Stein Variational Probabilistic Roadmaps

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. Video, additional material and results can be found here: https://sites.google.com/view/stein-prm.

Authors

• Alexander Lambert
• Brian Hou
• Rosario Scalise
• Siddhartha S. Srinivasa
• Byron Boots

Keywords

• Geometry
• Uncertainty
• Automation
• Autonomous systems
• Probabilistic logic
• Reliability engineering
• Planning
• Posterior Probability
• Regions Of Space
• Path Planning
• Feasible Set
• Configuration Space
• Variational Inference
• Occupancy Map
• Optimism
• Free Space
• Markov Chain Monte Carlo
• Random Generation
• Shortest Path
• Radial Basis Function Kernel
• Planning Phase
• Random Initialization
• Target Distribution
• Short Path
• Trajectory Optimization
• Simultaneous Localization And Mapping
• Part Of The Map
• Reproducing Kernel Hilbert Space
• Maximum Mean Discrepancy
• Safe Set
• Cost Path
• Cost Threshold
• Feasible Path
• Rejection Sampling
• Sample Distribution