Imitating Human Search Strategies for Assembly

We present a Learning from Demonstration method for teaching robots to perform search strategies imitated from humans in scenarios where alignment tasks fail due to position uncertainty. The method utilizes human demonstrations to learn both a state invariant dynamics model and an exploration distribution that captures the search area covered by the demonstrator. We present two alternative algorithms for computing a search trajectory from the exploration distribution, one based on sampling and another based on deterministic ergodic control. We augment the search trajectory with forces learnt through the dynamics model to enable searching both in force and position domains. An impedance controller with superposed forces is used for reproducing the learnt strategy. We experimentally evaluate the method on a KUKA LWR4+ performing a 2D peg-in-hole and a 3D electricity socket task. Results show that the proposed method can, with only few human demonstrations, learn to complete the search task.

[1]  Yuchen Zhao,et al.  Teaching Human Teachers to Teach Robot Learners , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[2]  Ville Kyrki,et al.  Learning compliant assembly motions from demonstration , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[3]  I. Mezić,et al.  Metrics for ergodicity and design of ergodic dynamics for multi-agent systems , 2011 .

[4]  De Chambrier,et al.  Learning Search Strategies from Human Demonstrations , 2016 .

[5]  Brett Browning,et al.  A survey of robot learning from demonstration , 2009, Robotics Auton. Syst..

[6]  Klas Kronander,et al.  Control and Learning of Compliant Manipulation Skills , 2015 .

[7]  Nolan Wagener,et al.  Learning contact-rich manipulation skills with guided policy search , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[8]  Nicholas Roy,et al.  Efficient Planning for Near-Optimal Compliant Manipulation Leveraging Environmental Contact , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[9]  Oliver Brock,et al.  Opening a lockbox through physical exploration , 2017, 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids).

[10]  Aude Billard,et al.  On Learning, Representing, and Generalizing a Task in a Humanoid Robot , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[11]  S. Schaal.  Dynamic Movement Primitives -A Framework for Motor Control in Humans and Humanoid Robotics , 2006 .

[12]  Holger Voos,et al.  Position Identification in Force-Guided Robotic Peg-in-Hole Assembly Tasks , 2014 .

[13]  Ales Ude,et al.  Solving peg-in-hole tasks by human demonstration and exception strategies , 2014 .

[14]  Anders Robertsson,et al.  Force controlled assembly of emergency stop button , 2011, 2011 IEEE International Conference on Robotics and Automation.

[15]  J. Hauser.  A PROJECTION OPERATOR APPROACH TO THE OPTIMIZATION OF TRAJECTORY FUNCTIONALS , 2002 .

[16]  George Marsaglia,et al.  A Fast, Easily Implemented Method for Sampling from Decreasing or Symmetric Unimodal Density Functions , 1984 .

[17]  G. Schreiber,et al.  The Fast Research Interface for the KUKA Lightweight Robot , 2022 .

[18]  Todd D. Murphey,et al.  Trajectory optimization for continuous ergodic exploration , 2013, 2013 American Control Conference.

[19]  Tesfamichael Hagos.  Estimation of Phases for Compliant Motion , 2017 .

[20]  Ville Kyrki,et al.  A geometric approach for learning compliant motions from demonstration , 2017, 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids).

[21]  Jan Peters,et al.  Reinforcement learning in robotics: A survey , 2013, Int. J. Robotics Res..

[22]  Mykel J. Kochenderfer,et al.  On the Optimality of Ergodic Trajectories for Information Gathering Tasks , 2018, 2018 Annual American Control Conference (ACC).