GPU-Accelerated Barrier-Rate Guided MPPI Control for Tractor-Trailer Systems

2025-08-07

AI Summary

To address the safety navigation challenge of articulated tractor-trailer systems during reverse parking in pedestrian-dense, narrow environments, this paper proposes a Barrier-Rate-guided Model Predictive Path Integral (BR-MPPI) control method. Our approach innovatively embeds Control Barrier Function (CBF) constraints directly into the importance sampling update of MPPI, dynamically reshaping the sampling distribution to enhance trajectory robustness and exploration capability. Integrating high-fidelity vehicle dynamics modeling, real-time collision avoidance, and GPU acceleration, the method achieves >100 Hz closed-loop control in CarMaker simulations. Experimental results demonstrate that BR-MPPI significantly improves parking success rate, trajectory smoothness, and safety over standard MPPI and collision-penalty baselines in multi-obstacle scenarios. This work establishes an efficient and reliable motion planning paradigm for autonomous heavy-vehicle parking in complex human-robot shared driving environments.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: Motion and Path Planning

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: Agentic search

Abstract

Articulated vehicles such as tractor-trailers, yard trucks, and similar platforms must often reverse and maneuver in cluttered spaces where pedestrians are present. We present how Barrier-Rate guided Model Predictive Path Integral (BR-MPPI) control can solve navigation in such challenging environments. BR-MPPI embeds Control Barrier Function (CBF) constraints directly into the path-integral update. By steering the importance-sampling distribution toward collision-free, dynamically feasible trajectories, BR-MPPI enhances the exploration strength of MPPI and improves robustness of resulting trajectories. The method is evaluated in the high-fidelity CarMaker simulator on a 12 [m] tractor-trailer tasked with reverse and forward parking in a parking lot. BR-MPPI computes control inputs in above 100 [Hz] on a single GPU (for scenarios with eight obstacles) and maintains better parking clearance than a standard MPPI baseline and an MPPI with collision cost baseline.

Problem

• Enables safe navigation for tractor-trailers in cluttered pedestrian areas
• Integrates Control Barrier Functions into path planning for collision avoidance
• Achieves real-time control at 100Hz on GPU for complex parking scenarios

Innovation

• GPU-accelerated BR-MPPI for real-time control
• CBF-embedded path-integral collision avoidance
• High-frequency tractor-trailer parking optimization