Operator training for preferred manipulator trajectories in a glovebox. Sharp, A., Horn, M. W., & Pryor, M. In 2017 IEEE Workshop on Advanced Robotics and its Social Impacts (ARSO), pages 1–6, March, 2017. ISSN: 2162-7576doi abstract bibtex 8 downloads We consider the problem of operator comfort levels with manipulator trajectories in heavily constrained, co-robotic environments. To reduce anxiety and improve user comfort, a trajectory learning system is trained using a set of specified environmental features associated with a waste sorting task. A Trajectory Preference Perceptron was used to learn desired operator feature weights via iterative suboptimal but improved trainer feedback thus relieving the operator of the burden of providing optimal feedback. The object feature weights are stored and updated to handle a fluid environmental scene. A pool of robotics students spanning a range of experience from novice undergraduates to doctoral candidates provided qualitative results to evaluate the method's effectiveness.
@inproceedings{sharp_operator_2017,
title = {Operator training for preferred manipulator trajectories in a glovebox},
doi = {10.1109/ARSO.2017.8025193},
abstract = {We consider the problem of operator comfort levels with manipulator trajectories in heavily constrained, co-robotic environments. To reduce anxiety and improve user comfort, a trajectory learning system is trained using a set of specified environmental features associated with a waste sorting task. A Trajectory Preference Perceptron was used to learn desired operator feature weights via iterative suboptimal but improved trainer feedback thus relieving the operator of the burden of providing optimal feedback. The object feature weights are stored and updated to handle a fluid environmental scene. A pool of robotics students spanning a range of experience from novice undergraduates to doctoral candidates provided qualitative results to evaluate the method's effectiveness.},
booktitle = {2017 {IEEE} {Workshop} on {Advanced} {Robotics} and its {Social} {Impacts} ({ARSO})},
author = {Sharp, Andrew and Horn, Matthew W. and Pryor, Mitch},
month = mar,
year = {2017},
note = {ISSN: 2162-7576},
keywords = {Collision avoidance, Grippers, Hardware, Manipulators, Training, Trajectory, anxiety reduction, control engineering education, data gloves, environmental features, feedback, fluid environmental scene, glovebox, heavily constrained corobotic environment, iterative learning control, iterative suboptimal, learning systems, manipulator trajectories, manipulators, operator comfort level, operator feature weights, operator training, optimal feedback, robotics students, suboptimal control, telerobotics, trainer feedback, training, trajectory control, trajectory learning system training, trajectory preference perceptron, user comfort, waste sorting task},
pages = {1--6},
}
Downloads: 8
{"_id":"Y9A2RaTuMNfYKkQGd","bibbaseid":"sharp-horn-pryor-operatortrainingforpreferredmanipulatortrajectoriesinaglovebox-2017","authorIDs":[],"author_short":["Sharp, A.","Horn, M. W.","Pryor, M."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Operator training for preferred manipulator trajectories in a glovebox","doi":"10.1109/ARSO.2017.8025193","abstract":"We consider the problem of operator comfort levels with manipulator trajectories in heavily constrained, co-robotic environments. To reduce anxiety and improve user comfort, a trajectory learning system is trained using a set of specified environmental features associated with a waste sorting task. A Trajectory Preference Perceptron was used to learn desired operator feature weights via iterative suboptimal but improved trainer feedback thus relieving the operator of the burden of providing optimal feedback. The object feature weights are stored and updated to handle a fluid environmental scene. A pool of robotics students spanning a range of experience from novice undergraduates to doctoral candidates provided qualitative results to evaluate the method's effectiveness.","booktitle":"2017 IEEE Workshop on Advanced Robotics and its Social Impacts (ARSO)","author":[{"propositions":[],"lastnames":["Sharp"],"firstnames":["Andrew"],"suffixes":[]},{"propositions":[],"lastnames":["Horn"],"firstnames":["Matthew","W."],"suffixes":[]},{"propositions":[],"lastnames":["Pryor"],"firstnames":["Mitch"],"suffixes":[]}],"month":"March","year":"2017","note":"ISSN: 2162-7576","keywords":"Collision avoidance, Grippers, Hardware, Manipulators, Training, Trajectory, anxiety reduction, control engineering education, data gloves, environmental features, feedback, fluid environmental scene, glovebox, heavily constrained corobotic environment, iterative learning control, iterative suboptimal, learning systems, manipulator trajectories, manipulators, operator comfort level, operator feature weights, operator training, optimal feedback, robotics students, suboptimal control, telerobotics, trainer feedback, training, trajectory control, trajectory learning system training, trajectory preference perceptron, user comfort, waste sorting task","pages":"1–6","bibtex":"@inproceedings{sharp_operator_2017,\n\ttitle = {Operator training for preferred manipulator trajectories in a glovebox},\n\tdoi = {10.1109/ARSO.2017.8025193},\n\tabstract = {We consider the problem of operator comfort levels with manipulator trajectories in heavily constrained, co-robotic environments. To reduce anxiety and improve user comfort, a trajectory learning system is trained using a set of specified environmental features associated with a waste sorting task. A Trajectory Preference Perceptron was used to learn desired operator feature weights via iterative suboptimal but improved trainer feedback thus relieving the operator of the burden of providing optimal feedback. The object feature weights are stored and updated to handle a fluid environmental scene. A pool of robotics students spanning a range of experience from novice undergraduates to doctoral candidates provided qualitative results to evaluate the method's effectiveness.},\n\tbooktitle = {2017 {IEEE} {Workshop} on {Advanced} {Robotics} and its {Social} {Impacts} ({ARSO})},\n\tauthor = {Sharp, Andrew and Horn, Matthew W. and Pryor, Mitch},\n\tmonth = mar,\n\tyear = {2017},\n\tnote = {ISSN: 2162-7576},\n\tkeywords = {Collision avoidance, Grippers, Hardware, Manipulators, Training, Trajectory, anxiety reduction, control engineering education, data gloves, environmental features, feedback, fluid environmental scene, glovebox, heavily constrained corobotic environment, iterative learning control, iterative suboptimal, learning systems, manipulator trajectories, manipulators, operator comfort level, operator feature weights, operator training, optimal feedback, robotics students, suboptimal control, telerobotics, trainer feedback, training, trajectory control, trajectory learning system training, trajectory preference perceptron, user comfort, waste sorting task},\n\tpages = {1--6},\n}\n\n","author_short":["Sharp, A.","Horn, M. W.","Pryor, M."],"key":"sharp_operator_2017","id":"sharp_operator_2017","bibbaseid":"sharp-horn-pryor-operatortrainingforpreferredmanipulatortrajectoriesinaglovebox-2017","role":"author","urls":{},"keyword":["Collision avoidance","Grippers","Hardware","Manipulators","Training","Trajectory","anxiety reduction","control engineering education","data gloves","environmental features","feedback","fluid environmental scene","glovebox","heavily constrained corobotic environment","iterative learning control","iterative suboptimal","learning systems","manipulator trajectories","manipulators","operator comfort level","operator feature weights","operator training","optimal feedback","robotics students","suboptimal control","telerobotics","trainer feedback","training","trajectory control","trajectory learning system training","trajectory preference perceptron","user comfort","waste sorting task"],"metadata":{"authorlinks":{}},"downloads":8,"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero-group/mpryor/246442","creationDate":"2020-06-09T10:47:20.999Z","downloads":8,"keywords":["collision avoidance","grippers","hardware","manipulators","training","trajectory","anxiety reduction","control engineering education","data gloves","environmental features","feedback","fluid environmental scene","glovebox","heavily constrained corobotic environment","iterative learning control","iterative suboptimal","learning systems","manipulator trajectories","manipulators","operator comfort level","operator feature weights","operator training","optimal feedback","robotics students","suboptimal control","telerobotics","trainer feedback","training","trajectory control","trajectory learning system training","trajectory preference perceptron","user comfort","waste sorting task"],"search_terms":["operator","training","preferred","manipulator","trajectories","glovebox","sharp","horn","pryor"],"title":"Operator training for preferred manipulator trajectories in a glovebox","year":2017,"dataSources":["E5HAvRxzabR7xmf5S","NnPAtJvwQQamPe64H"]}