Learning Controllers for Robust Bipedal Locomotion on Challenging Terrains

Learning Controllers for Robust Bipedal Locomotion on Challenging Terrains


Abstract

In this work, we demonstrate robust walking in the bipedal robot Digit on uneven terrains by just learning a single linear policy. In particular, we propose a new control pipeline, wherein the high-level trajectory modulator shapes the end-foot ellipsoidal trajectories, and the low-level gait controller regulates the torso and ankle orientation.The foot-trajectory modulator uses a linear policy and the regulator uses a linear PD control law. As opposed to neural network based policies, the proposed linear policy has only 13 learnable parameters, thereby not only guaranteeing sample efficient learning but also enabling simplicity and interpretability of the policy. This is achieved with no loss of performance on challenging terrains like slopes, stairs and outdoor landscapes.We first demonstrate robust walking in the custom simulation environment, MuJoCo, and then directly transfer to hardware with no modification of the control pipeline. We subject the biped to a series of pushes and terrain height changes, both indoors and outdoors, thereby validating the presented work.

incline
Slope: Incline
decline
Slope: Decline
vslope
Variying Slope
sine
Sinusoidal Terrain
spike
Undulated-Spiky Terrain
steps
Undulated-Step Terrain
stairs
Staircase
omni
Omni-Driectional Walking
omni
Unstructured Outdoors
omni
Push Recovery
omni
Hardware - Stairs
omni
Hardware - Command Control


Control Framework

drawing


Video




Citations

    @misc{krishna2021learning,
          title={Learning Linear Policies for Robust Bipedal Locomotion on Terrains with Varying Slopes}, 
          author={Lokesh Krishna and Utkarsh A. Mishra and Guillermo A. Castillo and Ayonga Hereid and Shishir Kolathaya},
          year={2021},
          eprint={2104.01662},
          archivePrefix={arXiv},
          primaryClass={cs.RO}
        }
    @misc{krishna2021linear,
          title={Linear Policies are Sufficient to Realize Robust Bipedal Walking on Challenging Terrains}, 
          author={Lokesh Krishna and Guillermo A. Castillo and Utkarsh A. Mishra and Ayonga Hereid and Shishir Kolathaya},
          year={2021},
          eprint={2109.12665},
          archivePrefix={arXiv},
          primaryClass={cs.RO}
        }



Related Papers

2021
September
PDF Learning Linear Policies for Robust Bipedal Locomotionon Terrains with Varying Slopes
Lokesh Krishna, Utkarsh A. Mishra, Guillermo A. Castillo, Ayonga Hereid, and Shishir Kolathaya
2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021), Prague, Czech Republic.
2021
October
PDF Linear Policies are Sufficient to Realize Robust Bipedal Walking on Challenging Terrains
Lokesh Krishna*, Guillermo A. Castillo*, Utkarsh A. Mishra, Ayonga Hereid, and Shishir Kolathaya
arXiv

People

Lokesh Krishna
B.Tech., IIT BHU
Utkarsh Mishra U
Utkarsh Mishra
IIT Roorkee
Shishir Kolathaya
Assistant Professor
Last updated: 2021-08-07