BiGym: A Demo-Driven Mobile Bi-Manual Manipulation Benchmark

Nikita Chernyadev, Nicholas Backshall, Xiao Ma, Yunfan Lu, Younggyo Seo, Stephen James

Keywords: Bi-Manual Manipulation, Mobile Manipulation, Benchmark

TL;DR: A Demo-Driven Mobile Bi-Manual Manipulation Benchmark & Learning Environment, for Reinforcement and Imitation Learning

Abstract: We introduce BiGym, a new benchmark and learning environment for mobile bi-manual demo-driven robotic manipulation. BiGym features 40 diverse tasks set in home environments, ranging from simple target reaching to complex kitchen cleaning. To capture the real-world performance accurately, we provide human-collected demonstrations for each task, reflecting the diverse modalities found in real-world robot trajectories. BiGym supports a variety of observations, including proprioceptive data and visual inputs such as RGB, and depth from 3 camera views. To validate the usability of BiGym, we thoroughly benchmark the state-of-the-art imitation learning algorithms and demo-driven reinforcement learning algorithms within the environment and discuss the future opportunities.

Video: https://youtu.be/jFhhPkFngwk?si=cjocyZ72Xtnb60l_

Website: https://chernyadev.github.io/bigym/

Code: https://github.com/chernyadev/bigym

Student Paper: no

Submission Number: 332