https://sourceforge.net/p/throughput/wiki/AdaptiveCollisionManipulator/Recent changes to AdaptiveCollisionManipulatorenWed, 10 Dec 2014 20:07:24 -0000https://sourceforge.net/p/throughput/wiki/AdaptiveCollisionManipulator/<div class="markdown_content"><pre>--- v47 +++ v48 @@ -35,9 +35,9 @@ - - - - - - ####Provider#### * **Name:** Martin F. Stoelen -* **Institution:** RoboticsLab, Universidad Carlos III de Madrid (UC3M), Spain -* **Contact:** mstoelen at ing.uc3m.es -* **Website:** &lt;http: roboticslab.uc3m.es=""&gt; +* **Institution:** Centre for Robotics and Neural Systems, University of Plymouth, UK +* **Contact:** martin dot stoelen at plymouth dot ac dot uk +* **Website:** &lt;http: www1.plymouth.ac.uk="" research="" crns=""&gt; - - - - - - [Home] [ExperimentalPapers] </pre> </div>Martin F. StoelenWed, 10 Dec 2014 20:07:24 -0000https://sourceforge.net2b2f898308895ea0fadf77b2f50e418861b7f866https://sourceforge.net/p/throughput/wiki/AdaptiveCollisionManipulator/<div class="markdown_content"><pre>--- v46 +++ v47 @@ -5,13 +5,20 @@ ####1. Description#### #####*Abstract*##### -TODO +An approach for adaptive shared control of an assistive manipulator is presented. A set of distributed collision and proximity sensors is used to aid in limiting collisions during direct control by the disabled user. Artificial neural networks adapt the use of the proximity sensors online, which limits movements in the direction of an obstacle before a collision occurs. The system learns by associating the different proximity sensors to the collision sensors where collisions are detected. This enables the user and the robot to adapt simultaneously and in real-time, with the objective of converging on a usage of the proximity sensors that increases performance for a given user, robot implementation and task-set. The system was tested in a controlled setting with a simulated 5 DOF assistive manipulator and showed promising reductions in the mean time on simplified manipulation tasks. It extends earlier work by showing that the approach can be applied to full multi-link manipulators. #####*Link*##### -This is so far unpublished work. Will update link when available... +* [Article link](http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6696494) #####*Citation*##### -This is so far unpublished work. Will update citation when available... +@Inproceedings{Stoelen2013, +author={Stoelen, M.F. and Tejada, V.F. and Victores, J.G. and Jardoan Huete, A. and Bonsignorio, F. and Balaguer, C.}, +booktitle={Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ International Conference on}, +title={Adaptive collision-limitation behavior for an assistive manipulator}, +year={2013}, +pages={1143-1148}, +doi={10.1109/IROS.2013.6696494}, +ISSN={2153-0858}} - - - - - - ####2. Data sets#### </pre> </div>Martin F. StoelenWed, 10 Dec 2014 20:03:56 -0000https://sourceforge.net23eaa01fd091ea44155911ef03c04f0e198ed33ehttps://sourceforge.net/p/throughput/wiki/AdaptiveCollisionManipulator/<div class="markdown_content"><pre>--- v45 +++ v46 @@ -5,7 +5,7 @@ ####1. Description#### #####*Abstract*##### -Assistive robot manipulators aim to allow disabled users to perform physical Daily Life Activities (DLA) safely and effectively in environments with both unknown and dynamic elements. Here the approach taken is to aid the user by means of an adaptive shared control, where collision and proximity sensors are used to aid in limiting collisions during teleoperation. An artificial neural network adapts the use of the proximity sensors online, which limits movements in the direction of an obstacle before a collision occurs. The system learns by associating the relevant proximity sensors to the relevant collision sensors every time a collision is detected, by implementing a Hebbian learning. This process enables the user and the robot to adapt simultaneously (online), with the objective of converging on a suitable usage of the proximity sensors for a given user, robot implementation and task-set. A set of infrared proximity sensors and a sensitive skin for a coarse collision detection were distributed over a virtual model of the ASIBOT robot. The system was tested in a controlled virtual setting and it extends earlier work by showing the concept can also be applied to full multi-link manipulators, by taking into account the known kinematic structure of the robot. +TODO #####*Link*##### This is so far unpublished work. Will update link when available... </pre> </div>Martin F. StoelenMon, 29 Apr 2013 18:36:26 -0000https://sourceforge.netbc4e7401a306ba663da50792898d402ef76e1ed9https://sourceforge.net/p/throughput/wiki/AdaptiveCollisionManipulator/<div class="markdown_content"><pre></pre> </div>Martin F. StoelenMon, 29 Apr 2013 18:33:56 -0000https://sourceforge.netf99291002ec1799ee920b2e119ce1705f1ea390d