Branch-Stochastic Model Predictive Control for Motion Planning under Multi-Modal Uncertainty with Scenario Clustering

2026-05-21

AI Summary

This work addresses the challenge of multimodal uncertainties—such as diverse intentions and trajectories of surrounding vehicles—in autonomous driving motion planning, which often render conventional approaches either overly conservative or computationally intractable. The authors propose a Branching Stochastic Model Predictive Control (Branch-SMPC) framework that integrates multimodal intention modeling with chance constraints to generate distinct trajectories tailored to different predicted intentions. A novel scene clustering method based on high-level decision similarity and an adaptive branching timing mechanism are introduced to significantly reduce conservatism while preserving safety guarantees. Extensive simulations in complex highway scenarios demonstrate that the proposed approach achieves superior performance by effectively balancing real-time computational feasibility, safety, and planning quality.

Abstract

Motion planning for autonomous driving must account for multi-modal uncertainty in both the intentions and trajectories of surrounding vehicles. Handling uncertainty in a worst-case manner guarantees robustness but often leads to excessive conservatism. Stochastic Model Predictive Control (SMPC) reduces trajectory-level conservatism through chance constraints, yet remains conservative with respect to intention uncertainty since constraints must hold across all intentions. We present a novel combination of SMPC and the branching structure, enabling the planner to generate distinct trajectories for different possible intentions while maintaining safety under trajectory uncertainty. A novel scenario clustering is proposed to merge prediction scenarios based on high-level decision similarity, thereby ensuring real-time tractability. Furthermore, an adaptive branching-time computation postpones commitment to separate plans until intention uncertainty is sufficiently reduced. Simulation studies in challenging highway scenarios demonstrate that the proposed method improves safety, reduces conservatism, and achieves real-time computational performance.

Problem

• multi-modal uncertainty
• motion planning
• intention uncertainty
• trajectory uncertainty
• autonomous driving

Innovation

• Branch-Stochastic MPC
• multi-modal uncertainty
• scenario clustering
• adaptive branching
• motion planning