Real-Time Adaptive Motion Planning via Point Cloud-Guided, Energy-Based Diffusion and Potential Fields
This work addresses motion planning for robotics in dynamic, obstacle-rich settings, but it appears incremental as it builds on existing diffusion and potential field methods.
The paper tackles real-time motion planning in complex environments by combining energy-based diffusion models with potential fields, processing obstacle information directly from point clouds without full geometric representations. It demonstrates effective performance in pursuit-evasion scenarios with partial observability, though no concrete numbers are provided.
Motivated by the problem of pursuit-evasion, we present a motion planning framework that combines energy-based diffusion models with artificial potential fields for robust real time trajectory generation in complex environments. Our approach processes obstacle information directly from point clouds, enabling efficient planning without requiring complete geometric representations. The framework employs classifier-free guidance training and integrates local potential fields during sampling to enhance obstacle avoidance. In dynamic scenarios, the system generates initial trajectories using the diffusion model and continuously refines them through potential field-based adaptation, demonstrating effective performance in pursuit-evasion scenarios with partial pursuer observability.