CAJun: Continuous Adaptive Jumping using a Learned Centroidal Controller

1University of Washington, 2Google Deepmind, 3Carnegie Mellon University
To be presented at CoRL 2023

CAJun achieves continuous, long-distance jumping for quadrupedal robots, and can jump over a 60cm-wide yoga mat with additional clearance.

Abstract

We present CAJun, a novel hierarchical learning and control framework that enables legged robots to jump continuously with adaptive jumping distances.

CAJun consists of a high-level centroidal policy and a low-level leg controller. In particular, we use reinforcement learning (RL) to train the centroidal policy, which specifies the gait timing, base velocity, and swing foot position for the leg controller. The leg controller optimizes motor commands for the swing and stance legs according to the gait timing to track the swing foot target and base velocity commands using optimal control. Additionally, we reformulate the stance leg optimizer in the leg controller to speed up policy training by an order of magnitude. Our system combines the versatility of learning with the robustness of optimal control.

By combining RL with optimal control methods, our system achieves the versatility of learning while enjoys the robustness from control methods, making it easily transferable to real robots. We show that after 20 minutes of training on a single GPU, CAJun can achieve continuous, long jumps with adaptive distances on a Go1 robot with small sim-to-real gaps. Moreover, the robot can jump across gaps with a maximum width of 70cm, which is over 40% wider than existing methods.

Video Overview

Long Distance Jumps

We implement two dynamic jumping gaits using CAJun. In Pronking, all legs jump simultaneously. In Bounding, the front and rear legs jump in alternating order.

Pronking

CoM displacement: 80cm per step.

Bounding

CoM displacement: 70cm per step.

Jumping with Adaptive Distance

In CAJun, the robot can adapt its jumping distance based on user command. For example, when the distance command shifts between 0.3m and 1m, the robot can alternate between long and short jumps.

LED Trajectories

Trajectories of the base, front leg and rear leg.

Pronking

Bounding

Jumping under Perturbation

The hierarchical design of CAJun makes it robust under external perturbations. For example, the robot is able to recover from an agressive backward leash-pull and resume jumping.

Pronking under Leash

Bounding under Leash

The robot suffers from foot slipping when entering grass from asphalt, but is able to resume normal jumping after a few steps.

Pronking on Grass

Bounding on Grass

Comparison with End-to-End RL

We compare CAJun with end-to-end RL methods trained under a similar setup, and find that end-to-end method cannot jump as far and suffers from a larger sim-to-real gap.

Pronking

CAJun: 4.76m, E2E: 1.67m

Bounding

CAJun: 4.34m, E2E: 3.47m

BibTeX

@article{yang2023cajun,
  title={CAJun: Continuous Adaptive Jumping using a Learned Centroidal Controller},
  author={Yang, Yuxiang and Shi, Guanya and Meng, Xiangyun and Yu, Wenhao and Zhang, Tingnan and Tan, Jie and Boots, Byron},
  journal={arXiv preprint arXiv:2306.09557},
  year={2023}
}