C-Free-Uniform: A Map-Conditioned Trajectory Sampler for Model Predictive Path Integral Control

2025-10-19

AI Summary

To address low navigation success rates caused by non-uniform trajectory sampling and poor environmental adaptability in complex polygonal environments, this paper proposes C-Free-Uniform (CFU), a novel trajectory sampling method. CFU formally defines uniformity over the free configuration space and introduces a map-conditioned control input distribution $ p(u mid x) $, enabling explicit geometric awareness of local obstacles during sampling. Integrated into the Model Predictive Path Integral (MPPI) framework, CFU yields CFU-MPPI—a new controller that synergistically combines stochastic optimization with geometry-aware sampling. Experiments in dense polygonal environments demonstrate that CFU-MPPI achieves state-of-the-art navigation success rates using significantly fewer trajectory samples than environment-agnostic baselines. This confirms its superior efficiency, robustness, and generalization capability across challenging geometric settings.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Trajectory sampling is a key component of sampling-based control mechanisms. Trajectory samplers rely on control input samplers, which generate control inputs u from a distribution p(u | x) where x is the current state. We introduce the notion of Free Configuration Space Uniformity (C-Free-Uniform for short) which has two key features: (i) it generates a control input distribution so as to uniformly sample the free configuration space, and (ii) in contrast to previously introduced trajectory sampling mechanisms where the distribution p(u | x) is independent of the environment, C-Free-Uniform is explicitly conditioned on the current local map. Next, we integrate this sampler into a new Model Predictive Path Integral (MPPI) Controller, CFU-MPPI. Experiments show that CFU-MPPI outperforms existing methods in terms of success rate in challenging navigation tasks in cluttered polygonal environments while requiring a much smaller sampling budget.

Problem

• Generates control inputs uniformly sampling free configuration space
• Conditions trajectory sampling on current local map information
• Improves navigation success rates in cluttered environments efficiently

Innovation

• Map-conditioned trajectory sampler for uniform free space coverage
• Integrates C-Free-Uniform sampling with Model Predictive Path Integral control
• Generates environment-aware control inputs using local map information