SONIC-K1: Porting a Universal Humanoid Motion Tracker to a Small Humanoid on a Single GPU

摘要变化

We present SONIC-K1, to our knowledge the first open-weights whole-body motion tracker for the Booster K1 humanoid, obtained by adapting the SONIC framework to this smaller robot. The policy is trained from scratch on a single shared GPU with a small fraction of the compute reported for SONIC. In MuJoCo simulation, it tracks nearly all held-out motion clips and performs well on external benchmarks under SONIC's success criterion. OnThe first release ran on the real K1 but shook its legs on steps and turns, itand trackssimulation uppershowed a sensitivity to actuation delay. We therefore post-bodytrained andit slowwith motionsthe robot's deployment torque limits, whilerandomized steppingactuation delay and turninggain remainrandomization. lessThe reliablepost-trained v1.0 stays robust under real-robot torque limits and large actuation delays, where the first release often failed, and simulationin experimentsour showruns aon sensitivitythe toreal actionK1 latencyit walks and turns without the earlier leg shaking. These results show broad motion-tracking capability on a modest training budget and point to quantitative hardware evaluation as the main next step. Code, policy weights and evaluation tools are publicly available.