RAVEN: Reinforcement-Adaptive Visibility-Graph Planning for Robust Humanoid Navigation with Collision-Free MPC

2026-07-17

AI Summary

This work addresses the challenge of safe navigation for humanoid robots in dynamic environments, where control delays, state estimation noise, and motion uncertainty often lead to collisions or constraint violations—even when geometrically optimal paths are planned. To mitigate these issues, the authors propose RAVEN, a novel framework that leverages reinforcement learning to adaptively tune geometric parameters—such as obstacle inflation—in a visibility graph planner. This reshapes the free-space topology to compensate for system-level errors while preserving the interpretability and long-horizon reasoning of geometric planning. The resulting reference trajectories are then tracked using a collision-avoidance model predictive controller explicitly designed to satisfy dynamic feasibility and safety constraints. Experimental results demonstrate that RAVEN significantly reduces overshoot in narrow passages and achieves more robust and reliable navigation under sensing noise and actuation delays compared to both hand-tuned baselines and end-to-end reinforcement learning policies.

Abstract

Humanoid navigation in dynamic environments requires long-horizon planning while respecting short-horizon dynamic and safety constraints. Classical visibility-graph planners combined with model predictive control (MPC) can efficiently generate collision-free trajectories, but their performance depends on manually tuned parameters and accurate system modeling. In real robotic systems, control delays, state-estimation noise, and locomotion uncertainties can cause overshoot and constraint violations even when the nominal path is geometrically optimal. We propose RAVEN, a hierarchical reinforcement learning (RL)-MPC framework for robust humanoid navigation. Unlike prior approaches that use learning to tune cost weights or replace planning entirely, RAVEN employs RL to adapt the geometric construction of a visibility-graph planner by modifying obstacle inflation and related graph parameters. By directly reshaping the free-space geometry, the learned planner alters the topology of the global path to compensate for delay and tracking imperfections. A collision-free MPC layer then tracks the planned trajectory while explicitly enforcing velocity bounds and obstacle-avoidance constraints. By training under realistic delays and observation noise, RAVEN learns planning adaptations that improve robustness while retaining explicit long-horizon geometric planning and constrained optimization, in contrast to end-to-end learning approaches. We evaluate RAVEN against a manually tuned visibility-graph MPC baseline and a pure RL navigation policy. Results demonstrate reduced overshoot near obstacles, improved robustness in narrow passages, and more reliable navigation under delay and noise. These findings indicate that reinforcement-adaptive graph construction combined with constrained MPC provides an effective and interpretable alternative to end-to-end learning for robust humanoid navigation.

Problem

• humanoid navigation
• collision avoidance
• control delay
• state-estimation noise
• safety constraints

Innovation

• Reinforcement Learning
• Visibility Graph
• Model Predictive Control
• Humanoid Navigation
• Adaptive Planning