Global Tensor Motion Planning

2024-11-28

arXiv.org

AI Summary

To address the low efficiency and limited trajectory diversity in batch motion planning for large-scale robotic learning, this paper proposes a fully tensorized global motion planning algorithm. Methodologically, we introduce a novel random multipartite graph discretization framework, theoretically guaranteeing probabilistic completeness; integrated with vectorized collision checking and graph-based batch search, the approach directly generates smooth spline trajectories without gradient-based optimization and natively supports GPU/TPU parallelization. Experiments on LiDAR occupancy grids and the MotionBenchMark benchmark demonstrate that our method achieves several-fold improvements in planning throughput over state-of-the-art baselines, while simultaneously delivering high real-time performance, strong robustness against environmental uncertainties, and enhanced trajectory diversity.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Batch planning is increasingly necessary to quickly produce diverse and high-quality motion plans for downstream learning applications, such as distillation and imitation learning. This paper presents Global Tensor Motion Planning (GTMP) -- a sampling-based motion planning algorithm comprising only tensor operations. We introduce a novel discretization structure represented as a random multipartite graph, enabling efficient vectorized sampling, collision checking, and search. We provide a theoretical investigation showing that GTMP exhibits probabilistic completeness while supporting modern GPU/TPU. Additionally, by incorporating smooth structures into the multipartite graph, GTMP directly plans smooth splines without requiring gradient-based optimization. Experiments on lidar-scanned occupancy maps and the MotionBenchMarker dataset demonstrate GTMP's computation efficiency in batch planning compared to baselines, underscoring GTMP's potential as a robust, scalable planner for diverse applications and large-scale robot learning tasks.

Problem

• Global Motion Planning
• Batch Planning
• Robotics Learning

Innovation

• GTMP (Global Tensor Motion Planning)
• Probabilistic Completeness
• Smooth Path Generation