Paolo Tripicchio
Scuola Superiore Sant'Anna, PERCRO, Department Member
Research Interests:
Inventory is one of the most problematic tasks in warehouses and retail areas. It is a challenging activity which still suffers from item losses, by producing waste in terms of time, materials, and profits. To solve this big problem, the... more
Inventory is one of the most problematic tasks in warehouses and retail areas. It is a challenging activity which still suffers from item losses, by producing waste in terms of time, materials, and profits. To solve this big problem, the Radio Frequency Identification (RFID) technology can be profitably employed, with great advantage if combined with robots by allowing the implementation of an automatic inventory system, without the need of operator actions. Besides, a localization system can be designed by exploiting a synthetic-array approach, with no additional cost, as proposed within the MONITOR Project here presented. In particular, this paper describes the main design criteria to implement an autonomous inventory and localization system through a UHF-RFID Robot.
Research Interests:
The use of radio frequency identification (RFID) technology for the traceability of products throughout the production chain, warehouse management, and the retail network is spreading in the last years, especially in those industries in... more
The use of radio frequency identification (RFID) technology for the traceability of products throughout the production chain, warehouse management, and the retail network is spreading in the last years, especially in those industries in line with the concept of Industry 4.0. The last decade has seen the development of increasingly precise and high-performance methods for the localization of goods. This work proposes a reliable 2-D localization methodology that is faster and provides a competitive accuracy, concerning the state-of-the-art techniques. The proposed method leverages a phase-distance model and exploits the synthetic aperture approach and unwrapping techniques for facing phase ambiguity and multipath phenomena. Trilateration applied on consecutive phase readings allows finding hyperbolae as the localization solution space. Analytic calculus is used to compute intersections among the conics that estimate the tag position. An algorithm computes intersections quality to select the best estimation. Experimental tests are conducted to assess the quality of the proposed strategy. A mobile robot equipped with a reader antenna localizes in 2-D the tags placed in an indoor scenario and reconstructs the map of the environment through a simultaneous localization and mapping (SLAM) algorithm. Note to Practitioners—A localization technology based on passive ultrahigh-frequency (UHF) radio frequency identification (RFID) is an enabling technology for intelligent warehouses, logistics, and retails. For this reason, this work presents a novel method to estimate the tag location with high accuracy. A reader antenna is mounted on an autonomous mobile robot that can move in an indoor or outdoor environment due to a simultaneous localization and mapping (SLAM) algorithm. The motion of the antenna generates a synthetic aperture. The system receives the phase measurements from the RFID tags and generates a distance model through phase unwrapping. In such a way, the possible locations of the tags in the environment are generated, creating conics. The trilateration step is performed analytically, intersecting the obtained conics. The resulting estimations are very accurate and not computation expensive. Therefore, the proposed approach can be employed in any application where localizing objects is fundamental even when reduced computational power is available, e.g., in warehouses where the products are at known heights, or where the items are placed on a fixed infrastructure, such as high-shelves logistics, to produce the inventory of the tagged objects within each shelf.
Research Interests:
Research Interests:
Abstract Robotics solutions in the industrial sector are moving towards higher degrees of flexibility. In such a context, the latest research is focused on grasping technologies for material handling. The aim of this work is to study a... more
Abstract Robotics solutions in the industrial sector are moving towards higher degrees of flexibility. In such a context, the latest research is focused on grasping technologies for material handling. The aim of this work is to study a possible way to overcome current limitations on the existing robotic solutions for picking objects in cluttered environments. Pick and place task is of fundamental relevance in automated manufacturing companies. Many industrial processes, have bin-picking as the preliminary phase task. The challenge is to handle mixed bins that contain multiple different types of parts with complex configuration to grasp in a more flexible way. Using intelligent autonomous robots for picking different kinds of objects allows industries to handle a wide assortment of parts without the need for any change in the hardware structure and design. In this manuscript, a collaborative robotics system is presented. It integrates a classical two-finger gripper and a custom designed soft-robotics end-effector in a commercial robot. In this work, 3D perception technologies are employed in order to select suitable geometries that exploit at best the advantages of the universal jamming gripper (UJG) in picking objects of different shapes. A custom algorithm that selects a suitable picking point, solving the perceptual issues posed by cluttered environments, is introduced. Grasping tests to assess the capabilities of the custom gripper have been conducted. The objects selected to test the performance of the proposed system do not consider just a specific manufacturing scenario but are meant to be more general, not limiting the analysis to a single case. Furthermore, a comparative study of the algorithmic performance of the proposed method with respect to state-of-the-art techniques is presented. The proposed system is shown to be able to grasp novel objects in a cluttered environment with a competitive success rate, determining a suitable grasping target in short time.
Research Interests:
This paper presents a multi-antenna approach of the phase-based SARFID method to locate static tags by employing two UHF-RFID reader antennas installed on an Unmanned Grounded Vehicle (UGV). The UGV is remotecontrolled and equipped with... more
This paper presents a multi-antenna approach of the phase-based SARFID method to locate static tags by employing two UHF-RFID reader antennas installed on an Unmanned Grounded Vehicle (UGV). The UGV is remotecontrolled and equipped with Laser Range Finder sensors to move inside an indoor environment, and the knowledge of its trajectory is achieved through a Simultaneous Localization And Mapping procedure. By processing the phase data collected from each reader antenna, different matching functions can be obtained and combined to improve the estimation of the bidimensional tag position. Differently from other localization techniques, neither reference tags (anchor tags), nor large phased array antennas are required.
Research Interests:
Research Interests:
Research Interests:
This paper presents the application of the phase-based SARFID technique to locate static tags through an Unmanned Grounded Vehicle (UGV) equipped with a UHF-RFID reader. The UGV is remote-controlled to move inside an indoor environment... more
This paper presents the application of the phase-based SARFID technique to locate static tags through an Unmanned Grounded Vehicle (UGV) equipped with a UHF-RFID reader. The UGV is remote-controlled to move inside an indoor environment and the knowledge of its trajectory is achieved through a Simultaneous Localization And Mapping (SLAM) procedure. The bi-dimensional tag position can be estimated with a location error in the order of few centimetres if the phase samples are collected in a proper spatial interval. Differently from other localization techniques, neither reference tags (anchor tags), nor large phased array antennas are required.
Research Interests:
Research Interests:
This paper proposes a multi-antenna approach of a synthetic aperture radar method for 3D UHF-RFID tag localization by exploiting an Unmanned Grounded Vehicle (UGV). The UGV is remote-controlled to move inside a complex indoor environment,... more
This paper proposes a multi-antenna approach of a synthetic aperture radar method for 3D UHF-RFID tag localization by exploiting an Unmanned Grounded Vehicle (UGV). The UGV is remote-controlled to move inside a complex indoor environment, and the knowledge of the reader antenna trajectories is achieved with millimeter accuracy through a commercial motion tracking system. By processing the tag backscattered signal phase data collected from two RFID reader antennas, coherent and non-coherent combining processing are performed to improve the estimation of the 3D tag position with respect to the case of a single antenna. Differently from other localization techniques, neither reference tags (anchor tags), nor large phased array antennas are required.
Research Interests:
Electroadhesion is a suitable technology for developing grippers for applications where fragile, compliant or variable shape objects need to be grabbed and where a retention action is typically preferred to a compression force. This... more
Electroadhesion is a suitable technology for developing grippers for applications where fragile, compliant or variable shape objects need to be grabbed and where a retention action is typically preferred to a compression force. This article presents a self-sensing technique for electroadhesive devices (EAD) based on the capacitance measure. Specifically, we demonstrate that measuring the variation of the capacitance between electrodes of an EAD during the adhesion can provide useful information to automatically detect the successful grip of an object and the possible loss of adhesion during manipulation. To this aim, a dedicated electronic circuit is developed that is able to measure capacitance variations while the high voltage required for the adhesion is activated. A test bench characterization is presented to evaluate the self-sensing of capacitance during different states: (1) the EAD is far away from the object to be grasped; (2) the EAD is in contact with the object, but the ...
Research Interests:
Recent advancements in human motion behaviors based on camera images made human motion tracking much more robust than in the past. However, they are still computationally expensive and they do not allow for online reconstruction of the... more
Recent advancements in human motion behaviors based on camera images made human motion tracking much more robust than in the past. However, they are still computationally expensive and they do not allow for online reconstruction of the human pose. In this context, this work presents a framework for tracking 2D human pose at high frequency while keeping the robustness of state-of-the-art methods. We achieved this result by combining, by means of Kalman filtering, the recent success of OpenPose, a robust deep learning-based 2D human pose estimation technique, with the fast Lucas-Kanade features matching method. This allows for processing images at different framerates, thus having multirate measurement updates. The frequency of the estimation is kept high by setting the Kalman filter time step. The method was tested on videos of several activities which include both fast and slow motion. The results show an improvement on the reconstruction error, an increased speed of reconstruction, better tracking during fast motion, and the capability to cover loss of tracking from the two measurements. The achieved intraframe (between two available measurements) trajectory estimation frequency was as high as 1 kHz.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Within the Industry 4.0 ecosystem, Inspection Robotics is one fundamental technology to speed up monitoring processes and obtain good accuracy and performance of the inspections while avoiding possible safety issues for human personnel.... more
Within the Industry 4.0 ecosystem, Inspection Robotics is one fundamental technology to speed up monitoring processes and obtain good accuracy and performance of the inspections while avoiding possible safety issues for human personnel. This manuscript investigates the robotics inspection of areas and surfaces employing Unmanned Aerial Vehicles (UAVs). The contribution starts by addressing the problem of coverage path planning and proposes a smoothing approach intended to reduce both flight time and memory consumption to store the target navigation path. Evaluation tests are conducted on a quadrotor equipped with a Model Predictive Control (MPC) policy and a Simultaneous Localization and Mapping (SLAM) algorithm to localize the UAV in the environment.
Research Interests:
Research Interests:
Research Interests:
Calibrating intrinsic and extrinsic camera parameters is a fundamental problem that is a preliminary task for a wide variety of applications, from robotics to computer vision to surveillance and industrial tasks. With the advent of... more
Calibrating intrinsic and extrinsic camera parameters is a fundamental problem that is a preliminary task for a wide variety of applications, from robotics to computer vision to surveillance and industrial tasks. With the advent of Internet of Things (IoT) technology and edge computing capabilities, the ability to track motion activities in large outdoor areas has become feasible. The proposed work presents a network of IoT camera nodes and a dissertation on two possible approaches for automatically estimating their poses. One approach follows the Structure from Motion (SfM) pipeline, while the other is marker-based. Both methods exploit the correspondence of features detected by cameras on synchronized frames. A preliminary indoor experiment was conducted to assess the performance of the two methods compared to ground truth measurements, employing a commercial tracking system of millimetric precision. Outdoor experiments directly compared the two approaches on a larger setup. The r...
Research Interests:
A novel methodology for Enactive Skills Transmission of Tai Chi movements through sound feed-back stimuli is presented in this paper. We adopted the paradigm of spatial sound and frequency distortion as an accurate methodology for the... more
A novel methodology for Enactive Skills Transmission of Tai Chi movements through sound feed-back stimuli is presented in this paper. We adopted the paradigm of spatial sound and frequency distortion as an accurate methodology for the transmission of error-feedback to the user. An intelligent system based on Gesture recognition methodologies and a real-time motion descriptor system is applied to engage and guide the users to perform an unknown movement in a natural and transparent process using the sound like the base of the transfer of the skill.
Research Interests:
This paper presents the application of the Particle Swarm Optimization (PSO) in synthetic aperture radar (SAR) methods exploiting multiple antennas for UHF-RFID tag localization. To perform 3D tag positioning, a robot equipped with two... more
This paper presents the application of the Particle Swarm Optimization (PSO) in synthetic aperture radar (SAR) methods exploiting multiple antennas for UHF-RFID tag localization. To perform 3D tag positioning, a robot equipped with two reader antennas is moving in the indoor scenario. The applicability of the PSO approach in SAR-based localization is discussed through a numerical analysis and an experimental campaign. A centimeter order localization error is achieved for 3D tag localization with a reduced computational cost with respect to classical SAR-based methods.
Research Interests:
The Fourth Industrial Revolution, or Industry 4.0, aims at automating traditional manufacturing and industrial practices using modern smart technology. Autonomous mobile robots equipped with different sensors and employing different... more
The Fourth Industrial Revolution, or Industry 4.0, aims at automating traditional manufacturing and industrial practices using modern smart technology. Autonomous mobile robots equipped with different sensors and employing different techniques have been proposed in the literature in the logistics and inventory warehouse management contexts. Efficient robot motion and planning depend on precise localization, mapping, and awareness of the environment. To properly localize items, recent attempts have been made employing radio frequency identification (RFID) in a 3D environment. This manuscript introduces four least mean squares methods to estimate the 3D positions of tags employing synthetic apertures and phase unwrapping. The proposed methods approach the localization problem solving a system of equations typical of multilateration methods to find the intersections of multiple hyperboloids. The novelty introduced here is the use of unwrapped phase distances to compute pseudo ranges fo...
Research Interests:
Welding automation is a fundamental process in manufacturing industries. Production lines integrate welding quality controls to reduce wastes and optimize the production chain. Early detection is fundamental as defects at any stage could... more
Welding automation is a fundamental process in manufacturing industries. Production lines integrate welding quality controls to reduce wastes and optimize the production chain. Early detection is fundamental as defects at any stage could determine the rejection of the entire product. In the last years, following the industry 4.0 paradigm, industrial automation lines have seen the introduction of modern technologies. Although the majority of the inspection systems still rely on traditional sensing and data processing, especially in the computer vision domain, some initiatives have been taken toward the employment of machine learning architectures. This chapter introduces deep neural networks in the context of welding defect detection, starting by analyzing common problems in the industrial applications of such technologies and discussing possible solutions in the specific case of quality checks in fuel injectors welding during the production stage.
Research Interests:
Research Interests:
The simulation of fabrics physics and its interaction with the human body has been largely studied in recent years to provide realistic-looking garments and wears specifically in the entertainment business. When the purpose of the... more
The simulation of fabrics physics and its interaction with the human body has been largely studied in recent years to provide realistic-looking garments and wears specifically in the entertainment business. When the purpose of the simulation is to obtain scientific measures and detailed mechanical properties of the interaction, the underlying physical models should be enhanced to obtain better simulation accuracy increasing the modeling complexity and relaxing the simulation timing constraints to properly solve the set of equations under analysis. However, in the specific field of haptic interaction, the desiderata are to have both physical consistency and high frame rate to display stable and coherent stimuli as feedback to the user requiring a tradeoff between accuracy and real-time interaction. This work introduces a haptic system for the evaluation of the fabric hand of specific garments either existing or yet to be produced in a virtual reality simulation. The modeling is based...
Research Interests:
This paper presents a particle swarm optimization (PSO) in synthetic aperture radar (SAR) methods to enable real-time 3D localization of UHF-RFID tags. The reader antennas are moved through a moving agent (e.g., unmanned aerial vehicle,... more
This paper presents a particle swarm optimization (PSO) in synthetic aperture radar (SAR) methods to enable real-time 3D localization of UHF-RFID tags. The reader antennas are moved through a moving agent (e.g., unmanned aerial vehicle, unmanned ground vehicle, robot) and they collect several phase samples by resembling a synthetic array. Thanks to the PSO approach the 3D matching function of the SAR method can be calculated in a reduced number of points by keeping an acceptable localization error. A numerical analysis demonstrates the method applicability through a comparison with conventional SAR methods based on the exhaustive search (3D dense grid) of the maximum point. Localization performance is investigated when an agent is equipped with a single antenna moving along a 3D trajectory or with two reader antennas at different height running a planar trajectory. Then, an experimental campaign in indoor scenario with an RFID-equipped unmanned ground vehicle shows the method effectiveness in performing real-time 3D positioning with centimeter-order localization error.
Research Interests:
Abstract Working machines in construction sites or emergency scenarios can operate in situations that can be dangerous for the operator. On the contrary, remote operation has been typically hindered by limited sense of presence of the... more
Abstract Working machines in construction sites or emergency scenarios can operate in situations that can be dangerous for the operator. On the contrary, remote operation has been typically hindered by limited sense of presence of the operator in the environment due to the reduced field of view of cameras. Starting from these considerations, this work introduces a novel real-time panoramic telepresence system for construction machines. This system does allow fully immersive operations in critical scenarios while keeping the operator in a safe location at safe distance from the construction operation. An omnidirectional stereo vision head mounted over the machine acquires and sends data to the operator with a streaming technique that focuses on the current direction of sight of the operator. The operator uses a head-mounted display to experience the remote site also with the possibility to view digital information overlaid to the remote scene as a type of augmented reality. The paper addresses the design and architecture of the system starting from the vision system and then proceeding to the immersive visualization.
Research Interests:
Learning to hand-write is a complex task and is fundamental for the development of a correct language reasoning and understanding. This work presents a multimodal system that focuses on the learning of handwriting movements. The system is... more
Learning to hand-write is a complex task and is fundamental for the development of a correct language reasoning and understanding. This work presents a multimodal system that focuses on the learning of handwriting movements. The system is thought for children usage providing engaging graphics and novel feedback technologies to accelerate the learning process. The main characteristic of the system architecture is its scalability in hardware and software components so that it is possible to employ it in any domestic setup.
Research Interests: Computer Science and IEEE
Research Interests:
Research Interests:
Capturing athletes performances with the purpose of skills training in the specific field of rowing sport is here presented, in particular The SPRINT multimodal system is introduced. This system is comprised of a mechanical reproduction... more
Capturing athletes performances with the purpose of skills training in the specific field of rowing sport is here presented, in particular The SPRINT multimodal system is introduced. This system is comprised of a mechanical reproduction of a rowing boat and of a virtual reality system with augmented feedback suited for novice and expert training. This paper details the implementation of an embedded acquisition system capable of measuring all the biomechanical data necessary for the rowing physical simulation increasing the performance with respect to the previous system.
Research Interests:
In this article we present a computer vision solution for stitching and reconstructing rolling stocks from a single fixed camera capture. First, we make a brief presentation of the State of the Art and the reason that make these algorithm... more
In this article we present a computer vision solution for stitching and reconstructing rolling stocks from a single fixed camera capture. First, we make a brief presentation of the State of the Art and the reason that make these algorithm not suitable for our purpose. Next, we presents the problem to solve and finally the results achieved.
This work presents a smart monitoring system capable of performing accurate quality checks to detect welding defects in fuel injectors. It employs classical computer vision for the geometrical analysis and Deep Neural Network for the... more
This work presents a smart monitoring system capable of performing accurate quality checks to detect welding defects in fuel injectors. It employs classical computer vision for the geometrical analysis and Deep Neural Network for the surface quality analysis. Despite the few training samples, the network has been trained successfully leveraging on the transfer learning and data augmentation techniques obtaining an accuracy of 97,22%.
This work presents a commercial collaborative robot that integrates a classical two-finger gripper and a custom-designed Universal Jamming Gripper (UJG). To exploit at best the potentialities of the UJG 3D perception technologies are... more
This work presents a commercial collaborative robot that integrates a classical two-finger gripper and a custom-designed Universal Jamming Gripper (UJG). To exploit at best the potentialities of the UJG 3D perception technologies are employed for selecting suitable geometries for picking objects of different shapes. The proposed system is competitive in terms of grasping success rate with respect to two state-of-the-art systems and outperforms them in the searching time of the grasping point.
Research Interests:
Research Interests:
Research Interests:
Visual quality inspection for defect detection is one of the main processes in modern industrial production facilities. In the last decades, artificial intelligence solutions took the place of classic computer vision techniques in the... more
Visual quality inspection for defect detection is one of the main processes in modern industrial production facilities. In the last decades, artificial intelligence solutions took the place of classic computer vision techniques in the production lines and specifically in tasks that, for their complexity, were usually demanded to human workers yet obtaining similar or greater performance of their counterparts. This work exploits a Deep Neural Network for a smart monitoring system capable of performing accurate quality checks to detect welding defects in fuel injectors during the production stage. The contribution focuses on a novel approach to cope with unforeseen changes in production quality introduced by the alteration of a particular machine or process. Results suggest that pre-filtering could avoid the retraining of custom-designed networks. Moreover, the introduction of a weighting strategy on the confusion matrix allows obtaining good performance estimations even in the case of small and unbalanced datasets. Concerning a specific demanding case of an imbalanced dataset with very few positive examples, the system displayed a 96.30% accuracy on defect classification.
Research Interests:
Modern industrial processes aim for the continuous production of small volumes tailored to the customer’s needs. Machines and robotic platforms have to be more and more adaptable, flexible, and able to cope with complex scenarios where... more
Modern industrial processes aim for the continuous production of small volumes tailored to the customer’s needs. Machines and robotic platforms have to be more and more adaptable, flexible, and able to cope with complex scenarios where sensing and manipulation capabilities are the key technology to succeed. The literature has plenty of capacitive, resistive, piezoelectric, and piezoresistive sensors used as tactile or force sensors. All of them present some drawbacks like non-linear behavior, sensitivity to temperature or electromagnetic noise, and hysteresis, among others. Other sensing systems are bulky and hard to integrate, sometimes jeopardizing the dexterity and manipulability of the gripper. In this context, the manuscript proposes fiber Bragg grating (FBG) optical fiber as a tactile sensing element to capture the interaction forces during material handling and object manipulation since it has numerous advantages compared with the other sensing devices. The work also offers a methodology to easily integrate the fiber in industrial grippers and introduces a set of tests useful to characterize the sensors. Custom gripper fingers have been realized in rapid prototyping to present a pictorial example of such an integration. Finally, the essay presents some experiments that assess the capability of a tactile sensor based on FBG optical fiber showing as it can correctly perceive the contact forces (NRMSE = 0.75%) and can recognize the material of the object that is being manipulated. The authors believe that the application of optical fiber sensor as tactile feedback can be useful in industrial scenarios to enable complex manipulation activities.
Research Interests:
Research Interests:
This work addresses the problem of semi-automatic inspection and navigation in confined environments. A system that overcomes many challenges at the state of the art is presented. It comprises a multirotor able to inspect an industrial... more
This work addresses the problem of semi-automatic inspection and navigation in confined environments. A system that overcomes many challenges at the state of the art is presented. It comprises a multirotor able to inspect an industrial combustion chamber thus working in a GPS-denied environment with poor lighting conditions, in the presence of magnetic and communication disturbances, iron dust and repetitive patterns on the structure walls. The presented system is able to pass through narrow entrances but still capable of acquiring high resolution images and to allow operators to perform inspection of the structures. Starting from the captured data, the system is able to provide a 3D reconstruction of the inspected environment for offline analysis.
Research Interests:
This paper introduces an intelligent system able to perform quality control assessment in an industrial production line. Deep learning techniques are employed and proved successful in a real application for the inspection of welding... more
This paper introduces an intelligent system able to perform quality control assessment in an industrial production line. Deep learning techniques are employed and proved successful in a real application for the inspection of welding defects on an assembly line of fuel injectors. Starting from state-of-the-art deep architectures and using the transfer learning technique, it is possible to train a network with about 7 million parameters using a reduced number of injector's images, obtaining an accuracy of 97.22%. The system is also configured in order to exploit new data, collected during operation, to extend the existing dataset and to improve further its performance. The developed system shows that deep neural networks can successfully perform quality inspection tasks that are usually demanded to humans.
Research Interests:
Working machines in construction sites or emergency scenarios can operate in situations that can be dangerous for the operator. On the contrary, remote operation has been typically hindered by limited sense of presence of the operator in... more
Working machines in construction sites or emergency scenarios can operate in situations that can be dangerous for the operator. On the contrary, remote operation has been typically hindered by limited sense of presence of the operator in the environment due to the reduced field of view of cameras. Starting from these considerations, this work introduces a novel real-time panoramic telepresence system for construction machines. This system does allow fully immersive operations in critical scenarios while keeping the operator in a safe location at safe distance from the construction operation. An omnidirectional stereo vision head mounted over the machine acquires and sends data to the operator with a streaming technique that focuses on the current direction of sight of the operator. The operator uses a head-mounted display to experience the remote site also with the possibility to view digital information overlaid to the remote scene as a type of augmented reality. The paper addresses the design and architecture of the system starting from the vision system and then proceeding to the immersive visualization.