Priority-Driven Safe Model Predictive Control Approach to Autonomous Driving Applications

2025-05-09

AI Summary

In autonomous driving, adaptive cruise control (ACC) and automated lane-changing systems struggle to simultaneously satisfy hard safety constraints—such as collision avoidance and lane-keeping—and soft comfort constraints under external disturbances. To address this, this paper proposes a priority-driven safe model predictive control (SMPC) framework. Methodologically, it integrates robust optimization with learning-based model approximation. Its key contributions are: (1) a novel priority-based constraint softening mechanism that rigorously guarantees zero violation of hard safety constraints; and (2) a learned approximate controller—implemented via neural networks—that enables real-time SMPC execution. Evaluated under realistic road scenarios with sudden disturbances, the approach achieves 100% satisfaction of all hard safety constraints, reduces control latency by 87%, and supports deployment on embedded automotive platforms.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements

Abstract

This paper demonstrates the applicability of the safe model predictive control (SMPC) framework to autonomous driving scenarios, focusing on the design of adaptive cruise control (ACC) and automated lane-change systems. Building on the SMPC approach with priority-driven constraint softening -- which ensures the satisfaction of emph{hard} constraints under external disturbances by selectively softening a predefined subset of adjustable constraints -- we show how the algorithm dynamically relaxes lower-priority, comfort-related constraints in response to unexpected disturbances while preserving critical safety requirements such as collision avoidance and lane-keeping. A learning-based algorithm approximating the time consuming SMPC is introduced to enable real-time execution. Simulations in real-world driving scenarios subject to unpredicted disturbances confirm that this prioritized softening mechanism consistently upholds stringent safety constraints, underscoring the effectiveness of the proposed method.

Problem

• Develops safe model predictive control for autonomous driving
• Ensures collision avoidance and lane-keeping under disturbances
• Enables real-time execution with learning-based SMPC approximation

Innovation

• Priority-driven constraint softening for safety
• Learning-based algorithm for real-time SMPC
• Dynamic relaxation of comfort-related constraints