SafeMove-RL: A Certifiable Reinforcement Learning Framework for Dynamic Motion Constraints in Trajectory Planning

2025-05-19

AI Summary

Local motion planning under dynamic uncertainty and partial observability remains challenging due to the difficulty of ensuring real-time safety, feasibility, and adaptability simultaneously. Method: This paper proposes a reinforcement learning framework grounded in Dynamic Safety Margin (DSM), integrating real-time nonlinear trajectory optimization with adaptive gap analysis to enable online DSM generation. The DSM mechanism concurrently refines trajectories and preserves control invariance in unknown environments, thereby guaranteeing verifiable feasibility and safety. Contribution/Results: Evaluated in both simulation and on real robotic platforms, the framework achieves a 23.6% improvement in task success rate and reduces computational latency by 41.2% compared to state-of-the-art methods. It is the first approach to achieve a provable triadic balance among safety, real-time performance, and environmental adaptability under partial observability—establishing a novel paradigm for safety-critical autonomous systems.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: Motion and Path Planning
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This study presents a dynamic safety margin-based reinforcement learning framework for local motion planning in dynamic and uncertain environments. The proposed planner integrates real-time trajectory optimization with adaptive gap analysis, enabling effective feasibility assessment under partial observability constraints. To address safety-critical computations in unknown scenarios, an enhanced online learning mechanism is introduced, which dynamically corrects spatial trajectories by forming dynamic safety margins while maintaining control invariance. Extensive evaluations, including ablation studies and comparisons with state-of-the-art algorithms, demonstrate superior success rates and computational efficiency. The framework's effectiveness is further validated on both simulated and physical robotic platforms.

Problem

• Ensures safe trajectory planning in dynamic uncertain environments
• Integrates real-time optimization with adaptive feasibility assessment
• Dynamically corrects trajectories while maintaining safety margins

Innovation

• Dynamic safety margin-based reinforcement learning for motion planning
• Real-time trajectory optimization with adaptive gap analysis
• Enhanced online learning for dynamic safety margin correction