控制(管理)
理论(学习稳定性)
计算机科学
控制系统
工程类
人工智能
机器学习
电气工程
作者
Shaozhong Guo,Jun Guo,Yunqing Zhang,Jinglai Wu
出处
期刊:SAE technical paper series
日期:2024-04-09
被引量:1
摘要
<div class="section abstract"><div class="htmlview paragraph">Intelligent vehicle-to-everything connectivity is an important development trend in the automotive industry. Among various active safety systems, Autonomous Emergency Braking (AEB) has garnered widespread attention due to its outstanding performance in reducing traffic accidents. AEB effectively avoids or mitigates vehicle collisions through automatic braking, making it a crucial technology in autonomous driving. However, the majority of current AEB safety models exhibit limitations in braking modes and fail to fully consider the overall vehicle stability during braking. To address these issues, this paper proposes an improved AEB control system based on a risk factor (AERF). The upper-level controller introduces the risk factor (RF) and proposes a multi-stage warning/braking control strategy based on preceding vehicle dynamic characteristics, while also calculating the desired acceleration. Furthermore, a lower-level PID-based controller is designed to track the desired acceleration and compute the corresponding brake master cylinder pressure and throttle opening using an established inverse longitudinal dynamics model. Furthermore, to address vehicle stability during braking, an Anti-lock Braking System (ABS) controller is integrated with the proposed AERF. The effectiveness of the AERF is validated through software co-simulation and hardware-in-the-loop testing (HIL). The results demonstrate that the AERF can maintain a safe braking distance within 2 meters under Euro NCAP standard conditions, with excellent tracking performance of the actual braking deceleration and an error rate below 5%, ensuring a high level of system safety.</div></div>
科研通智能强力驱动
Strongly Powered by AbleSci AI