Abstract
Reliable bipedal walking over complex terrain is a challenging problem, using a curriculum can help learning. Curriculum learning is the idea of starting with an achievable version of a task and increasing the difficulty as a success criteria is met. We propose a 3-stage curriculum to train Deep Reinforcement Learning policies for bipedal walking over various challenging terrains. In the first stage, while applying forces from a target policy to the robot joints and base, the agent starts on an easy terrain and the terrain difficulty is gradually increased. In the second stage, the guiding forces are gradually reduced to zero. Finally, in the third stage, random perturbations with increasing magnitude are applied to the robot base, so the robustness of the policies are improved. In simulation experiments, we show that our approach is effective in learning separate walking policies for five terrain types: flat, hurdles, gaps, stairs, and steps. Moreover, we demonstrate that in the absence of human demonstrations, a simple hand designed walking trajectory is a sufficient prior to learn to traverse complex terrain types. In ablation studies, we show that taking out any one of the three stages of the curriculum degrades the learning performance.
Original language | English |
---|---|
Title of host publication | ACRA 2020 Proceedings |
Editors | Paulo Borges |
Place of Publication | Christchurch NZ |
Publisher | Australian Robotics and Automation Association (ARAA) |
Number of pages | 10 |
Publication status | Published - 2020 |
Event | Australasian Conference on Robotics and Automation 2020 - CSIRO’s Queensland Centre for Advanced Technologies (QCAT), Brisbane, Australia Duration: 8 Dec 2020 → 10 Dec 2020 https://www.araa.asn.au/conference/acra-2020/ https://www.dropbox.com/s/2wq817ubujy1u7e/ACRA2020Proceedings.zip https://research.csiro.au/robotics/acra2020/ |
Publication series
Name | Australasian Conference on Robotics and Automation, ACRA |
---|---|
Publisher | Australian Robotics and Automation Association (ARAA) |
Volume | 2020-December |
ISSN (Print) | 1448-2053 |
Conference
Conference | Australasian Conference on Robotics and Automation 2020 |
---|---|
Abbreviated title | ACRA 2020 |
Country/Territory | Australia |
City | Brisbane |
Period | 8/12/20 → 10/12/20 |
Internet address |