Real-time robot learning with locally weighted statistical learning

Publisher Website
Google Scholar
Abstract
Locally weighted learning (LWL) is a class of statistical learning techniques that provides useful repre- sentations and training algorithms for learning about com- plex phenomena during autonomous adaptive control of ro- botic systems. This paper introduces several LWL algo- rithms that have been tested successfully in real-time learn- ing of complex robot tasks. We discuss two major classes of LWL, memory-based LWL and purely incremental LWL that does not need to remember any data explicitly. In contrast to the traditional beliefs that LWL methods cannot work well in high-dimensional spaces, we provide new algorithms that have been tested in up to 50 dimensional learning problems. The applicability of our LWL algorithms is dem- onstrated in various robot learning examples, including the learning of devil-sticking, pole-balancing of a humanoid ro- bot arm, and inverse-dynamics learning for a seven degree- of-freedom robot.

This publication has 13 references indexed in Scilit: