Print Email Facebook Twitter Action learning from human demonstrations for personal robots Title Action learning from human demonstrations for personal robots Author Rozemuller, C.G. Contributor Jonker, P.P. (mentor) Rudinac, M. (mentor) Faculty Mechanical, Maritime and Materials Engineering Department BioMechanical Engineering Programme BMD Date 2013-12-10 Abstract Household robots need to perform tasks specific for the owner. With Learning from demonstration (LfD) a robot can learn new tasks from human demonstrations, without requiring programming skills. This thesis investigates a novel representation of actions that can be learned by using only a 3d camera and an object tracker. The action representation is objectbased so it is independent of the morphology of the robot. The actions are represented using the average and standard deviation of multiple demonstrated trajectories with six degrees of freedom. The standard deviation serves as a weight factor for the required accuracy of the recognized or synthesized trajectory. Three novel methods proposed in this thesis aim to reduce variances in the demonstration that are not specific to the action. First the demonstrations are aligned in time using a novel action signature and a novel time warp algorithm. The time warp algorithm can approximate the alignment of multiple multidimensional signals in quadratic computing time. The third novel technique is a dynamically optimized choice of reference frame so variations in start and end position have little influence on the variance in trajectory. This method has been tested on a database of five actions repeatedly demonstrated by six subjects. The results show that it is possible to have a 90 percent action recognition rate with only three demonstrations in the database. It is also shown that a robot can use this action representation to synthesize four out of five actions with varying object positions. Subject learning from demonstrations To reference this document use: http://resolver.tudelft.nl/uuid:4dbcecf9-e143-4f16-b315-f4a509140b7d Embargo date 2014-02-12 Part of collection Student theses Document type master thesis Rights (c) 2013 Rozemuller, C.G. Files PDF Thesis_Chris_Rozemuller.pdf 6.27 MB Close viewer /islandora/object/uuid:4dbcecf9-e143-4f16-b315-f4a509140b7d/datastream/OBJ/view