Robust Quadrupedal Locomotion

In this work, with a view toward fast deployment of locomotion gaits in low-cost hardware, we use a linear policy for realizing end-foot trajectories in the quadruped robot, Stoch2. In particular, the parameters of the end-foot trajectories are shaped via a linear feedback policy that takes the torso orientation and the terrain slope as inputs. The corresponding desired joint angles are obtained via an inverse kinematics solver, and tracked via a PID control law. Augmented Random Search, a model-free and a gradient-free learning algorithm, is used to train this linear policy. Simulation results show that the resulting walking is robust to external pushes and terrain slope variations. This methodology is not only computationally light-weight, but also uses minimal sensing and actuation capabilities in the robot, thereby justifying the approach.

This work is in collaboration with Dr. Bharadwaj Amrutur, Dr. Shalabh Bhatnagar and Dr. Ashitava Ghosal from IISc.

Reference:

Kartik Paigwar, Lokesh Krishna, Sashank Tirumala, Naman Khetan, Aditya Sagi, Ashish Joglekar, Shalabh Bhatnagar, Ashitava Ghosal, Bharadwaj Amrutur, and Shishir Kolathaya, “Robust Quadrupedal Locomotion on Sloped Terrains: A Linear Policy Approach” 4th Conference on Robot Learning (CoRL 2020), Cambridge MA, USA.

https://stochlab.github.io/img/CORL.gif