← 研踪 PaperTrailarXiv:2610.39327 · v1
SONIC-K1: Porting a Universal Humanoid Motion Tracker to a Small Humanoid on a Single GPU
首次提交 2026-10-29 12:16
版本更新 2026-10-29 12:16
雷达观测 2026-10-30 11:25
摘要
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. On the real K1, it tracks upper-body and slow motions, while stepping and turning remain less reliable, and simulation experiments show a sensitivity to action latency. 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.
作者备注：Project page: https://kyry.me/sonic-k1
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-04 01:41 · 最新观测版本
SONIC-K1: Porting a Universal Humanoid Motion Tracker to a Small Humanoid on a Single GPU
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. On the real K1, it tracks upper-body and slow motions, while stepping and turning remain less reliable, and simulation experiments show a sensitivity to action latency. 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.
查看此版本解读 →查看此版本原文 ↗v1 · 2026-10-29 12:16
SONIC-K1: Porting a Universal Humanoid Motion Tracker to a Small Humanoid on a Single GPU
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. On the real K1, it tracks upper-body and slow motions, while stepping and turning remain less reliable, and simulation experiments show a sensitivity to action latency. 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.
查看此版本解读 →查看此版本原文 ↗