融合安全完整性框架与强化学习的自动驾驶换道决策方法
By combining a safety integrity framework with reinforcement learning, quantifying safety indicators and embedding the PPO algorithm, the shortcomings of the autonomous driving lane-changing decision model in safety constraint design are solved, and a balance between safety and efficiency of autonomous vehicles in complex environments is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HARBIN INST OF TECH
- Filing Date
- 2025-10-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing autonomous driving lane-changing decision-making models lack a unified paradigm in safety constraint design and do not strictly adhere to functional safety standards, resulting in safety hazards in complex and dynamic traffic environments.
A safety integrity framework is constructed, and exposure, severity, and controllability indicators are quantified. A total reward function for reinforcement learning is designed and integrated into the objective function of the PPO algorithm to form an adaptive safety constraint that ensures lane-changing behavior is within the functional safety boundary.
It achieves coordinated optimization of safety and driving efficiency of autonomous vehicles in complex traffic environments, ensuring that lane-changing behavior is within the absolute safety boundary and improving overall traffic efficiency.
Smart Images
Figure CN121291489B_ABST