Learning Safety-Constrained Whole-Body Loco-Manipulation for Lunar Surface Operations
This paper presents a Constrained Reinforcement Learning framework for quadrupedal mobile manipulators that learns a single whole-body policy for tracking a 6D end-effector pose specified in task space, with operational limits enforced during learning through a constrained learning mechanism.
ProofPaper ↗
Key points
- Robotic systems for lunar surface operations must act autonomously and preserve their own integrity in an environment where failures are unrecoverable and continuous supervision is impossible.
- The framework is evaluated in a lunar-analog simulation spanning three terrain morphologies at five difficulty levels under lunar gravity, with regolith contact properties calibrated to the JSC-1a simulant and varied from loose to dense.
- The learned policy attains mean position and orientation errors of 3.22 cm and 6.83 deg. under nominal conditions, retains this precision across regolith densities and over the operationally representative range of terrain difficulty, and reduces soft-constraint violation time by up to an order of magnitude relative to an unconstrained baseline while incurring no hard-constraint violations.
Sources (1)
- [1]Learning Safety-Constrained Whole-Body Loco-Manipulation for Lunar Surface OperationsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 11, 09:38 AM
This paper presents a Constrained Reinforcement Learning framework for quadrupedal mobile manipulators that learns a single whole-body policy for tracking a 6D end-effector pose specified in task space, with operational limits enforced during learning through a constrained learning mechanism.
Robotic systems for lunar surface operations must act autonomously and preserve their own integrity in an environment where failures are unrecoverable and continuous supervision is impossible.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 11, 2026Pessimistic Latent Task-aware Optimization for Robust Offline Meta-Reinforcement Learning
- Oct 11, 2026RSIPO: Recursive Self-Improving Policy Optimization for Agentic Reinforcement Learning via Skill-Policy Co-Evolution
- Oct 11, 2026ANTE: Anticipate, Then Execute -- Learning Internal World Models for Coding Agents from Verified Beliefs
- Oct 9, 2026The Thinking Leakage in Reinforcement Learning for LLMs
- Oct 9, 2026Rethinking Relative Policy Differences Transfer of On-Policy Distillation
- Oct 9, 2026SimVTA: Simulation-Grounded Video-Trajectory Alignment for Driving World Action Models