Humanoid robot drives and drifts a go-kart via Direct Perception Control
机器之心 · wechat · 2026-08-17
Symbiosis Robotics demoed a humanoid that climbs into a cramped go-kart, steers through corners while coordinating throttle and brake, and even drifts at the finish. The core is Direct Perception Control (DPC): it removes the intermediate motion representation and separate Whole-Body Tracker, instead computing joint and hand targets directly from vision, task language and body feedback, with a two-stage DriftDistill training (offline BC init + online rollout correction) to handle real-world drift.
The team also curated 15,010 hours of embodied data (6,781h human egocentric, 4,024h armed robots, 3,660h wheeled humanoids, 545h bipedal), all converted into unified G1-executable joint actions. The write-up contrasts DPC with layered systems like Figure Helix, Gemini Robotics 2, ω-0 and Ψ₀, and shows additional demos of mobile pick-and-place and whole-body manipulation in confined spaces.
More from Embodied
- Think tank warns of 'China shock' in robotics as physical AI race heats up — pstAsiatech · 2026-08-17
- Neuralink bypasses spinal cord injuries to restore movement — r0ck3t23 · 2026-08-17
- FireSat satellites use AI to detect wildfires every 20 minutes globally — nordicinst · 2026-08-17
- Robot demo showcases super agility with room for improvement — i_bioloid · 2026-08-17
- $9B poured into World Models, narrative outpaces reality — 快鲤鱼 · 2026-08-17
- Humanoid robot future: tens of thousands to Mars, hundreds of millions to homes, billions to workplaces — davidpattersonx · 2026-08-17