2008 Journal article Open Access

See at: Archive ouverte UNIGE | The Visual Computer | RERO DOC Digital Library | Infoscience - EPFL scientific publications | The Visual Computer | www.springerlink.com | CNR ExploRA

See at: ieeexplore.ieee.org | CNR ExploRA

See at: CNR ExploRA

See at: CNR ExploRA

2006 Report Open Access

See at: ISTI Repository | CNR ExploRA

Predicting user movements in heterogeneous indoor environments by reservoir computing
Bacciu D., Barsocchi P., Chessa S., Gallicchio C., Micheli A.
Anticipating user localization by making accurate predictions on its indoor movement patterns is a fundamental challenge for achieving higher degrees of personalization and reactivity in smart-home environments. We propose an approach to real-time movement forecasting founding on the efficient Reservoir Computing paradigm, predicting user movements based on streams of Received Signal Strengths collected by wireless motes distributed in the home environment. The ability of the system to generalize its predictive performance to unseen ambient configurations is experimentally assessed in challenging conditions, comprising external test scenarios collected in home environments that are not included in the training set. Experimental results suggest that the system can effectively generalize acquired knowledge to novel smart-home setups, thereby delivering an higher level of personalization while decreasing costs for installation and setup.Source: Space, Time and Ambient Intelligence Workshop. International Joint Conference on Artificial Intelligence, pp. 1–6, Barcelona, Spain, 16 July 2011
Project(s): RUBICON

See at: ijcai-11.iiia.csic.es | CNR ExploRA

User movements forecasting by reservoir computing using signal streams produced by mote-class sensors
Barsocchi Paolo, Chessa Stefano, Gallicchio Claudio, Micheli Alessio
Real-time, indoor user localization, although limited to the current user position, is of great practical importance in many Ambient Assisted Living (AAL) applications. Moreover, an accurate prediction of the user next position (even with a short advice) may open a number of new AAL applications that could timely provide the right services in the right place even before the user request them. However, the problem of forecasting the user position is complicated due to the intrinsic difficulty of localization in indoor environments, and to the fact that different paths of the user may intersect at a given point, but they may end in different places. We tackle with this problem by modeling the localization information stream obtained from a Wireless Sensor Network (WSN) using Recurrent Neural Networks implemented as efficient Echo State Networks (ESNs), within the Reservoir Computing paradigm. In particular, we have set up an experimental test-bed in which the WSN produces localization information of a user that moves along a number of different paths, and in which the ESN collects localization information to predict a future position of the user at some given mark points. Our results show that, with an appropriate configuration of the ESN, the system reaches a good accuracy of the prediction also with a small WSN, and that the accuracy scales well with the WSN size. Furthermore, the accuracy is also reasonably robust to variations in the deployment of the sensors. For these reason our solution can be configured to meet the desired trade-off between cost and accuracySource: ISTI Technical reports, 2011
Project(s): RUBICON

See at: CNR ExploRA

In gara le tecnologie per la casa sicura
Barsocchi P.
Le prime due tappe di 'Evaal', il concorso internazionale dedicato alle applicazioni più 'smart' per l'assistenza domestica automatizzata organizzato per il secondo anno consecutivo dal laboratorio 'Wireless network' (Wn Lab) dell'Istituto di scienze e tecnologie dell'informazione 'A. Faedo' (Isti) del Cnr di Pisa, si sono svolte a Madrid ('Localizzazione in ambiente domestico') e a Valencia ('Riconoscimento delle attività domestiche'). L'ultima fase, che ha visto la proclamazione e la premiazione dei vincitori, si è tenuta lo scorso settembre a Eindhoven.Source: Almanacco della scienza 16 (2012).
Project(s): UNIVERSAAL

See at: www.almanacco.cnr.it | CNR ExploRA

EvAAL, Evaluating AAL systems through competitive benchmarking, the experience of the 1st competition.
Salvi D., Làzaro J. P., Arredondo M. T., Barsocchi P.
As Ambient Assisted Living (AAL) emerges as a need for our ageing societies, many barriers are still in place against its wide adop- tion. One of the main issues related to the creation of an AAL market is the lack of consensus around well established technologies which should e ectively cover real needs of the population. EvAAL (Evaluating AAL Systems Through Competitive Benchmarking) is a newborn initiative aimed at evaluating solutions related to Ambient Assisted Living by or- ganizing annual international competitions. Its main objectives are the creation of a community of stakeholders around AAL and the creation of metrics and benchmarks for both innovative prototypes and commer- cial solutions. EvAAL focuses not only on comparison of algorithms or speci c hardware issues, but also of user acceptance, deployment and installation e ort, integrability, etc. In its rst versions, the competition is focusing on speci c technical aspects of AAL but aims, in the near future, at joining heterogeneous "ambient" technologies in a common evaluation framework. In July 2011, the rst EvAAL competition took place in Valencia, Spain, on Indoor Localization and Tracking for AAL. This paper describes how EvAAL is designed, its principles and how it is internally organized, and goes though an evaluation of this structure though the experience gained during the rst competition.Source: ISTI Technical reports, pp.1–13, 2012
Project(s): UNIVERSAAL

See at: CNR ExploRA

An Experimental Evaluation of Reservoir Computation for Ambient Assisted Living.
Bacciu D., Barsocchi P., Chessa S., Gallicchio C., Micheli A.
In this paper we investigate the introduction of Reservoir Computing (RC) neural network models in the context of AAL (Ambient Assisted Living) and self-learning robot ecologies, with a focus on the computational constraints related to the implementation over a network of sensors. Specifically, we experimentally study the relationship between architectural parameters influencing the computational cost of the models and the performance on a task of user movements prediction from sensors signal streams. The RC shows favorable scaling properties results for the analyzed AAL task.Source: ISTI Technical reports, pp.1–10, 2012
Project(s): RUBICON

See at: CNR ExploRA

EvAAL, Evaluating AAL systems through competitive benchmarking, the experience of the 1st competition
Salvi D., Barsocchi P., Arredondo M. T., Làzaro Ramos J. P.
As Ambient Assisted Living (AAL) emerges as a need for our ageing societies, many barriers are still in place against its wide adop- tion. One of the main issues related to the creation of an AAL market is the lack of consensus around well established technologies which should e ectively cover real needs of the population. EvAAL (Evaluating AAL Systems Through Competitive Benchmarking) is a newborn initiative aimed at evaluating solutions related to Ambient Assisted Living by or- ganizing annual international competitions. Its main objectives are the creation of a community of stakeholders around AAL and the creation of metrics and benchmarks for both innovative prototypes and commer- cial solutions. EvAAL focuses not only on comparison of algorithms or speci c hardware issues, but also of user acceptance, deployment and installation e ort, integrability, etc. In its rst versions, the competition is focusing on speci c technical aspects of AAL but aims, in the near future, at joining heterogeneous "ambient" technologies in a common evaluation framework. In July 2011, the rst EvAAL competition took place in Valencia, Spain, on Indoor Localization and Tracking for AAL. This paper describes how EvAAL is designed, its principles and how it is internally organized, and goes though an evaluation of this structure though the experience gained during the first competition.Source: Evaluating AAL Systems Through Competitive Benchmarking. Indoor Localization and Tracking, edited by Stefano Chessa, Stefan Knauth, pp. 14–25. Berlin: Springer, 2012
DOI: 10.1007/978-3-642-33533-4
Project(s): UNIVERSAAL
Metrics:

See at: doi.org | link.springer.com | CNR ExploRA

Activity recognition by reservoir computing using signal streams produced by wireless sensor devices
Barsocchi P., Palumbo F. Gallicchio C., Chessa S., Micheli A.
Ambient Assisted Living facilities provide assistance and care for the elderly, where it is useful to infer their daily activity for ensuring their safety and successful ageing. In this work, we present an activity recognition system that classifies a set of common daily activities exploit- ing both the data sampled by accelerometer sensors carried out by the user, and the Receive Signal Strength (RSS) values coming from wireless sensors devices deployed in the environment. To this end, accelerometer and RSS streams, obtained from a Wireless Sensor Network (WSN), are treated using Recurrent Neural Networks implemented as efficient Echo State Networks (ESNs), within the Reservoir Computing paradigm. Our results show that, with an appropriate configuration of the ESN, the system reaches a good accuracy with a low energy cost.Source: ISTI Technical reports, 2013
Project(s): UNIVERSAAL

See at: CNR ExploRA

EvAAL'12 awards
Alvarez-Garcia J. A., Barsocchi P., Chessa S., Salvi D.
"Evaluating Ambient Assisted Living Systems through Competitive Benchmarking" (EvAAL) is an annual international competition that addresses the "grand" challenge of evaluation and comparison of Ambient Assisted Living (AAL) systems and plat- forms, with the final goal to assess the autonomy, in- dependent living and quality of life that AAL systems may grant to their end users. The EvAAL event is a live competition and this year edition is focused on two pillars of AAL: Indoor localization and activity recognition.Source: Journal of ambient intelligence and smart environments (Print) 4 (2012): 565–567. doi:10.3233/AIS-2012-0182
DOI: 10.3233/ais-2012-0182
Project(s): UNIVERSAAL
Metrics:

See at: iospress.metapress.com | CNR ExploRA

2014 Journal article Open Access

An experimental characterization of reservoir computing in ambient assisted living applications
Bacciu D., Barsocchi P., Chessa S., Gallicchio C., Micheli A.
In this paper, we present an introduction and critical experimental evaluation of a reservoir computing (RC) approach for ambient assisted living (AAL) applica- tions. Such an empirical analysis jointly addresses the issues of efficiency, by analyzing different system config- urations toward the embedding into computationally con- strained wireless sensor devices, and of efficacy, by analyzing the predictive performance on real-world appli- cations. First, the approach is assessed on a validation scheme where training, validation and test data are sampled in homogeneous ambient conditions, i.e., from the same set of rooms. Then, it is introduced an external test set involving a new setting, i.e., a novel ambient, which was not available in the first phase of model training and vali- dation. The specific test-bed considered in the paper allows us to investigate the capability of the RC approach to discriminate among user movement trajectories from received signal strength indicator sensor signals. This capability can be exploited in various AAL applications targeted at learning user indoor habits, such as in the roposed indoor movement forecasting task. Such a joint analysis of the efficiency/efficacy trade-off provides novel insight in the concrete successful exploitation of RC for AAL tasks and for their distributed implementation into wireless sensor networks.Source: Neural computing & applications (Print) 24 (2014): 1451–1464. doi:10.1007/s00521-013-1364-4
DOI: 10.1007/s00521-013-1364-4
Project(s): RUBICON
Metrics:

See at: Neural Computing and Applications | Neural Computing and Applications | link.springer.com | CNR ExploRA

An experimental evaluation of reservoir computation for ambient assisted living
Bacciu D., Chessa S., Gallicchio C., Micheli A., Barsocchi P.
In this paper we investigate the introduction of Reservoir Computing (RC) neural network models in the context of AAL (Ambient Assisted Living) and self-learning robot ecologies, with a focus on the computational constraints related to the implementation over a network of sensors. Specifically, we experimentally study the relationship between architectural parameters influencing the computational cost of the models and the performance on a task of user movements prediction from sensors signal streams. The RC shows favorable scaling properties results for the analyzed AAL task.Source: Neural Nets and Surroundings, edited by Bruno Apolloni, Simone Bassis, Anna Esposito, Francesco Carlo Morabito, pp. 41–50. Heidelberg: Springer, 2013
DOI: 10.1007/978-3-642-35467-0_5
Project(s): RUBICON
Metrics:

See at: doi.org | link.springer.com | CNR ExploRA

Evaluating AAL systems through competitive benchmarking - International competitions and final workshop
Botía J. A., Álvarez García J. A., Fujinami K., Barsocchi P., Riedel T.
This book is a compilation of papers describing the systems participating in the third edition of the EvAAL competition. The first edition in 2011 was centered on indoor localization. The second edition in 2012 included a track on activity recognition. This third edition in 2013 maintained both tracks.Source: Berlin: Springer-Verlag, 2013
DOI: 10.1007/978-3-642-41043-7
Project(s): UNIVERSAAL
Metrics:

See at: doi.org | link.springer.com | CNR ExploRA

2013 Journal article Open Access

Position recognition to support bedsores prevention
Barsocchi P.
A feasibility study where small wireless devices are used to classify some typical users positions in the bed is presented. Wearable wireless low-cost commercial transceivers operating at 2.4 GHz are supposed to be widely deployed in indoor settings and on peoples bodies in tomorrows pervasive computing environments. The key idea of this work is to leverage their presence by collecting the received signal strength (RSS) measured among fixed devices, deployed in the environment, and the wearable one. The RSS measurements are used to classify a set of users positions in the bed, monitoring the activities of patients unable to make the desirable bodily movements. The collected data are classified using both Support Vector Machine and K-Nearest Neighbour methods, in order to recognize the different users position, and thus supporting the bedsores issue.Source: IEEE Journal of Biomedical and Health Informatics 17 (2013): 53–59. doi:10.1109/TITB.2012.2220374 ?
DOI: 10.1109/titb.2012.2220374
Project(s): UNIVERSAAL
Metrics:

See at: ISTI Repository | IEEE Journal of Biomedical and Health Informatics | ieeexplore.ieee.org | CNR ExploRA

See at: CNR ExploRA | wnet.isti.cnr.it

See at: CNR ExploRA

See at: CNR ExploRA