Task-oriented optimal Perception-Action coupling for robotic systems
Topic: Adaptive control | All
Thursday 6th May 2021, 2pm – 3pm (Paris time).
SPEAKER
Paolo Salaris, Department of Information Engineering, University of Pisa, Italy (https://www.centropiaggio.unipi.it/~salaris)
ABSTRACT
The perception-action loop is arguably one of the defining features of robotics: a robot needs to move for realizing a task, and its choice for a motion plan depends on its internal knowledge of the ‘world’. However, onboard sensors can typically only provide partial information about the world and, therefore, some level of reconstruction/inference/estimation is needed online for recovering any information not directly available from the raw sensor readings. One of the aspects that makes the action-perception loop scientifically very interesting (and challenging) is the tight coupling between action and perception performance: in order to successfully realize a task, the robot requires an accurate world model (which can include internal state estimation, parameter self-calibration, reconstruction of the surrounding environment, and so on). However, the chosen motion and actions can also have a strong influence on the accuracy of the estimated world model, with some motions/actions more informative than other ones. Therefore, there exists a need of finding a balance between choosing motion/action plans that are ‘optimal’ for the sake of world modelling but, at the same time, are also effective in accomplishing the tasks of interest. In this talk, the problem of optimal action/perception coupling will be discussed and a solution described as well as its possible extension to the problem of shared control and multi-robot systems.
BIO
Paolo Salaris received his Degree in Electrical Engineering from the University of Pisa in 2007. He got the Doctoral degree in Robotics, Automation and Bioengineering at the Research Center “E.Piaggio” of the University of Pisa in June 2011 (advisor Antonio Bicchi). During his PhD, he has been Visiting Scholar at Beckman Institute for Advanced Science and Technology, University of Illinois, Urbana-Champaign (US-IL) from March to October 2009 (advisor Seth Hutchinson). He has been a PostDoc at the Research Center “E.Piaggio” from June 2011 to January 2014, working within the robotics group headed by Antonio Bicchi and from February 2014 to July 2015 at LAAS-CNRS in Toulouse within the Gepetto Team headed by Jean-Paul Laumond. He spent few months at College de France, Paris, in 2013. From October 2015 to August 2019 he has been an Inria researcher (CRCN) at Inria Sophia Antipolis-Méditerranée, France, within the Lagadic Team, headed by François Chaumette, until December 2017 and within the CHORALE Team, headed by Philippe Martinet, from January 2018 to August 2019.
He is currently an Assistant Professor (RTD-B) at the Department of Information Engineering (DII) and the Research Centre “E. Piaggio”, University of Pisa.
His research interests include optimal control and path planning for robots with limited sensory systems, visual servoing, map building of indoor and outdoor complex environments, observability and controllability analysis, optimal sensing design, optimal estimation and active sensing control, action-perception coupling and sensors-based feedback control and planning.
SPEAKER
Paolo Salaris, Department of Information Engineering, University of Pisa, Italy (https://www.centropiaggio.unipi.it/~salaris)
ABSTRACT
The perception-action loop is arguably one of the defining features of robotics: a robot needs to move for realizing a task, and its choice for a motion plan depends on its internal knowledge of the ‘world’. However, onboard sensors can typically only provide partial information about the world and, therefore, some level of reconstruction/inference/estimation is needed online for recovering any information not directly available from the raw sensor readings. One of the aspects that makes the action-perception loop scientifically very interesting (and challenging) is the tight coupling between action and perception performance: in order to successfully realize a task, the robot requires an accurate world model (which can include internal state estimation, parameter self-calibration, reconstruction of the surrounding environment, and so on). However, the chosen motion and actions can also have a strong influence on the accuracy of the estimated world model, with some motions/actions more informative than other ones. Therefore, there exists a need of finding a balance between choosing motion/action plans that are ‘optimal’ for the sake of world modelling but, at the same time, are also effective in accomplishing the tasks of interest. In this talk, the problem of optimal action/perception coupling will be discussed and a solution described as well as its possible extension to the problem of shared control and multi-robot systems.
BIO
Paolo Salaris received his Degree in Electrical Engineering from the University of Pisa in 2007. He got the Doctoral degree in Robotics, Automation and Bioengineering at the Research Center “E.Piaggio” of the University of Pisa in June 2011 (advisor Antonio Bicchi). During his PhD, he has been Visiting Scholar at Beckman Institute for Advanced Science and Technology, University of Illinois, Urbana-Champaign (US-IL) from March to October 2009 (advisor Seth Hutchinson). He has been a PostDoc at the Research Center “E.Piaggio” from June 2011 to January 2014, working within the robotics group headed by Antonio Bicchi and from February 2014 to July 2015 at LAAS-CNRS in Toulouse within the Gepetto Team headed by Jean-Paul Laumond. He spent few months at College de France, Paris, in 2013. From October 2015 to August 2019 he has been an Inria researcher (CRCN) at Inria Sophia Antipolis-Méditerranée, France, within the Lagadic Team, headed by François Chaumette, until December 2017 and within the CHORALE Team, headed by Philippe Martinet, from January 2018 to August 2019.
He is currently an Assistant Professor (RTD-B) at the Department of Information Engineering (DII) and the Research Centre “E. Piaggio”, University of Pisa.
His research interests include optimal control and path planning for robots with limited sensory systems, visual servoing, map building of indoor and outdoor complex environments, observability and controllability analysis, optimal sensing design, optimal estimation and active sensing control, action-perception coupling and sensors-based feedback control and planning.