MP-RBFN: Learning-based Vehicle Motion Primitives using Radial Basis Function Networks

2025-07-14

AI Summary

Balancing trajectory accuracy and computational efficiency remains a key challenge in autonomous driving optimal control. To address this, we propose MP-RBFN—a motion primitive learning framework based on Radial Basis Function Networks (RBFNs). MP-RBFN uniquely bridges optimization-based high-fidelity dynamical modeling with sampling-based efficient trajectory generation, being the first to leverage RBFNs for learning parameterized motion primitives directly from optimal control solutions. It guarantees strict adherence to vehicle dynamics constraints while enabling millisecond-level inference. Experimental results demonstrate that MP-RBFN improves trajectory accuracy by 7× over state-of-the-art semi-analytical methods and achieves real-time planning performance. The implementation is open-sourced and has been successfully integrated into an industrial-grade trajectory planner, validating its practical efficacy.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

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

Abstract

This research introduces MP-RBFN, a novel formulation leveraging Radial Basis Function Networks for efficiently learning Motion Primitives derived from optimal control problems for autonomous driving. While traditional motion planning approaches based on optimization are highly accurate, they are often computationally prohibitive. In contrast, sampling-based methods demonstrate high performance but impose constraints on the geometric shape of trajectories. MP-RBFN combines the strengths of both by coupling the high-fidelity trajectory generation of sampling-based methods with an accurate description of vehicle dynamics. Empirical results show compelling performance compared to previous methods, achieving a precise description of motion primitives at low inference times. MP-RBFN yields a seven times higher accuracy in generating optimized motion primitives compared to existing semi-analytic approaches. We demonstrate the practical applicability of MP-RBFN for motion planning by integrating the method into a sampling-based trajectory planner. MP-RBFN is available as open-source software at https://github.com/TUM-AVS/RBFN-Motion-Primitives.

Problem

• Efficiently learning motion primitives for autonomous driving
• Combining high-fidelity trajectory generation with accurate dynamics
• Reducing computational cost while maintaining high accuracy

Innovation

• Uses Radial Basis Function Networks for learning
• Combines sampling and optimization-based methods
• Achieves high accuracy with low inference time