Dynamic Bipedal Turning through Sim-to-Real Reinforcement Learning
Authors:
- 28 November 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 903-910
Abstract
For legged robots to match the athletic capabilities of humans and animals, they must not only produce robust periodic walking and running, but also seamlessly switch between nominal locomotion gaits and more specialized transient maneuvers. Despite recent advancements in controls of bipedal robots, there has been little focus on producing highly dynamic behaviors. Recent work utilizing reinforcement learning to produce policies for control of legged robots have demonstrated success in producing robust walking behaviors. However, these learned policies have difficulty expressing a multitude of different behaviors on a single network. Inspired by conventional optimization-based control techniques for legged robots, this work applies a recurrent policy to execute four-step, 90° turns trained using reference data generated from optimized single rigid body model trajectories. We present a training framework using epilogue terminal rewards for learning specific behaviors from pre-computed trajectory data and demonstrate a successful transfer to hardware on the bipedal robot Cassie.Keywords
All Related Versions
Funding Information
- NSF (IIS-1849343,1314109-DGE)
- DARPA (-16-1-0002)
This publication has 25 references indexed in Scilit:
- Mini Cheetah: A Platform for Pushing the Limits of Dynamic Quadruped Control2019
- Optimized Jumping on the MIT Cheetah 3 Robot2019
- Learning agile and dynamic motor skills for legged robotsScience Robotics, 2019
- Physics-based motion capture imitation with deep reinforcement learning2018
- Bipedal Hopping: Reduced-Order Model Embedding via Optimization-Based Control2018
- Fast Online Trajectory Optimization for the Bipedal Robot Cassie2018
- Design of Dynamic Legged RobotsFoundations and Trends in Robotics, 2017
- Optimization-based locomotion planning, estimation, and control design for the atlas humanoid robotAutonomous Robots, 2015
- Whole-body motion planning with centroidal dynamics and full kinematics2014
- A direct method for trajectory optimization of rigid bodies through contactThe International Journal of Robotics Research, 2013