TAMPAgent at CoRL: coding agents + TAMP planning for zero-shot long-horizon mobile manipulation
dhadfieldmenell · x · 2026-11-05
Researchers will present TAMPAgent at CoRL 2026, an agentic planning framework that equips a coding agent with a belief-space task and motion planning (TAMP) model to tackle long-horizon mobile manipulation under partial observability—targets inside fridges or blocked by objects—zero-shot and without simulation rehearsal.
- Motivation: coding agents calling robot tools mostly work only on fully observable tabletops
- Method: combines LLM flexibility and common-sense reasoning with robot planners' time and token efficiency
- Validation: evaluated in simulation and on a real RB-Y1 mobile manipulator and a Unitree G1 humanoid
- Featured in CoRL 2026 Late-Breaking Results and the LEAP workshop
More from Embodied
- Robotics sim frontier names from World Labs, NVIDIA, PI to meet in SF on sim2real gap — jia_seed · 2026-11-05
- Robotics RL still on Isaac Lab PhysX despite mjwarp's 2x+ training speedup — carlosdponx · 2026-11-05
- Widespread broken code and training data in academic robotics projects — carlosdponx · 2026-11-05
- General Intuition: game clips (1.5T action tokens/month) train one model to control 10+ robots — AI Engineer · 2026-11-05
- CoRL paper GUILD: actively guiding robot data collection toward useful auxiliary tasks — RobobertoMM · 2026-11-05
- Raven Prism launches: a full Linux computer in glasses, controlled by eyes and voice — Scobleizer · 2026-11-05