Daniel Roggen
University of Sussex, Engineering and Design, Faculty Member
Research Interests:
ABSTRACT There is a need for event organizers and emergency response personnel to detect emerging, potentially critical crowd situations at an early stage during city-wide mass gatherings. In this work, we introduce and describe... more
ABSTRACT There is a need for event organizers and emergency response personnel to detect emerging, potentially critical crowd situations at an early stage during city-wide mass gatherings. In this work, we introduce and describe mathematical methods based on pedestrian-behavior models to infer and visualize crowd conditions from pedestrians' GPS location traces. We tested our approach during the 2011 Lord Mayor's Show in London by deploying a system able to infer and visualize in real-time crowd density, crowd turbulence, crowd velocity and crowd pressure. To collection location updates from festival visitors, a mobile phone app that supplies the user with event-related information and periodically logs the device's location was distributed. We collected around four million location updates from over 800 visitors. The City of London Police consulted the crowd condition visualization to monitor the event. As an evaluation of the usefulness of our approach, we learned through interviews with police officers that our approach helps to assess occurring crowd conditions and to spot critical situations faster compared to the traditional video-based methods. With that, appropriate measure can be deployed quickly helping to resolve a critical situation at an early stage.
Research Interests:
Abstract. A robust activity and context-recognition system must be capable of operating over a long period of time, exploiting new sources of information as they become available and evolving in an autonomous manner, coping with user... more
Abstract. A robust activity and context-recognition system must be capable of operating over a long period of time, exploiting new sources of information as they become available and evolving in an autonomous manner, coping with user variability and changes in the number and ...
Research Interests:
Research Interests:
Snowboarding is one of the fastest growing sports in the world. However, it is rather difficult to learn. Snowboarders, who have reached an average level of expertise, often find it difficult to improve their style without taking... more
Snowboarding is one of the fastest growing sports in the world. However, it is rather difficult to learn. Snowboarders, who have reached an average level of expertise, often find it difficult to improve their style without taking expensive private lessons. Moreover, some of their ...
Research Interests:
Research Interests:
Research Interests:
ABSTRACT Flying is exciting - but can also be exhausting or even harmful. Dehydration due to dry air and lack of movement promote the development of deep vein thrombosis (DVT). Moreover, many passengers feel uneasy in airplanes whereas... more
ABSTRACT Flying is exciting - but can also be exhausting or even harmful. Dehydration due to dry air and lack of movement promote the development of deep vein thrombosis (DVT). Moreover, many passengers feel uneasy in airplanes whereas some even experience severe anxiety attacks. Within the EU project SEAT, we developed pervasive sensing technologies and integrated them into an airplane seat. They allow to sense physiological parameters related to the passenger’s comfort and affective state. This video shows how pervasive computing can provide support to airplane passengers, right when they need it.
Research Interests:
Research Interests:
Research Interests: Computer Science, Artificial Intelligence, Machine Learning, Human Factors, Pattern Recognition, and 15 moreActivity Recognition, Modalities, Best practice, Human behavior, Data acquisition, Lead, Pattern Classification, Self Organization, Lessons Learned, Humidity, Human Activity, Packet Loss, Large Scale, Data Synchronization, and Heterogeneous Network
Research Interests:
ABSTRACT A real-time understanding of the behavior of pedestrian crowds in physical spaces is important for crowd monitoring and management during large-scale mass gatherings. Thanks to the proliferation of location-aware smartphones in... more
ABSTRACT A real-time understanding of the behavior of pedestrian crowds in physical spaces is important for crowd monitoring and management during large-scale mass gatherings. Thanks to the proliferation of location-aware smartphones in our society, we see a big potential in inferring crowd behavior patterns by tracking the location of attendees via their mobile phones. This chapter describes a framework to infer and visualize crowd behavior patterns in real-time, using a specially developed smartphone app. Attendees at an event voluntarily provide their location updates and in return may receive timely, targeted and personalized notifications directly from the security personnel which can be of help during an emergency situation. Users also have access to event-related information including travel advice to the location. We conducted a systems trial during the Lord Mayor’s Show 2011 in London, UK and the Notte Bianca festival 2011 in Valletta, Malta. In this chapter, besides verifying the technological feasibility, we report on interviews conducted with app users and police forces that were accessing the monitoring tools during the event. We learned from both sides that the created feedback loop between the attendees of the event running the app and the security personnel is seen as a strong incentive to follow such a participatory sensing approach. The researchers worked closely with policy makers, the emergency services and event organisers and policy implications of using the Socionical App will be discussed; as well as the response of users to being guided by an AmI device during a possible emergency.
Research Interests:
There is a need for event organizers and emergency response personnel to detect emerging, potentially critical crowd situations at an early stage during city-wide mass gatherings. In this work, we present a framework to infer and... more
There is a need for event organizers and emergency response personnel to detect emerging, potentially critical crowd situations at an early stage during city-wide mass gatherings. In this work, we present a framework to infer and visualize crowd behavior patterns in real-time from pedestrians’ GPS location traces. We deployed and tested our framework during the 2011 Lord Mayor’s Show in London. To collection location updates from festival visitors, a mobile phone app that supplies the user with event-related information and periodically logs the device’s location was distributed. We collected around four million location updates from over 800 visitors. The City of London Police consulted the crowd condition visualization to monitor the event. We learned from the police officers that our framework helps to assess occurring crowd conditions and to spot critical situations faster compared to the traditional video-based methods. With that, appropriate measure can be deployed quickly helping to resolve a critical situation at an early stage.
Research Interests:
Research Interests:
Summary Objectives: In this work the effect of quasi-stationary movements on the electrodermal activity (EDA) after a startle event has been investigated and evaluated. In previous EDA research there is a discrepancy between the use of... more
Summary Objectives: In this work the effect of quasi-stationary movements on the electrodermal activity (EDA) after a startle event has been investigated and evaluated. In previous EDA research there is a discrepancy between the use of controlled environment studies and daily life surveys. This paper aims to address this by expanding the knowledge about EDA in real life applications. Methods: A minimally obtrusive body-worn measurement device was designed and produced that simultaneously records EDA and finger movements. During this study, five subjects walked at different speeds and listened to startling sound events. The EDA response to these startle events was analyzed for different walking speeds using crosscorrelograms and cumulative frequency plots. Results: The measured response to the startle event is consistent with the signal characteristics described in the literature. The results show that the faster a person is walking the more the signal property of the phasic part of ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Abstract. Wearable computers, embedded in clothing or seamlessly in-tegrated in devices we carry with us, have a tremendous advantage to become the main gateway to personal health management. Current state-of-the-art devices allow to... more
Abstract. Wearable computers, embedded in clothing or seamlessly in-tegrated in devices we carry with us, have a tremendous advantage to become the main gateway to personal health management. Current state-of-the-art devices allow to monitor basic physical or ...
Research Interests:
... Daniel Roggen, Wearable Computing Lab, ETH Zurich, droggen@ife.ee.ethz.ch Alois Ferscha, Institute for Pervasive Computing, University of Linz Gerhard Tröster, Wearable Computing Lab, ETH ... R. Murty, G. Mainland, I. Rose, AR... more
... Daniel Roggen, Wearable Computing Lab, ETH Zurich, droggen@ife.ee.ethz.ch Alois Ferscha, Institute for Pervasive Computing, University of Linz Gerhard Tröster, Wearable Computing Lab, ETH ... R. Murty, G. Mainland, I. Rose, AR Chowdhury, A. Gosain, J. Bers, and M. Welsh. ...
Abstract—The SOCIONICAL project aims to develop complexity science based modeling, prediction and simulation methods for large scale socio-technical systems in an AmI based smart environment. Focusing on one of the two scenarios of the... more
Abstract—The SOCIONICAL project aims to develop complexity science based modeling, prediction and simulation methods for large scale socio-technical systems in an AmI based smart environment. Focusing on one of the two scenarios of the project (ie crowd evacuation), in this paper, we have presented the scenario based modeling approach utilized. The position of ambient technology of choice (ie LifeBelt) in the modeling approach is discussed with realization of required enhancement to support sensing of the collective ...
Research Interests:
AbstractWearable computing aims to empower people by providing intelligence embedded within garments. It relies on sensors placed at different locations of the body. To foster user-acceptance sensors should be small, light, and... more
AbstractWearable computing aims to empower people by providing intelligence embedded within garments. It relies on sensors placed at different locations of the body. To foster user-acceptance sensors should be small, light, and unobtrusive. In this paper we present a wearable ...
Research Interests:
Research Interests: Engineering and Sensors
Research Interests:
Research Interests:
ABSTRACT Download Citation Email Print Request Permissions Save to Project We present SmartActionSLAM, an Android smartphone application that performs location tracking in home and office environments. It uses the integrated motion... more
ABSTRACT Download Citation Email Print Request Permissions Save to Project We present SmartActionSLAM, an Android smartphone application that performs location tracking in home and office environments. It uses the integrated motion sensors of the smartphone and an optional foot-mounted inertial measurement unit to track a person. The application implements an instance of the ActionSLAM algorithmic framework. ActionSLAM combines pedestrian dead reckoning with the observation of activities (in SmartActionSLAM: sitting and standing still) to build and update a local landmark map of the user's environment. This map is used to compensate for error accumulation of dead reckoning in a particle filter framework. We show that it is possible to execute the ActionSLAM algorithm in real-time on a smartphone without platform-specific optimizations. Furthermore, we analyze the localization performance of the application in six constrained and two real-life recordings. When using only the smartphone's internal sensors, tracking was adequate in most constrained setups, but failed in the real-world scenarios because of errors in recognizing irregular leg movements. By including the foot-mounted sensor, mapping with a mean landmark positioning error of < 0.5m and robustness > 90% was achieved in all environments. Smart-ActionSLAM is fully wearable and requires no infrastructure in the environment. The approach is therefore ideally suited for rapid deployment in home and office environments, as for example required in patient monitoring studies.
Research Interests:
Community intelligence is often manifested in distinct collective behavior patterns. We investigate on the exemplary use case of paragliding how real-time participatory mobile sensing can be exploited to infer collective behavior patterns... more
Community intelligence is often manifested in distinct collective behavior patterns. We investigate on the exemplary use case of paragliding how real-time participatory mobile sensing can be exploited to infer collective behavior patterns and to conclude about community intelligence. In particular, we present a system to simultaneously aggregate flight information from many paraglider pilots using their location-aware mobile phones. We show
Research Interests:
Research Interests: Artificial Intelligence, Machine Learning, Ubiquitous Computing, Wearable Computing, Human Factors, and 21 morePattern Recognition, Wireless Sensor Networks, Activity Recognition, Best practice, Human behavior, Data acquisition, Lead, Sensor Network, Pattern Classification, Sensorimotor synchronisation, Wearable Computer, Self Organization, Lessons Learned, Wireless Sensor, Humidity, Human Activity, Packet Loss, Large Scale, Sensor system, Data Synchronization, and Heterogeneous Network
ABSTRACT The next generation of ambient intelligence environments will support people throughout their lives with much greater autonomy and capacity to assist than nowadays. To achieve this, a shift is required away from one mainstream... more
ABSTRACT The next generation of ambient intelligence environments will support people throughout their lives with much greater autonomy and capacity to assist than nowadays. To achieve this, a shift is required away from one mainstream view in ubiquitous computing: the ``handcrafting'' of extremely application specific and narrowly defined "context-aware assistants". Instead, ambient intelligence and physically embedded intelligent systems (PEIS) will converge. A key for this to happen is for PEIS to be context-aware: they must especially be able to infer the activities, behaviors, intentions, and even emotions of users in their daily life. As the environment can be very dynamic, with changing resources or user expectations, so must the PEIS. We approach this problem by an ecology of Context Cells: items with communication and processing capabilities geared at discovering and classifying patterns of human behaviors. The physica implementation of a Context Cell is a situated sensor-actuator node. Individual "specialized" cells are independently capable of context awareness, and capable of adaptation. Cells cooperate with each other to recognize more complex pieces of context, increase performance, or increase robustness to faults. Finally, cells can train other "undifferentiated" cells which are newly discovered in the ambient intelligence environment. This allows to extend the coverage of context awareness through the environment by spreading capabilities among cells. The ecology of Context Cells forms the substrate for ambient intelligence environments capable of lifelong adaptation and evolution guided by the user's expectations.
ABSTRACT The field of research on activity recognition is relatively young compared to others, like computer vision. In more mature fields, algorithms are usually tested on standardized, reference datasets. This way, algorithms coming... more
ABSTRACT The field of research on activity recognition is relatively young compared to others, like computer vision. In more mature fields, algorithms are usually tested on standardized, reference datasets. This way, algorithms coming from different groups can be tested in a fair manner, which accelerates the process of developing new knowledge. Collecting a reference dataset under realistic settings for activity recognition poses many challenges due to the large amount of sensors and sensor modalities which are needed to provide a sufficiently complete playground. We here report on some lessons learned while collecting such a reference dataset with a heterogeneous setup. We argue for the importance of a few principles to obtain a clean dataset, starting from the sampling and acquisition, down to the synchronization and labeling of the data.
Research Interests:
Research Interests:
Research Interests:
Abstract. Quality of Context (QoC) in context-aware computing im-proves reasoning and decision making. Activity recognition in wearable computing enables context-aware assistance. Wearable systems must in-clude QoC to participate in... more
Abstract. Quality of Context (QoC) in context-aware computing im-proves reasoning and decision making. Activity recognition in wearable computing enables context-aware assistance. Wearable systems must in-clude QoC to participate in context processing frameworks common in ...
Research Interests:
ABSTRACT We present and evaluate a microcontroller-optimized limited-memory implementation of a Warping Longest Common Subsequence algorithm (WarpingLCSS). It permits to spot patterns within noisy sensor data in real-time in resource... more
ABSTRACT We present and evaluate a microcontroller-optimized limited-memory implementation of a Warping Longest Common Subsequence algorithm (WarpingLCSS). It permits to spot patterns within noisy sensor data in real-time in resource constrained sensor nodes. It allows variability in the sensed system dynamics through warping; it uses only integer operations; it can be applied to various sensor modalities; and it is suitable for embedded training to recognize new patterns. We illustrate the method on 3 applications from wearable sensing and activity recognition using 3 sensor modalities: spotting the QRS complex in ECG, recognizing gestures in everyday life, and analyzing beach volleyball. We implemented the system on a low-power 8-bit AVR wireless node and a 32-bit ARM Cortex M4 microcontroller. Up to 67 or 140 10-second gestures can be recognized simultaneously in real-time from a 10Hz motion sensor on the AVR and M4 using 8mW and 10mW respectively. A single gesture spotter uses as few as 135μW on the AVR. The method allows low data rate distributed in-network recognition and we show a 100 fold data rate reduction in a complex activity recognition scenario. The versatility and low complexity of the method makes it well suited as a generic pattern recognition method and could be implemented as part of sensor front-ends.
Detecting pedestrians moving together through public spaces can provide relevant information for many location-based social applications. In this work we present an online method to detect such pedestrian flocks by spatio-temporal... more
Detecting pedestrians moving together through public spaces can provide relevant information for many location-based social applications. In this work we present an online method to detect such pedestrian flocks by spatio-temporal clustering of location trajectories. Compared to prior work, our method provides increased robustness against the influence of noisy and missing GPS data often encountered in urban environments. To assess
Research Interests:
Research Interests:
ABSTRACT Previous work on the recognition of human movement patterns has mainly focused on movements of individuals. This paper addresses the joint identification of the indoor movement of multiple persons forming a cohesive whole -... more
ABSTRACT Previous work on the recognition of human movement patterns has mainly focused on movements of individuals. This paper addresses the joint identification of the indoor movement of multiple persons forming a cohesive whole - specifically a flock - with clustering approaches operating on features derived from multiple sensor modalities of modern smartphones. Automatic detection of flocks has several important applications, including evacuation management and socially aware computing. The novelty of this paper is, firstly, to use data fusion techniques to combine several sensor modalities (WiFi, accelerometer and compass) to improve recognition accuracy over previous unimodal approaches. Secondly, improve the recognition of flocks using hierarchical clustering. We use a dataset comprising 16 subjects forming one to four flocks walking in a building on single and multiple floors. With the best settings, we achieve a F-score accuracy of up to 87 percent an improvement of up to twelve percent points over existing approaches.