CoCoNav: Conformal Control for Safe Robot Navigation in Crowds

2026-08-07

AI Summary

This work addresses the challenges of unsafe navigation and infeasible planning in dense crowds caused by uncertainty in prediction errors. The authors propose CoCoNav, a framework that integrates online conformal calibration with runtime-certified planning through a “relax-and-verify” mechanism to dynamically adjust trajectory error bounds. It first generates a nominal trajectory using soft-constrained model predictive control (MPC) and then independently verifies the safety of both the trajectory and its contingency maneuvers. A key innovation is the introduction of a time-domain-specific conformal PI controller that adaptively regulates time-varying prediction errors while guaranteeing long-term empirical coverage. Simulations and real-world experiments on a quadrupedal robot demonstrate that CoCoNav significantly outperforms existing baselines in collision avoidance safety, task success rate, and navigation efficiency.

Abstract

Safe and efficient robot navigation in crowds requires anticipating pedestrian motion despite uncertain and potentially shifting prediction errors. Existing reactive methods can produce oscillatory behavior, while predictive planners often treat forecasts as exact or rely on restrictive error models. Incorporating conservative uncertainty sets as hard constraints can also render model predictive control (MPC) infeasible. We propose \textit{CoCoNav}, a crowd-navigation framework that combines online conformal calibration with runtime-certified planning. A horizon-specific conformal proportional--integral controller adapts trajectory-error bounds to regulate long-run empirical coverage, enabling the framework to respond to changing prediction errors. A \textit{relax-then-verify} planner preserves solver feasibility by generating nominal trajectories with soft-constrained MPC and separately certifying them, together with contingency maneuvers, against the calibrated bounds before execution. Simulations and quadruped experiments show that CoCoNav achieves a favorable balance among collision avoidance, task success, and navigation efficiency relative to the evaluated baselines.

Problem

• robot navigation
• crowd navigation
• prediction uncertainty
• safety
• motion planning

Innovation

• conformal prediction
• model predictive control
• safe navigation
• uncertainty calibration
• crowd robotics