Towards Orientation Learning and Adaptation in Cartesian Space
Yanlong Huang, Fares J. Abu-Dakka, João Silvério, and Darwin G. Caldwell
We propose an approach that is capable of learning multiple orientation trajectories and adapting learned orientation skills to new situations (e.g., via-points and end-points), where both orientation and angular velocity are considered. In addition, we extend our approach to the learning of quaternions with angular acceleration or jerk constraints as well as the learning of quaternions associated with high-dimensional inputs.
Overview
Overview
Experiments
Experiments
Painting Task
Painting Task
Handover Task
Handover Task
Interpolation Comparison
Interpolation Comparison