Efficient Exploratory Learning of Inverse Kinematics on a Bionic Elephant Trunk
Open Access
- 20 November 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 25 (6), 1147-1160
- https://doi.org/10.1109/tnnls.2013.2287890
Abstract
We present an approach to learn the inverse kinematics of the “bionic handling assistant”-an elephant trunk robot. This task comprises substantial challenges including high dimensionality, restrictive and unknown actuation ranges, and nonstationary system behavior. We use a recent exploration scheme, online goal babbling, which deals with these challenges by bootstrapping and adapting the inverse kinematics on the fly. We show the success of the method in extensive real-world experiments on the nonstationary robot, including a novel combination of learning and traditional feedback control. Simulations further investigate the impact of nonstationary actuation ranges, drifting sensors, and morphological changes. The experiments provide the first substantial quantitative real-world evidence for the success of goal-directed bootstrapping schemes, moreover with the challenge of nonstationary system behavior. We thereby provide the first functioning control concept for this challenging robot platform.Keywords
This publication has 25 references indexed in Scilit:
- Learning versatile sensorimotor coordination with goal babbling and neural associative dynamicsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Constant curvature continuum kinematics as fast approximate model for the Bionic Handling AssistantPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2012
- Goal Babbling Permits Direct Learning of Inverse KinematicsIEEE Transactions on Autonomous Mental Development, 2010
- Maturationally-constrained competence-based intrinsically motivated learningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2010
- Robust intrinsically motivated exploration and active learningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Reinforcement learning by reward-weighted regression for operational space controlPublished by Association for Computing Machinery (ACM) ,2007
- A Self-Organizing Neural Model of Motor Equivalent Reaching and Tool Use by a Multijoint ArmJournal of Cognitive Neuroscience, 1993
- Forward models: Supervised learning with a distal teacherCognitive Science, 1992
- Feedback-Error-Learning Neural Network for Supervised Motor LearningPublished by Elsevier BV ,1990
- Eye–hand coordination in the newborn.Developmental Psychology, 1982