Stein Variational Probabilistic Roadmaps
This addresses the challenge of reliable motion planning for autonomous systems in partially observable environments, representing an incremental advancement by integrating probabilistic methods into existing roadmap techniques.
The paper tackles the problem of generating global path plans for autonomous systems in uncertain environments by proposing Stein Variational Probabilistic Roadmaps (SV-PRM), which uses particle-based variational inference to efficiently cover feasible regions in configuration space, resulting in sample-efficient planning graphs and large improvements over traditional sampling approaches.
Efficient and reliable generation of global path plans are necessary for safe execution and deployment of autonomous systems. In order to generate planning graphs which adequately resolve the topology of a given environment, many sampling-based motion planners resort to coarse, heuristically-driven strategies which often fail to generalize to new and varied surroundings. Further, many of these approaches are not designed to contend with partial-observability. We posit that such uncertainty in environment geometry can, in fact, help drive the sampling process in generating feasible, and probabilistically-safe planning graphs. We propose a method for Probabilistic Roadmaps which relies on particle-based Variational Inference to efficiently cover the posterior distribution over feasible regions in configuration space. Our approach, Stein Variational Probabilistic Roadmap (SV-PRM), results in sample-efficient generation of planning-graphs and large improvements over traditional sampling approaches. We demonstrate the approach on a variety of challenging planning problems, including real-world probabilistic occupancy maps and high-dof manipulation problems common in robotics.