Legged robots can learn expressive whole-body skills from the motions of humans and animals. Due to the morphology gap between the source and the robot, however, the motion must be tailored to the dynamic properties of the robot. In particular, dynamic motions such as a jump require careful adjustment, since their timi...
Jaeryeong Kim, Taerim Yoon, Jin Cheng et al.· 0 citations
This work proposes a hierarchical framework that uses motion keyframes as an intermediate representation between Vision-Language Model (VLM) planning and Reinforcement Learning (RL) control and introduces a saliency-based keyframe sampling strategy for low-level policy training that improves end-to-end task success rat...
Si-Tong Chen, Fatemeh Zargarbashi, Jin Cheng et al.· 1 citation
Bundled Contact Gradients is proposed, a contact-local randomized smoothing framework for differentiable policy learning that evaluates a local bundle of randomized perturbation rollouts around the stiff contact configuration and aggregates their gradient signal thereby reducing gradient variance.
Dyuman Aditya, Jin Cheng, Clemens Schwarke et al.· arXiv.org· 1 citation
Time-Adaptive Robotic Control (TARC), a reinforcement learning framework in which the policy jointly predicts a control action and its duration of application, is introduced, and matches the performance of high-frequency discrete-time controllers while operating at less than half their control frequency.
Arnav Sukhija, Lenart Treven, Jin Cheng et al.· arXiv.org· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.