submitted 19 May, 2023; originally announced May 2023.

Comments: Published by ApJ

Journal ref: ApJ 963 91 (2024)

Autonomous Navigation in Rows of Trees and High Crops with Deep Semantic Segmentation

Authors: Alessandro Navone, Mauro Martini, Andrea Ostuni, Simone Angarano, Marcello Chiaberge

Abstract: Segmentation-based autonomous navigation has recently been proposed as a promising methodology to guide robotic platforms through crop rows without requiring precise GPS localization. However, existing methods are limited to scenarios where the centre of the row can be identified thanks to the sharp distinction between the plants and the sky. However, GPS signal obstruction mainly occurs in the ca… ▽ More Segmentation-based autonomous navigation has recently been proposed as a promising methodology to guide robotic platforms through crop rows without requiring precise GPS localization. However, existing methods are limited to scenarios where the centre of the row can be identified thanks to the sharp distinction between the plants and the sky. However, GPS signal obstruction mainly occurs in the case of tall, dense vegetation, such as high tree rows and orchards. In this work, we extend the segmentation-based robotic guidance to those scenarios where canopies and branches occlude the sky and hinder the usage of GPS and previous methods, increasing the overall robustness and adaptability of the control algorithm. Extensive experimentation on several realistic simulated tree fields and vineyards demonstrates the competitive advantages of the proposed solution. △ Less

Submitted 18 April, 2023; originally announced April 2023.

Domain Generalization for Crop Segmentation with Standardized Ensemble Knowledge Distillation

Authors: Simone Angarano, Mauro Martini, Alessandro Navone, Marcello Chiaberge

Abstract: In recent years, precision agriculture has gradually oriented farming closer to automation processes to support all the activities related to field management. Service robotics plays a predominant role in this evolution by deploying autonomous agents that can navigate fields while performing tasks such as monitoring, spraying, and harvesting without human intervention. To execute these precise act… ▽ More In recent years, precision agriculture has gradually oriented farming closer to automation processes to support all the activities related to field management. Service robotics plays a predominant role in this evolution by deploying autonomous agents that can navigate fields while performing tasks such as monitoring, spraying, and harvesting without human intervention. To execute these precise actions, mobile robots need a real-time perception system that understands their surroundings and identifies their targets in the wild. Existing methods, however, often fall short in generalizing to new crops and environmental conditions. This limit is critical for practical applications where labeled samples are rarely available. In this paper, we investigate the problem of crop segmentation and propose a novel approach to enhance domain generalization using knowledge distillation. In the proposed framework, we transfer knowledge from a standardized ensemble of models individually trained on source domains to a student model that can adapt to unseen realistic scenarios. To support the proposed method, we present a synthetic multi-domain dataset for crop segmentation containing plants of variegate species and covering different terrain styles, weather conditions, and light scenarios for more than 70,000 samples. We demonstrate significant improvements in performance over state-of-the-art methods and superior sim-to-real generalization. Our approach provides a promising solution for domain generalization in crop segmentation and has the potential to enhance a wide variety of agriculture applications. △ Less

Submitted 14 April, 2024; v1 submitted 3 April, 2023; originally announced April 2023.

Online Learning of Wheel Odometry Correction for Mobile Robots with Attention-based Neural Network

Authors: Alessandro Navone, Mauro Martini, Simone Angarano, Marcello Chiaberge

Abstract: Modern robotic platforms need a reliable localization system to operate daily beside humans. Simple pose estimation algorithms based on filtered wheel and inertial odometry often fail in the presence of abrupt kinematic changes and wheel slips. Moreover, despite the recent success of visual odometry, service and assistive robotic tasks often present challenging environmental conditions where visua… ▽ More Modern robotic platforms need a reliable localization system to operate daily beside humans. Simple pose estimation algorithms based on filtered wheel and inertial odometry often fail in the presence of abrupt kinematic changes and wheel slips. Moreover, despite the recent success of visual odometry, service and assistive robotic tasks often present challenging environmental conditions where visual-based solutions fail due to poor lighting or repetitive feature patterns. In this work, we propose an innovative online learning approach for wheel odometry correction, paving the way for a robust multi-source localization system. An efficient attention-based neural network architecture has been studied to combine precise performances with real-time inference. The proposed solution shows remarkable results compared to a standard neural network and filter-based odometry correction algorithms. Nonetheless, the online learning paradigm avoids the time-consuming data collection procedure and can be adopted on a generic robotic platform on-the-fly. △ Less

Submitted 21 March, 2023; originally announced March 2023.

AGN feedback in an infant galaxy cluster: the LOFAR-Chandra view of the giant FRII radio galaxy J103025+052430 at z=1.7

Authors: M. Brienza, R. Gilli, I. Prandoni, Q. D'Amato, K. Rajpurohit, F. Calura, M. Chiaberge, A. Comastri, K. Iwasawa, G. Lanzuisi, E. Liuzzo, S. Marchesi, M. Mignoli, G. Miley, C. Norman, A. Peca, M. Raciti, T. Shimwell, P. Tozzi, C. Vignali, F. Vitello, F. Vito

Abstract: In the nearby universe jets from AGN are observed to have a dramatic impact on their surrounding extragalactic environment. Their effect at the `cosmic noon' (z>1.5), the epoch when star formation and AGN activity peak, is instead much less constrained. Here we present a study of the giant (750 kpc) radio galaxy 103025+052430 located at the centre of a protocluster at redshift z=1.7, with a focus… ▽ More In the nearby universe jets from AGN are observed to have a dramatic impact on their surrounding extragalactic environment. Their effect at the `cosmic noon' (z>1.5), the epoch when star formation and AGN activity peak, is instead much less constrained. Here we present a study of the giant (750 kpc) radio galaxy 103025+052430 located at the centre of a protocluster at redshift z=1.7, with a focus on its interaction with the external medium. We present new LOFAR observations at 144 MHz, which we combine with VLA 1.4 GHz and 0.5-7 keV Chandra archival data. The new map at 144 MHz confirms that the source has a complex morphology, possibly consistent with the `hybrid morphology' classification. The large size of the source gave us the possibility to perform a resolved radio spectral index analysis, a very unique opportunity for a source at such high redshift. This reveals a tentative flattening of the radio spectral index at the edge of the backflow in the Western lobe, which might be indicating plasma compression. The spatial coincidence between this region and the thermal X-ray bubble C suggests a causal connection between the two. Contrary to previous estimates for the bright X-ray component A, we find that inverse Compton scattering between the radio-emitting plasma of the Eastern lobe and the CMB photons can account for a large fraction (~45%-80%) of its total 0.5-7 keV measured flux. Finally, the X-ray bubble C, which is consistent with a thermal origin, is found to be significantly overpressurised with respect to the ambient medium. This suggests that it will tend to expand and release its energy in the surroundings, contributing to the overall intracluster medium heating. Overall, 103025+052430 gives us the chance to investigate the interaction between AGN jets and the surrounding gas in a system that is likely the predecessor of the rich galaxy clusters we all well know at z=0. △ Less

Submitted 24 February, 2023; v1 submitted 23 February, 2023; originally announced February 2023.

Comments: 12 pages, 5 figures, 2 tables, accepted for publication in A&A

PIC4rl-gym: a ROS2 modular framework for Robots Autonomous Navigation with Deep Reinforcement Learning

Authors: Mauro Martini, Andrea Eirale, Simone Cerrato, Marcello Chiaberge

Abstract: Learning agents can optimize standard autonomous navigation improving flexibility, efficiency, and computational cost of the system by adopting a wide variety of approaches. This work introduces the \textit{PIC4rl-gym}, a fundamental modular framework to enhance navigation and learning research by mixing ROS2 and Gazebo, the standard tools of the robotics community, with Deep Reinforcement Learnin… ▽ More Learning agents can optimize standard autonomous navigation improving flexibility, efficiency, and computational cost of the system by adopting a wide variety of approaches. This work introduces the \textit{PIC4rl-gym}, a fundamental modular framework to enhance navigation and learning research by mixing ROS2 and Gazebo, the standard tools of the robotics community, with Deep Reinforcement Learning (DRL). The paper describes the whole structure of the PIC4rl-gym, which fully integrates DRL agent's training and testing in several indoor and outdoor navigation scenarios and tasks. A modular approach is adopted to easily customize the simulation by selecting new platforms, sensors, or models. We demonstrate the potential of our novel gym by benchmarking the resulting policies, trained for different navigation tasks, with a complete set of metrics. △ Less

Submitted 19 November, 2022; originally announced November 2022.

Deep Instance Segmentation and Visual Servoing to Play Jenga with a Cost-Effective Robotic System

Authors: Luca Marchionna, Giulio Pugliese, Mauro Martini, Simone Angarano, Francesco Salvetti, Marcello Chiaberge

Abstract: The game of Jenga represents an inspiring benchmark for developing innovative manipulation solutions for complex tasks. Indeed, it encouraged the study of novel robotics methods to successfully extract blocks from the tower. A Jenga game round undoubtedly embeds many traits of complex industrial or surgical manipulation tasks, requiring a multi-step strategy, the combination of visual and tactile… ▽ More The game of Jenga represents an inspiring benchmark for developing innovative manipulation solutions for complex tasks. Indeed, it encouraged the study of novel robotics methods to successfully extract blocks from the tower. A Jenga game round undoubtedly embeds many traits of complex industrial or surgical manipulation tasks, requiring a multi-step strategy, the combination of visual and tactile data, and the highly precise motion of the robotic arm to perform a single block extraction. In this work, we propose a novel, cost-effective architecture for playing Jenga with e.Do, a 6-DOF anthropomorphic manipulator manufactured by Comau, a standard depth camera, and an inexpensive monodirectional force sensor. Our solution focuses on a visual-based control strategy to accurately align the end-effector with the desired block, enabling block extraction by pushing. To this aim, we train an instance segmentation deep learning model on a synthetic custom dataset to segment each piece of the Jenga tower, allowing visual tracking of the desired block's pose during the motion of the manipulator. We integrate the visual-based strategy with a 1D force sensor to detect whether the block can be safely removed by identifying a force threshold value. Our experimentation shows that our low-cost solution allows e.DO to precisely reach removable blocks and perform up to 14 consecutive extractions in a row. △ Less

Submitted 9 January, 2023; v1 submitted 15 November, 2022; originally announced November 2022.

Journal ref: Sensors. 2023; 23(2):752

RL-DWA Omnidirectional Motion Planning for Person Following in Domestic Assistance and Monitoring

Authors: Andrea Eirale, Mauro Martini, Marcello Chiaberge

Abstract: Robot assistants are emerging as high-tech solutions to support people in everyday life. Following and assisting the user in the domestic environment requires flexible mobility to safely move in cluttered spaces. We introduce a new approach to person following for assistance and monitoring. Our methodology exploits an omnidirectional robotic platform to detach the computation of linear and angular… ▽ More Robot assistants are emerging as high-tech solutions to support people in everyday life. Following and assisting the user in the domestic environment requires flexible mobility to safely move in cluttered spaces. We introduce a new approach to person following for assistance and monitoring. Our methodology exploits an omnidirectional robotic platform to detach the computation of linear and angular velocities and navigate within the domestic environment without losing track of the assisted person. While linear velocities are managed by a conventional Dynamic Window Approach (DWA) local planner, we trained a Deep Reinforcement Learning (DRL) agent to predict optimized angular velocities commands and maintain the orientation of the robot towards the user. We evaluate our navigation system on a real omnidirectional platform in various indoor scenarios, demonstrating the competitive advantage of our solution compared to a standard differential steering following. △ Less

Submitted 13 January, 2023; v1 submitted 9 November, 2022; originally announced November 2022.

A deep 1.4 GHz survey of the J1030 equatorial field: a new window on radio source populations across cosmic time

Authors: Q. D'Amato, I. Prandoni, R. Gilli, C. Vignali, M. Massardi, E. Liuzzo, P. Jagannathan, M. Brienza, R. Paladino, M. Mignoli, S. Marchesi, A. Peca, M. Chiaberge, G. Mazzolari, C. Norman

Abstract: We present deep L-Band observations of the equatorial field centered on the z=6.3 SDSS QSO, reaching a 1 sigma sensitivity of ~2.5 uJy at the center of the field. We extracted a catalog of 1489 radio sources down to a flux density of ~12.5 uJy (5 sigma) over a field of view of ~ 30' diameter. We derived the source counts accounting for catalog reliability and completeness, and compared them with o… ▽ More We present deep L-Band observations of the equatorial field centered on the z=6.3 SDSS QSO, reaching a 1 sigma sensitivity of ~2.5 uJy at the center of the field. We extracted a catalog of 1489 radio sources down to a flux density of ~12.5 uJy (5 sigma) over a field of view of ~ 30' diameter. We derived the source counts accounting for catalog reliability and completeness, and compared them with others available in the literature. Our source counts are among the deepest available so far, and, overall, are consistent with recent counts' determinations and models. We detected for the first time in the radio band the SDSS J1030+0524 QSO (26 +/- 5 uJy). We derived its optical radio loudness R_O = 0.62 +/- 0.12, which makes it the most radio quiet AGN at z >~ 6 discovered so far and detected at radio wavelengths. We unveiled extended diffuse radio emission associated with the lobes of a bright FRII radio galaxy located close to the center of the J1030 field, which is likely to become the future BCG of a protocluster at z=1.7. The lobes' complex morphology, coupled with the presence of X-ray diffuse emission detected around the FRII galaxy lobes, may point toward an interaction between the radio jets and the external medium. We also investigated the relation between radio and X-ray luminosity for a sample of 243 X-ray-selected objects obtained from 500 ks Chandra observations of the same field, and spanning a wide redshift range (0 ~< z ~< 3). Focused on sources with a spectroscopic redshift and classification, we found that sources hosted by ETG and AGN follow Log(L_R)/Log(L_X) linear correlations with slopes of ~0.6 and ~0.8, respectively. This is interpreted as a likely signature of different efficiency in the accretion process. Finally, we found that most of these sources (>~87%) show a radio-to-X-ray radio loudness R_X < -3.5, classifying these objects as radio quiet. △ Less

Submitted 27 October, 2022; originally announced October 2022.

Comments: 18 pages, 15 figures, 5 table. Accepted for publication in A&A

Journal ref: A&A 668, A133 (2022)

The MURALES survey. VII. Optical spectral properties of the nuclei of 3C radio sources at 0.3<z<0.82

Authors: A. Capetti, B. Balmaverde, R. D. Baldi, S. Baum, M. Chiaberge, P. Grandi, A. Marconi, C. O'Dea, G. Venturi

Abstract: This seventh paper of the MUse RAdio Loud Emission lines Snapshot (MURALES) project presents the results of the observations obtained with the VLT/MUSE integral field spectrograph of 3C radio sources and discusses the optical spectral properties of the nuclei of 26 objects with 0.3<z<0.82 (median redshift 0.51). At these redshifts the H$α$ and [NII] emission lines are not covered by optical spectr… ▽ More This seventh paper of the MUse RAdio Loud Emission lines Snapshot (MURALES) project presents the results of the observations obtained with the VLT/MUSE integral field spectrograph of 3C radio sources and discusses the optical spectral properties of the nuclei of 26 objects with 0.3<z<0.82 (median redshift 0.51). At these redshifts the H$α$ and [NII] emission lines are not covered by optical spectra and alternative diagnostic diagrams are needed to separate the different spectroscopic sub-classes. We derive a robust spectroscopic classification into high and low excitation galaxies (HEGs and LEGs) by only using ratios of emission lines in the rest frame UV and blue portion of the spectra. A key result is that FRII/LEGs are found also at the highest level of radio power (up to L$_{178} \sim 2\times 10^{35}$ erg/s/Hz), among the most luminous radio sources in the Universe. Furthermore, their fraction within the FRII RGs population does not strongly depend on radio luminosity. This suggests that the jet properties in powerful FRII radio sources do not depend on the accretion mode and on the structure of the accretion disk as expected if the jet launching process is due to the extraction of the rotational energy of the supermassive black hole. The alternative possibility of recurrent transitions between a LEG and a HEG phase is disfavored based on the variation timescales of the various AGN components. △ Less

Submitted 11 October, 2022; originally announced October 2022.

Comments: Pre-proofs version - Accepted for publication in A&A

Journal ref: A&A 671, A32 (2023)

Generative Adversarial Super-Resolution at the Edge with Knowledge Distillation

Authors: Simone Angarano, Francesco Salvetti, Mauro Martini, Marcello Chiaberge

Abstract: Single-Image Super-Resolution can support robotic tasks in environments where a reliable visual stream is required to monitor the mission, handle teleoperation or study relevant visual details. In this work, we propose an efficient Generative Adversarial Network model for real-time Super-Resolution, called EdgeSRGAN (code available at https://github.com/PIC4SeR/EdgeSRGAN). We adopt a tailored arch… ▽ More Single-Image Super-Resolution can support robotic tasks in environments where a reliable visual stream is required to monitor the mission, handle teleoperation or study relevant visual details. In this work, we propose an efficient Generative Adversarial Network model for real-time Super-Resolution, called EdgeSRGAN (code available at https://github.com/PIC4SeR/EdgeSRGAN). We adopt a tailored architecture of the original SRGAN and model quantization to boost the execution on CPU and Edge TPU devices, achieving up to 200 fps inference. We further optimize our model by distilling its knowledge to a smaller version of the network and obtain remarkable improvements compared to the standard training approach. Our experiments show that our fast and lightweight model preserves considerably satisfying image quality compared to heavier state-of-the-art models. Finally, we conduct experiments on image transmission with bandwidth degradation to highlight the advantages of the proposed system for mobile robotic applications. △ Less

Submitted 9 May, 2023; v1 submitted 7 September, 2022; originally announced September 2022.

Ultra-low-power Range Error Mitigation for Ultra-wideband Precise Localization

Authors: Simone Angarano, Francesco Salvetti, Vittorio Mazzia, Giovanni Fantin, Dario Gandini, Marcello Chiaberge

Abstract: Precise and accurate localization in outdoor and indoor environments is a challenging problem that currently constitutes a significant limitation for several practical applications. Ultra-wideband (UWB) localization technology represents a valuable low-cost solution to the problem. However, non-line-of-sight (NLOS) conditions and complexity of the specific radio environment can easily introduce a… ▽ More Precise and accurate localization in outdoor and indoor environments is a challenging problem that currently constitutes a significant limitation for several practical applications. Ultra-wideband (UWB) localization technology represents a valuable low-cost solution to the problem. However, non-line-of-sight (NLOS) conditions and complexity of the specific radio environment can easily introduce a positive bias in the ranging measurement, resulting in highly inaccurate and unsatisfactory position estimation. In the light of this, we leverage the latest advancement in deep neural network optimization techniques and their implementation on ultra-low-power microcontrollers to introduce an effective range error mitigation solution that provides corrections in either NLOS or LOS conditions with a few mW of power. Our extensive experimentation endorses the advantages and improvements of our low-cost and power-efficient methodology. △ Less

Submitted 7 September, 2022; originally announced September 2022.

Journal ref: SAI: Computing Conference 2022 Proceedings - Springer

NGC 3147: a prototypical low-luminosity active galactic nucleus with double-peaked optical and ultraviolet lines

Authors: Stefano Bianchi, Marco Chiaberge, Ari Laor, Robert Antonucci, Atharva Bagul, Alessandro Capetti

Abstract: A previous narrow-slit ($0.1$ arcsec) Hubble Space Telescope observation unveiled a broad relativistic H$α$ profile in NGC3147, a low-luminosity ($\mathrm{L_{bol}}\sim10^{42}$ erg s$^{-1}$), low-Eddington ratio ($\mathrm{L_{bol}/L_{Edd}}\sim10^{-4}$) active galactic nucleus (AGN), formerly believed to be a candidate true type 2 AGN intrinsically lacking the broad-line region. The new observations… ▽ More A previous narrow-slit ($0.1$ arcsec) Hubble Space Telescope observation unveiled a broad relativistic H$α$ profile in NGC3147, a low-luminosity ($\mathrm{L_{bol}}\sim10^{42}$ erg s$^{-1}$), low-Eddington ratio ($\mathrm{L_{bol}/L_{Edd}}\sim10^{-4}$) active galactic nucleus (AGN), formerly believed to be a candidate true type 2 AGN intrinsically lacking the broad-line region. The new observations presented here confirm the double-peaked profile of the H$α$ line, which further shows variability both in flux and in the inner radius of the emitting disc with respect to the previous epoch. Similar disc line profiles are also found in prominent ultraviolet (UV) lines, in particular Ly$α$ and C IV. The new data also allow us to build a simultaneous subarcsec optical-to-X-ray spectral energy distribution of NGC3147, which is characterized by the absence of a thermal UV bump, and an emission peak in the X-rays. The resulting very flat $α_{ox}=-0.82$ is typical of low-luminosity AGN, and is in good agreement with the extrapolation to low luminosities of the well-known trend with luminosity observed in standard AGN. Indeed, we are possibly observing the accretion disc emission in NGC3147 in the optical, close to the expected peak. On the other hand, the steep -2 UV power law may be Comptonization of that cold disc by a warm corona, what is instead generally observed as a `soft excess' in more luminous AGN. △ Less

Submitted 5 September, 2022; originally announced September 2022.

Comments: Accepted for publication in MNRAS

Back-to-Bones: Rediscovering the Role of Backbones in Domain Generalization

Authors: Simone Angarano, Mauro Martini, Francesco Salvetti, Vittorio Mazzia, Marcello Chiaberge

Abstract: Domain Generalization (DG) studies the capability of a deep learning model to generalize to out-of-training distributions. In the last decade, literature has been massively filled with training methodologies that claim to obtain more abstract and robust data representations to tackle domain shifts. Recent research has provided a reproducible benchmark for DG, pointing out the effectiveness of naiv… ▽ More Domain Generalization (DG) studies the capability of a deep learning model to generalize to out-of-training distributions. In the last decade, literature has been massively filled with training methodologies that claim to obtain more abstract and robust data representations to tackle domain shifts. Recent research has provided a reproducible benchmark for DG, pointing out the effectiveness of naive empirical risk minimization (ERM) over existing algorithms. Nevertheless, researchers persist in using the same outdated feature extractors, and no attention has been given to the effects of different backbones yet. In this paper, we start back to the backbones proposing a comprehensive analysis of their intrinsic generalization capabilities, which so far have been ignored by the research community. We evaluate a wide variety of feature extractors, from standard residual solutions to transformer-based architectures, finding an evident linear correlation between large-scale single-domain classification accuracy and DG capability. Our extensive experimentation shows that by adopting competitive backbones in conjunction with effective data augmentation, plain ERM outperforms recent DG solutions and achieves state-of-the-art accuracy. Moreover, our additional qualitative studies reveal that novel backbones give more similar representations to same-class samples, separating different domains in the feature space. This boost in generalization capabilities leaves marginal room for DG algorithms. It suggests a new paradigm for investigating the problem, placing backbones in the spotlight and encouraging the development of consistent algorithms on top of them. The code is available at https://github.com/PIC4SeR/Back-to-Bones. △ Less

Submitted 9 May, 2023; v1 submitted 2 September, 2022; originally announced September 2022.

Compact dust emission in a gravitationally lensed massive quiescent galaxy at z = 2.15 revealed in ~130 pc-resolution observations by ALMA

Authors: Takahiro Morishita, Abdurro'uf, Hiroyuki Hirashita, Andrew B. Newman, Massimo Stiavelli, Marco Chiaberge

Abstract: We present new observations of MRG-M2129, a quiescent galaxy at z = 2.15 with the Atacama Large Millimeter/submillimeter Array (ALMA). With the combination of the gravitational lensing effect by the foreground cluster and the angular resolution provided by ALMA, our data reveal 1.2 mm continuum emission at $\sim130$ pc angular resolution. Compact dust continuum is detected at 7.9 $σ$ in the target… ▽ More We present new observations of MRG-M2129, a quiescent galaxy at z = 2.15 with the Atacama Large Millimeter/submillimeter Array (ALMA). With the combination of the gravitational lensing effect by the foreground cluster and the angular resolution provided by ALMA, our data reveal 1.2 mm continuum emission at $\sim130$ pc angular resolution. Compact dust continuum is detected at 7.9 $σ$ in the target but displaced from its stellar peak position by $62 \pm 38$ mas, or $\sim169 \pm 105$ pc in the source plane. We find considerably high dust-to-stellar mass ratio, $4 \times 10^{-4}$. From non-detection of the [C i] 3P2 -> 3P1 line, we then derive $3 σ$ upper limits on the molecular gas-to-dust mass ratio $δ_\mathrm{GDR} < 60$ and the molecular gas-to-stellar mass ratio fH2 < 2.3%. The derived $δ_\mathrm{GDR}$ is $>2\times$ smaller than the typical value assumed for quiescent galaxies in the literature. Our study supports that there exists a broad range of $δ_\mathrm{GDR}$ and urges submillimeter follow-up observations of quenching/recently quenched galaxies at similar redshifts. Based on the derived low $δ_\mathrm{GDR}$ and other observed dust properties, we argue that the central black hole is still active and regulates star formation in the system. Our study exhibits a rare case of a gravitationally lensed type 2 QSO harbored by a quiescent galaxy. △ Less

Submitted 7 September, 2022; v1 submitted 22 August, 2022; originally announced August 2022.

Comments: Accepted for publication in ApJ

Position-Agnostic Autonomous Navigation in Vineyards with Deep Reinforcement Learning

Authors: Mauro Martini, Simone Cerrato, Francesco Salvetti, Simone Angarano, Marcello Chiaberge

Abstract: Precision agriculture is rapidly attracting research to efficiently introduce automation and robotics solutions to support agricultural activities. Robotic navigation in vineyards and orchards offers competitive advantages in autonomously monitoring and easily accessing crops for harvesting, spraying and performing time-consuming necessary tasks. Nowadays, autonomous navigation algorithms exploit… ▽ More Precision agriculture is rapidly attracting research to efficiently introduce automation and robotics solutions to support agricultural activities. Robotic navigation in vineyards and orchards offers competitive advantages in autonomously monitoring and easily accessing crops for harvesting, spraying and performing time-consuming necessary tasks. Nowadays, autonomous navigation algorithms exploit expensive sensors which also require heavy computational cost for data processing. Nonetheless, vineyard rows represent a challenging outdoor scenario where GPS and Visual Odometry techniques often struggle to provide reliable positioning information. In this work, we combine Edge AI with Deep Reinforcement Learning to propose a cutting-edge lightweight solution to tackle the problem of autonomous vineyard navigation without exploiting precise localization data and overcoming task-tailored algorithms with a flexible learning-based approach. We train an end-to-end sensorimotor agent which directly maps noisy depth images and position-agnostic robot state information to velocity commands and guides the robot to the end of a row, continuously adjusting its heading for a collision-free central trajectory. Our extensive experimentation in realistic simulated vineyards demonstrates the effectiveness of our solution and the generalization capabilities of our agent. △ Less

Submitted 28 June, 2022; originally announced June 2022.

Journal ref: 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE), 2022, pp. 477-484

Waypoint Generation in Row-based Crops with Deep Learning and Contrastive Clustering

Authors: Francesco Salvetti, Simone Angarano, Mauro Martini, Simone Cerrato, Marcello Chiaberge

Abstract: The development of precision agriculture has gradually introduced automation in the agricultural process to support and rationalize all the activities related to field management. In particular, service robotics plays a predominant role in this evolution by deploying autonomous agents able to navigate in fields while executing different tasks without the need for human intervention, such as monito… ▽ More The development of precision agriculture has gradually introduced automation in the agricultural process to support and rationalize all the activities related to field management. In particular, service robotics plays a predominant role in this evolution by deploying autonomous agents able to navigate in fields while executing different tasks without the need for human intervention, such as monitoring, spraying and harvesting. In this context, global path planning is the first necessary step for every robotic mission and ensures that the navigation is performed efficiently and with complete field coverage. In this paper, we propose a learning-based approach to tackle waypoint generation for planning a navigation path for row-based crops, starting from a top-view map of the region-of-interest. We present a novel methodology for waypoint clustering based on a contrastive loss, able to project the points to a separable latent space. The proposed deep neural network can simultaneously predict the waypoint position and cluster assignment with two specialized heads in a single forward pass. The extensive experimentation on simulated and real-world images demonstrates that the proposed approach effectively solves the waypoint generation problem for both straight and curved row-based crops, overcoming the limitations of previous state-of-the-art methodologies. △ Less

Submitted 23 June, 2022; originally announced June 2022.

Comments: Accepted at ECML PKDD 2022

Journal ref: Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2022. Lecture Notes in Computer Science(), vol 13718, Springer

The Host Galaxy of the Recoiling Black Hole Candidate in 3C 186: An Old Major Merger Remnant at the Center of a z=1 Cluster

Authors: T. Morishita, M. Chiaberge, B. Hilbert, E. Lambrides, L. Blecha, S. Baum, S. Bianchi, A. Capetti, G. Castignani, F. D. Macchetto, G. K. Miley, C. P. O'Dea, C. A. Norman

Abstract: 3C186, a radio-loud quasar at $z=1.0685$, was previously reported to have both velocity and spatial offsets from its host galaxy, and has been considered as a promising candidate for a gravitational wave recoiling black hole triggered by a black hole merger. Another possible scenario is that 3C186 is in an on-going galaxy merger, exhibiting a temporary displacement. In this study, we present analy… ▽ More 3C186, a radio-loud quasar at $z=1.0685$, was previously reported to have both velocity and spatial offsets from its host galaxy, and has been considered as a promising candidate for a gravitational wave recoiling black hole triggered by a black hole merger. Another possible scenario is that 3C186 is in an on-going galaxy merger, exhibiting a temporary displacement. In this study, we present analyses of new deep HST/WFC3-IR and ACS images, aiming to characterize the host galaxy and test this alternative scenario. We carefully measure the light-weighted center of the host and reveal a significant spatial offset from the quasar core ($11.1\pm0.1$kpc). The direction of the confirmed offset aligns almost perpendicularly to the radio jet. We do not find evidence of a recent merger, such as a young starburst in disturbed outskirts, but only marginal light concentration in F160W at $\sim30$kpc. The host consists of matured ($>200$Myr) stellar populations and one compact star-forming region. We compare with hydro-dynamical simulations and find that those observed features are consistently seen in late-stage merger remnants. Taken together, those pieces of evidence indicate that the system is not an on-going/young merger remnant, suggesting that the recoiling black hole scenario is still a plausible explanation for the puzzling nature of 3C186. △ Less

Submitted 26 April, 2022; originally announced April 2022.

Comments: Accepted for publication in ApJ. 14 pages, 8 figures, 2 tables

NOEMA observations support a recoiling black hole in 3C186

Authors: G. Castignani, E. Meyer, M. Chiaberge, F. Combes, T. Morishita, R. Decarli, A. Capetti, M. Dotti, G. R. Tremblay, C. A. Norman

Abstract: 3C186 is a powerful radio loud quasar (QSO) at the center of a cool-core cluster at z=1.06. Previous studies reported evidence for a projected spatial offset of ~1'' between the isophotal center of the galaxy and the point-source QSO as well as a spectral shift of ~2000 km/s between the narrow and broad line region of the system. In this work we report high-resolution molecular gas CO(4-3) observa… ▽ More 3C186 is a powerful radio loud quasar (QSO) at the center of a cool-core cluster at z=1.06. Previous studies reported evidence for a projected spatial offset of ~1'' between the isophotal center of the galaxy and the point-source QSO as well as a spectral shift of ~2000 km/s between the narrow and broad line region of the system. In this work we report high-resolution molecular gas CO(4-3) observations of the system taken with NOEMA interferometer. We clearly detect a large reservoir of molecular gas, $M_{H_2}\sim8\times10^{10}~M_\odot$, that is cospatial with the host galaxy and likely associated with a rotating disk-like structure. We firmly confirm both the spatial offset of the galaxy's gas reservoir with respect to the continuum emission of the QSO and the spectral offset with respect to the redshift of the broad line region. Our morphological and kinematical analysis confirms that the most likely scenario to explain the 3C186 system is that the QSO is a kicked super-massive black hole (SMBH), which we believe may have resulted from a strong gravitational wave recoil as two SMBHs coalesced after the merger of their host galaxies. △ Less

Submitted 12 April, 2022; originally announced April 2022.

Comments: 9 pages, Astronomy & Astrophysics Letters in press

The MURALES survey. VI. Properties and origin of the extended line emission structures in radio galaxies

Authors: Barbara Balmaverde, Alessandro Capetti, R. D. Baldi, S. Baum, M. Chiaberge, R. Gilli, Ana Jimenez-Gallardo, Alessandro Marconi, Francesco Massaro, E. Meyer, C. O'Dea, G. Speranza, E. Torresi, Giacomo Venturi

Abstract: This is the sixth paper presenting the results of the MUse RAdio Loud Emission line Snapshot survey (MURALES). We observed 37 radio sources from the 3C sample with z<0.3 and declination <20 degrees with the MUSE optical integral field spectrograph at the VLT. We here focus on the properties of the extended emission line regions (EELRs) that can be studied with unprecedented detail thanks to the de… ▽ More This is the sixth paper presenting the results of the MUse RAdio Loud Emission line Snapshot survey (MURALES). We observed 37 radio sources from the 3C sample with z<0.3 and declination <20 degrees with the MUSE optical integral field spectrograph at the VLT. We here focus on the properties of the extended emission line regions (EELRs) that can be studied with unprecedented detail thanks to the depth of these observations. Line emission in the 10 FRIs is, in most cases, confined to within <4 kpc, while large-scale (>4 kpc) ionized gas is seen in all but two of the 26 FRIIs. It usually takes the form of elongated or filamentary structures, typically extending between 10 and 30 kpc, but also reaching distances of ~80 kpc. We find that 1) the large-scale ionized gas structures show a tendency to be oriented at large angles from the radio axis, and 2) the gas on a scale of a few kpc from the nucleus often shows ordered rotation with a kinematical axis forming a median angle of 65 degrees with the radio axis. We also discuss the velocity field and ionization properties of the EELRs. The observed emission line structures appear to be associated with gaseous "superdisks" formed after a gas rich merger. The different properties of the EELR can be explained with a combination of the source evolutionary state and the orientation of the "superdisk" with respect to the radio axis. The general alignment between the superdisks and radio axis might be produced by stable and coherent accretion maintained over long timescales. △ Less

Submitted 1 April, 2022; originally announced April 2022.

Comments: Pre-proofs version - Accepted for publication in A&A

Journal ref: A&A 662, A23 (2022)

Marvin: an Innovative Omni-Directional Robotic Assistant for Domestic Environments

Authors: Andrea Eirale, Mauro Martini, Luigi Tagliavini, Dario Gandini, Marcello Chiaberge, Giuseppe Quaglia

Abstract: Population ageing and pandemics recently demonstrate to cause isolation of elderly people in their houses, generating the need for a reliable assistive figure. Robotic assistants are the new frontier of innovation for domestic welfare, and elderly monitoring is one of the services a robot can handle for collective well-being. Despite these emerging needs, in the actual landscape of robotic assista… ▽ More Population ageing and pandemics recently demonstrate to cause isolation of elderly people in their houses, generating the need for a reliable assistive figure. Robotic assistants are the new frontier of innovation for domestic welfare, and elderly monitoring is one of the services a robot can handle for collective well-being. Despite these emerging needs, in the actual landscape of robotic assistants there are no platform which successfully combines a reliable mobility in cluttered domestic spaces, with lightweight and offline Artificial Intelligence (AI) solutions for perception and interaction. In this work, we present Marvin, a novel assistive robotic platform we developed with a modular layer-based architecture, merging a flexible mechanical design with cutting-edge AI for perception and vocal control. We focus the design of Marvin on three target service functions: monitoring of elderly and reduced-mobility subjects, remote presence and connectivity, and night assistance. Compared to previous works, we propose a tiny omnidirectional platform, which enables agile mobility and effective obstacle avoidance. Moreover, we design a controllable positioning device, which easily allows the user to access the interface for connectivity and extends the visual range of the camera sensor. Nonetheless, we delicately consider the privacy issues arising from private data collection on cloud services, a critical aspect of commercial AI-based assistants. To this end, we demonstrate how lightweight deep learning solutions for visual perception and vocal command can be adopted, completely running offline on the embedded hardware of the robot. △ Less

Submitted 14 July, 2022; v1 submitted 10 December, 2021; originally announced December 2021.

Comments: 20 pages, 9 figures, 3 table

Journal ref: Sensors 2022, 22(14), 5261

A Deep Learning Driven Algorithmic Pipeline for Autonomous Navigation in Row-Based Crops

Authors: Simone Cerrato, Vittorio Mazzia, Francesco Salvetti, Mauro Martini, Simone Angarano, Alessandro Navone, Marcello Chiaberge

Abstract: Expensive sensors and inefficient algorithmic pipelines significantly affect the overall cost of autonomous machines. However, affordable robotic solutions are essential to practical usage, and their financial impact constitutes a fundamental requirement to employ service robotics in most fields of application. Among all, researchers in the precision agriculture domain strive to devise robust and… ▽ More Expensive sensors and inefficient algorithmic pipelines significantly affect the overall cost of autonomous machines. However, affordable robotic solutions are essential to practical usage, and their financial impact constitutes a fundamental requirement to employ service robotics in most fields of application. Among all, researchers in the precision agriculture domain strive to devise robust and cost-effective autonomous platforms in order to provide genuinely large-scale competitive solutions. In this article, we present a complete algorithmic pipeline for row-based crops autonomous navigation, specifically designed to cope with low-range sensors and seasonal variations. Firstly, we build on a robust data-driven methodology to generate a viable path for the autonomous machine, covering the full extension of the crop with only the occupancy grid map information of the field. Moreover, our solution leverages on latest advancement of deep learning optimization techniques and synthetic generation of data to provide an affordable solution that efficiently tackles the well-known Global Navigation Satellite System unreliability and degradation due to vegetation growing inside rows. Extensive experimentation and simulations against computer-generated environments and real-world crops demonstrated the robustness and intrinsic generalizability of our methodology that opens the possibility of highly affordable and fully autonomous machines. △ Less

Submitted 14 September, 2023; v1 submitted 7 December, 2021; originally announced December 2021.

Comments: Submitted to Robotics and Autonomous Systems (Elsevier)

The MURALES survey. V. Jet-induced star formation in 3C 277.3 (Coma A)

Authors: A. Capetti, B. Balmaverde, C. Tadhunter, A. Marconi, G. Venturi, M. Chiaberge, R. D. Baldi, S. Baum, R. Gilli, P. Grandi, Eileen T. Meyer, G. Miley, C. O'Dea, W. Sparks, E. Torresi, G. Tremblay

Abstract: We present observations obtained with the VLT/MUSE optical integral field spectrograph of the radio source 3C277.3, located at a redshift of 0.085 and associated with the galaxy Coma A. An emission line region fully enshrouds the double-lobed radio source, which is ~60 kpc x 90 kpc in size. Based on the emission line ratios, we identified five compact knots in which the gas ionization is powered b… ▽ More We present observations obtained with the VLT/MUSE optical integral field spectrograph of the radio source 3C277.3, located at a redshift of 0.085 and associated with the galaxy Coma A. An emission line region fully enshrouds the double-lobed radio source, which is ~60 kpc x 90 kpc in size. Based on the emission line ratios, we identified five compact knots in which the gas ionization is powered by young stars located as far as ~60 kpc from the host. The emission line filaments surrounding the radio emission are compatible with ionization from fast shocks (with a velocity of 350-500 km/s), but a contribution from star formation occurring at the edges of the radio source is likely. Coma A might be a unique example in the local Universe in which the expanding outflow triggers star formation throughout the whole radio source. △ Less

Submitted 2 November, 2021; originally announced November 2021.

Comments: Pre-proofs version - Accepted for publication in A&A

Redshift identification of X-ray selected active galactic nuclei in the J1030 field: searching for large-scale structures and high-redshift sources

Authors: Stefano Marchesi, Marco Mignoli, Roberto Gilli, Alessandro Peca, Micol Bolzonella, Riccardo Nanni, Marianna Annunziatella, Barbara Balmaverde, Marcella Brusa, Francesco Calura, Letizia P. Cassarà, Marco Chiaberge, Andrea Comastri, Felice Cusano, Quirino D'Amato, Kazushi Iwasawa, Giorgio Lanzuisi, Danilo Marchesini, Takahiro Morishita, Isabella Prandoni, Andrea Rossi, Paolo Tozzi, Cristian Vignali, Fabio Vito, Giovanni Zamorani , et al. (1 additional authors not shown)

Abstract: We publicly release the spectroscopic and photometric redshift catalog of the sources detected with Chandra in the field of the $z$=6.3 quasar SDSS J1030+0525. This is currently the fifth deepest X-ray field, and reaches a 0.5-2 keV flux limit $f_{\rm 0.5-2}$=6$\times$10$^{-17}$ erg s$^{-1}$ cm$^{-2}$. By using two independent methods, we measure a photometric redshift for 243 objects, while 123 (… ▽ More We publicly release the spectroscopic and photometric redshift catalog of the sources detected with Chandra in the field of the $z$=6.3 quasar SDSS J1030+0525. This is currently the fifth deepest X-ray field, and reaches a 0.5-2 keV flux limit $f_{\rm 0.5-2}$=6$\times$10$^{-17}$ erg s$^{-1}$ cm$^{-2}$. By using two independent methods, we measure a photometric redshift for 243 objects, while 123 (51%) sources also have a spectroscopic redshift, 110 of which coming from an INAF-Large Binocular Telescope (LBT) Strategic Program. We use the spectroscopic redshifts to determine the quality of the photometric ones, and find it in agreement with that of other X-ray surveys which used a similar number of photometric data-points. In particular, we measure a sample normalized median absolute deviation $σ_{NMAD}$=1.48||$z_{phot}$-$z_{spec}$||/(1+$z_{spec}$)=0.065. We use these new spectroscopic and photometric redshifts to study the properties of the Chandra J1030 field. We observe several peaks in our spectroscopic redshift distribution between $z$=0.15 and $z$=1.5, and find that the sources in each peak are often distributed across the whole Chandra field of view. This evidence confirms that X-ray selected AGN can efficiently track large-scale structures over physical scales of several Mpc. Finally, we computed the Chandra J1030 $z>$3 number counts: while the spectroscopic completeness at high-redshift of our sample is limited, our results point towards a potential source excess at $z\geq$4, which we plan to either confirm or reject in the near future with dedicated spectroscopic campaigns. △ Less

Submitted 16 September, 2021; originally announced September 2021.

Comments: 21 pages, 15 figures. Accepted for publication in Astronomy and Astrophysics

Journal ref: A&A 656, A117 (2021)

Lower-Luminosity Obscured AGN Host Galaxies are Not Predominantly in Major-Merging Systems at Cosmic Noon

Authors: Erini Lambrides, Marco Chiaberge, Timothy Heckman, Allison Kirkpatrick, Eileen T. Meyer, Andreea Petric, Kirsten Hall, Arianna Long, Duncan J. Watts, Roberto Gilli, Raymond Simons, Kirill Tchernyshyov, Vicente Rodriguez-Gomez, Fabio Vito, Alexander De La Vega, Jeffrey R. Davis, Dale D Kocevski, Colin Norman

Abstract: For over 60 years, the scientific community has studied actively growing central super-massive black holes (active galactic nuclei -- AGN) but fundamental questions on their genesis remain unanswered. Numerical simulations and theoretical arguments show that black hole growth occurs during short-lived periods ($\sim$ 10$^{7}$ -10$^{8}$ yr) of powerful accretion. Major mergers are commonly invoked… ▽ More For over 60 years, the scientific community has studied actively growing central super-massive black holes (active galactic nuclei -- AGN) but fundamental questions on their genesis remain unanswered. Numerical simulations and theoretical arguments show that black hole growth occurs during short-lived periods ($\sim$ 10$^{7}$ -10$^{8}$ yr) of powerful accretion. Major mergers are commonly invoked as the most likely dissipative process to trigger the rapid fueling of AGN. If the AGN-merger paradigm is true, we expect galaxy mergers to coincide with black hole accretion during a heavily obscured AGN phase (N$_H$ $ > 10^{23}$ cm$^{-2}$). Starting from one of the largest samples of obscured AGN at 0.5 $<$ $z$ $<$ 3.1, we select 40 non-starbursting lower-luminosity obscured AGN. We then construct a one-to-one matched redshift- and near-IR magnitude-matched non-starbursting inactive galaxy control sample. Combining deep color \textit{Hubble Space Telescope} imaging and a novel method of human classification, we test the merger-AGN paradigm prediction that heavily obscured AGN are strongly associated with galaxies undergoing a major merger. On the total sample of 80 galaxies, we estimate each individual classifier's accuracy at identifying merging galaxies/post-merging systems and isolated galaxies. We calculate the probability of each galaxy being in either a major merger or isolated system, given the accuracy of the human classifiers and the individual classifications of each galaxy. We do not find statistically significant evidence that obscured AGN at cosmic noon are predominately found in systems with evidence of significant merging/post-merging features. △ Less

Submitted 15 July, 2021; originally announced July 2021.

Comments: 19 pages, 11 figures, accepted in ApJ

Deep Semantic Segmentation at the Edge for Autonomous Navigation in Vineyard Rows

Authors: Diego Aghi, Simone Cerrato, Vittorio Mazzia, Marcello Chiaberge

Abstract: Precision agriculture is a fast-growing field that aims at introducing affordable and effective automation into agricultural processes. Nowadays, algorithmic solutions for navigation in vineyards require expensive sensors and high computational workloads that preclude large-scale applicability of autonomous robotic platforms in real business case scenarios. From this perspective, our novel propose… ▽ More Precision agriculture is a fast-growing field that aims at introducing affordable and effective automation into agricultural processes. Nowadays, algorithmic solutions for navigation in vineyards require expensive sensors and high computational workloads that preclude large-scale applicability of autonomous robotic platforms in real business case scenarios. From this perspective, our novel proposed control leverages the latest advancement in machine perception and edge AI techniques to achieve highly affordable and reliable navigation inside vineyard rows with low computational and power consumption. Indeed, using a custom-trained segmentation network and a low-range RGB-D camera, we are able to take advantage of the semantic information of the environment to produce smooth trajectories and stable control in different vineyards scenarios. Moreover, the segmentation maps generated by the control algorithm itself could be directly exploited as filters for a vegetative assessment of the crop status. Extensive experimentations and evaluations against real-world data and simulated environments demonstrated the effectiveness and intrinsic robustness of our methodology. △ Less

Submitted 1 July, 2021; originally announced July 2021.

Comments: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021)

Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

Authors: Vittorio Mazzia, Simone Angarano, Francesco Salvetti, Federico Angelini, Marcello Chiaberge

Abstract: Deep neural networks based purely on attention have been successful across several domains, relying on minimal architectural priors from the designer. In Human Action Recognition (HAR), attention mechanisms have been primarily adopted on top of standard convolutional or recurrent layers, improving the overall generalization capability. In this work, we introduce Action Transformer (AcT), a simple,… ▽ More Deep neural networks based purely on attention have been successful across several domains, relying on minimal architectural priors from the designer. In Human Action Recognition (HAR), attention mechanisms have been primarily adopted on top of standard convolutional or recurrent layers, improving the overall generalization capability. In this work, we introduce Action Transformer (AcT), a simple, fully self-attentional architecture that consistently outperforms more elaborated networks that mix convolutional, recurrent and attentive layers. In order to limit computational and energy requests, building on previous human action recognition research, the proposed approach exploits 2D pose representations over small temporal windows, providing a low latency solution for accurate and effective real-time performance. Moreover, we open-source MPOSE2021, a new large-scale dataset, as an attempt to build a formal training and evaluation benchmark for real-time, short-time HAR. The proposed methodology was extensively tested on MPOSE2021 and compared to several state-of-the-art architectures, proving the effectiveness of the AcT model and laying the foundations for future work on HAR. △ Less

Submitted 10 January, 2022; v1 submitted 1 July, 2021; originally announced July 2021.

Comments: Published by Pattern Recognition, Elsevier

Journal ref: Pattern Recognition, Volume 124, April 2022, 108487

Merger or Not: Accounting for Human Biases in Identifying Galactic Merger Signatures

Authors: Erini Lambrides, Duncan J. Watts, Marco Chiaberge, Kirill Tchernyshyov, Allison Kirkpatrick, Eileen T. Meyer, Timothy Heckman, Raymond Simons, Oz Amram, Kirsten R. Hall, Arianna Long, Colin Norman

Abstract: Significant galaxy mergers throughout cosmic time play a fundamental role in theories of galaxy evolution. The widespread usage of human classifiers to visually assess whether galaxies are in merging systems remains a fundamental component of many morphology studies. Studies that employ human classifiers usually construct a control sample, and rely on the assumption that the bias introduced by usi… ▽ More Significant galaxy mergers throughout cosmic time play a fundamental role in theories of galaxy evolution. The widespread usage of human classifiers to visually assess whether galaxies are in merging systems remains a fundamental component of many morphology studies. Studies that employ human classifiers usually construct a control sample, and rely on the assumption that the bias introduced by using humans will be evenly applied to all samples. In this work, we test this assumption and develop methods to correct for it. Using the standard binomial statistical methods employed in many morphology studies, we find that the merger fraction, error, and the significance of the difference between two samples are dependent on the intrinsic merger fraction of any given sample. We propose a method of quantifying merger biases of individual human classifiers and incorporate these biases into a full probabilistic model to determine the merger fraction and the probability of an individual galaxy being in a merger. Using 14 simulated human responses and accuracies, we are able to correctly label a galaxy as ''merger'' or ''isolated'' to within 1\% of the truth. Using 14 real human responses on a set of realistic mock galaxy simulation snapshots our model is able to recover the pre-coalesced merger fraction to within 10\%. Our method can not only increase the accuracy of studies probing the merger state of galaxies at cosmic noon, but also can be used to construct more accurate training sets in machine learning studies that use human classified data-sets. △ Less

Submitted 29 June, 2021; originally announced June 2021.

Comments: 23 pages, 16 figures, accepted for publication in ApJ, https://github.com/elambrid/merger_or_not

The MURALES survey. IV. Searching for nuclear outflows in 3C radio galaxies at z < 0.3 with MUSE observations

Authors: Giovanna Speranza, Barbara Balmaverde, Alessandro Capetti, Francesco Massaro, G. Tremblay, Alessandro Marconi, Giacomo Venturi, M. Chiaberge, R. D. Baldi, S. Baum, P. Grandi, Eileen T. Meyer, C. 0'Dea, W. Sparks, B. A. Terrazas, E. Torresi

Abstract: We analyze VLT/MUSE observations of 37 radio galaxies from the Third Cambridge catalogue (3C) with redshift $<$0.3 searching for nuclear outflows of ionized gas. These observations are part of the MURALES project (a MUse RAdio Loud Emission line Snapshot survey), whose main goal is to explore the feedback process in the most powerful radio-loud AGN. We applied a nonparametric analysis to the [O~II… ▽ More We analyze VLT/MUSE observations of 37 radio galaxies from the Third Cambridge catalogue (3C) with redshift $<$0.3 searching for nuclear outflows of ionized gas. These observations are part of the MURALES project (a MUse RAdio Loud Emission line Snapshot survey), whose main goal is to explore the feedback process in the most powerful radio-loud AGN. We applied a nonparametric analysis to the [O~III] $λ$5007 emission line, whose asymmetries and high-velocity wings reveal signatures of outflows. We find evidence of nuclear outflows in 21 sources, with velocities between $\sim$400 - 1000 km s$^{-1}$, outflowing masses of $\sim 10^5-10^7$ M$_\odot$, and a kinetic energy in the range $\sim 10^{53} - 10^{56}$ erg. In addition, evidence for extended outflows is found in the 2D gas velocity maps of 13 sources of the subclasses of high-excitation (HEG) and broad-line (BLO) radio galaxies, with sizes between 0.4 and 20 kpc. We estimate a mass outflow rate in the range 0.4 - 30 M$_\odot$ yr$^{-1}$ and an energy deposition rate of ${\dot E}_{kin} \sim 10^{42}-10^{45} $ erg s$^{-1}$. Comparing the jet power, the nuclear luminosity of the active galactic nucleus, and the outflow kinetic energy rate, we find that outflows of HEGs and BLOs are likely radiatively powered, while jets likely only play a dominant role in galaxies with low excitation. The low loading factors we measured suggest that these outflows are driven by momentum and not by energy. Based on the gas masses, velocities, and energetics involved, we conclude that the observed ionized outflows have a limited effect on the gas content or the star formation in the host. In order to obtain a complete view of the feedback process, observations exploring the complex multiphase structure of outflows are required. △ Less

Submitted 17 June, 2021; originally announced June 2021.

Comments: 40 pages; accepted for publication on A&A;

Report number: AA/2021/40686

Journal ref: A&A 653, A150 (2021)

Domain-Adversarial Training of Self-Attention Based Networks for Land Cover Classification using Multi-temporal Sentinel-2 Satellite Imagery

Authors: Mauro Martini, Vittorio Mazzia, Aleem Khaliq, Marcello Chiaberge

Abstract: The increasing availability of large-scale remote sensing labeled data has prompted researchers to develop increasingly precise and accurate data-driven models for land cover and crop classification (LC&CC). Moreover, with the introduction of self-attention and introspection mechanisms, deep learning approaches have shown promising results in processing long temporal sequences in the multi-spectra… ▽ More The increasing availability of large-scale remote sensing labeled data has prompted researchers to develop increasingly precise and accurate data-driven models for land cover and crop classification (LC&CC). Moreover, with the introduction of self-attention and introspection mechanisms, deep learning approaches have shown promising results in processing long temporal sequences in the multi-spectral domain with a contained computational request. Nevertheless, most practical applications cannot rely on labeled data, and in the field, surveys are a time consuming solution that poses strict limitations to the number of collected samples. Moreover, atmospheric conditions and specific geographical region characteristics constitute a relevant domain gap that does not allow direct applicability of a trained model on the available dataset to the area of interest. In this paper, we investigate adversarial training of deep neural networks to bridge the domain discrepancy between distinct geographical zones. In particular, we perform a thorough analysis of domain adaptation applied to challenging multi-spectral, multi-temporal data, accurately highlighting the advantages of adapting state-of-the-art self-attention based models for LC&CC to different target zones where labeled data are not available. Extensive experimentation demonstrated significant performance and generalization gain in applying domain-adversarial training to source and target regions with marked dissimilarities between the distribution of extracted features. △ Less

Submitted 30 June, 2021; v1 submitted 1 April, 2021; originally announced April 2021.

Journal ref: Remote Sensing 13.13 (2021): 2564

Efficient-CapsNet: Capsule Network with Self-Attention Routing

Authors: Vittorio Mazzia, Francesco Salvetti, Marcello Chiaberge

Abstract: Deep convolutional neural networks, assisted by architectural design strategies, make extensive use of data augmentation techniques and layers with a high number of feature maps to embed object transformations. That is highly inefficient and for large datasets implies a massive redundancy of features detectors. Even though capsules networks are still in their infancy, they constitute a promising s… ▽ More Deep convolutional neural networks, assisted by architectural design strategies, make extensive use of data augmentation techniques and layers with a high number of feature maps to embed object transformations. That is highly inefficient and for large datasets implies a massive redundancy of features detectors. Even though capsules networks are still in their infancy, they constitute a promising solution to extend current convolutional networks and endow artificial visual perception with a process to encode more efficiently all feature affine transformations. Indeed, a properly working capsule network should theoretically achieve higher results with a considerably lower number of parameters count due to intrinsic capability to generalize to novel viewpoints. Nevertheless, little attention has been given to this relevant aspect. In this paper, we investigate the efficiency of capsule networks and, pushing their capacity to the limits with an extreme architecture with barely 160K parameters, we prove that the proposed architecture is still able to achieve state-of-the-art results on three different datasets with only 2% of the original CapsNet parameters. Moreover, we replace dynamic routing with a novel non-iterative, highly parallelizable routing algorithm that can easily cope with a reduced number of capsules. Extensive experimentation with other capsule implementations has proved the effectiveness of our methodology and the capability of capsule networks to efficiently embed visual representations more prone to generalization. △ Less

Submitted 6 July, 2021; v1 submitted 29 January, 2021; originally announced January 2021.

Comments: Accepted by Scientific Reports

Robust Ultra-wideband Range Error Mitigation with Deep Learning at the Edge

Authors: Simone Angarano, Vittorio Mazzia, Francesco Salvetti, Giovanni Fantin, Marcello Chiaberge

Abstract: Ultra-wideband (UWB) is the state-of-the-art and most popular technology for wireless localization. Nevertheless, precise ranging and localization in non-line-of-sight (NLoS) conditions is still an open research topic. Indeed, multipath effects, reflections, refractions, and complexity of the indoor radio environment can easily introduce a positive bias in the ranging measurement, resulting in hig… ▽ More Ultra-wideband (UWB) is the state-of-the-art and most popular technology for wireless localization. Nevertheless, precise ranging and localization in non-line-of-sight (NLoS) conditions is still an open research topic. Indeed, multipath effects, reflections, refractions, and complexity of the indoor radio environment can easily introduce a positive bias in the ranging measurement, resulting in highly inaccurate and unsatisfactory position estimation. This article proposes an efficient representation learning methodology that exploits the latest advancement in deep learning and graph optimization techniques to achieve effective ranging error mitigation at the edge. Channel Impulse Response (CIR) signals are directly exploited to extract high semantic features to estimate corrections in either NLoS or LoS conditions. Extensive experimentation with different settings and configurations has proved the effectiveness of our methodology and demonstrated the feasibility of a robust and low computational power UWB range error mitigation. △ Less

Submitted 28 April, 2021; v1 submitted 30 November, 2020; originally announced November 2020.

Comments: Submitted to Engineering Applications of Artificial Intelligence

Journal ref: Engineering Applications of Artificial Intelligence, Volume 102, June 2021, 104278

Indoor Point-to-Point Navigation with Deep Reinforcement Learning and Ultra-wideband

Authors: Enrico Sutera, Vittorio Mazzia, Francesco Salvetti, Giovanni Fantin, Marcello Chiaberge

Abstract: Indoor autonomous navigation requires a precise and accurate localization system able to guide robots through cluttered, unstructured and dynamic environments. Ultra-wideband (UWB) technology, as an indoor positioning system, offers precise localization and tracking, but moving obstacles and non-line-of-sight occurrences can generate noisy and unreliable signals. That, combined with sensors noise,… ▽ More Indoor autonomous navigation requires a precise and accurate localization system able to guide robots through cluttered, unstructured and dynamic environments. Ultra-wideband (UWB) technology, as an indoor positioning system, offers precise localization and tracking, but moving obstacles and non-line-of-sight occurrences can generate noisy and unreliable signals. That, combined with sensors noise, unmodeled dynamics and environment changes can result in a failure of the guidance algorithm of the robot. We demonstrate how a power-efficient and low computational cost point-to-point local planner, learnt with deep reinforcement learning (RL), combined with UWB localization technology can constitute a robust and resilient to noise short-range guidance system complete solution. We trained the RL agent on a simulated environment that encapsulates the robot dynamics and task constraints and then, we tested the learnt point-to-point navigation policies in a real setting with more than two-hundred experimental evaluations using UWB localization. Our results show that the computational efficient end-to-end policy learnt in plain simulation, that directly maps low-range sensors signals to robot controls, deployed in combination with ultra-wideband noisy localization in a real environment, can provide a robust, scalable and at-the-edge low-cost navigation system solution. △ Less

Submitted 18 November, 2020; originally announced November 2020.

Comments: Accepted by ICAART 2021 - http://www.icaart.org/

DeepWay: a Deep Learning Waypoint Estimator for Global Path Generation

Authors: Vittorio Mazzia, Francesco Salvetti, Diego Aghi, Marcello Chiaberge

Abstract: Agriculture 3.0 and 4.0 have gradually introduced service robotics and automation into several agricultural processes, mostly improving crops quality and seasonal yield. Row-based crops are the perfect settings to test and deploy smart machines capable of monitoring and manage the harvest. In this context, global path generation is essential either for ground or aerial vehicles, and it is the star… ▽ More Agriculture 3.0 and 4.0 have gradually introduced service robotics and automation into several agricultural processes, mostly improving crops quality and seasonal yield. Row-based crops are the perfect settings to test and deploy smart machines capable of monitoring and manage the harvest. In this context, global path generation is essential either for ground or aerial vehicles, and it is the starting point for every type of mission plan. Nevertheless, little attention has been currently given to this problem by the research community and global path generation automation is still far to be solved. In order to generate a viable path for an autonomous machine, the presented research proposes a feature learning fully convolutional model capable of estimating waypoints given an occupancy grid map. In particular, we apply the proposed data-driven methodology to the specific case of row-based crops with the general objective to generate a global path able to cover the extension of the crop completely. Extensive experimentation with a custom made synthetic dataset and real satellite-derived images of different scenarios have proved the effectiveness of our methodology and demonstrated the feasibility of an end-to-end and completely autonomous global path planner. △ Less

Submitted 21 January, 2021; v1 submitted 30 October, 2020; originally announced October 2020.

Comments: Submitted to Computers and Electronics in Agriculture

Journal ref: Volume 184, May 2021, 106091