基于融合感知的自动驾驶汽车AEB控制研究

doi:10.3969/j.issn.1674-8484.2025.04.013

摘要/Abstract

摘要：

针对现有自动紧急制动（AEB）系统在复杂场景中易出现障碍物误识别、未充分考量前车加速度干扰及控制精度不足等问题，该文提出了一种融合视觉与激光雷达（LiDAR）感知的障碍物检测方法。通过设计基于模型预测控制（MPC）的分级 AEB 控制策略获得期望的制动减速度，并采用比例-积分-微分（PID）控制器对车辆的制动器主缸压力进行控制。结果表明：提出的障碍物检测方法能够在复杂场景中准确识别障碍物；控制器能够使车辆在不同 AEB 测试工况下的减速率达到100%，制动加速度能够按期望值输出；提出的方法能够有效保证自动紧急制动过程的安全性和舒适性。

关键词: 汽车安全, 智能汽车, 自动紧急制动（AEB）, 障碍物检测, 模型预测控制（MPC）

Abstract:

To address the limitations of existing automatic emergency braking (AEB) systems—such as the susceptibility to obstacle misidentification in complex scenarios, the insufficient consideration of the preceding vehicle's acceleration, and the lack of control precision—this paper proposed an obstacle detection approach that integrated visual and LiDAR perception. A hierarchical AEB control strategy based on model predictive control (MPC) was designed to determine the desired braking deceleration, and a proportional-integral-derivative (PID) controller was employed to regulate the vehicle's brake master cylinder pressure. The results show that the proposed obstacle detection method can accurately identify obstacles in complex scenarios. Furthermore, the controller enables the vehicle to achieve a 100% deceleration rate across various AEB test scenarios, with braking acceleration being output as intended. The proposed methodology effectively enhances both safety and ride comfort during the automatic emergency braking process.

Key words: vehicle safety, Intelligent vehicle, autonomous emergency break (AEB), obstruction detection, model predictive control (MPC)

中图分类号:

U461.91

高超俊, 李祎承, 蔡英凤, 王海, 蒋金. 基于融合感知的自动驾驶汽车AEB控制研究[J]. 汽车安全与节能学报, 2025, 16(4): 629-637.

GAO Chaojun, LI Yicheng, CAI Yingfeng, WANG Hai, JIANG Jin. Research on AEB control of autonomous vehicles based on sensor fusion perception[J]. Journal of Automotive Safety and Energy, 2025, 16(4): 629-637.

图/表 13

图1 纵向车间动力学模型

整备质量，m	1 412 kg
车长，L	4.50 m
轴距，L₀	2.91 m
车宽，B	1.90 m
路面附着系数，μ	0.85 C
最大制动力，p_bm	15 MPa

表1 整车主要参数

图2 技术路线框架图

场景	v_s / (m·s^-1)	v_f / (m·s^-1)	d_r0 / m	a_f / (m·s^-2)
CCR_s1	5.56	0	50	0
CCR_s2	11.11	0	50	0
CCR_m1	8.33	5.56	50	0
CCR_m2	18.06	5.56	50	0
CCR_b1	13.90	13.90	40	-2
CCR_b2	13.90	13.90	40	-4
CCR_b3	20.00	13.90	40	-6
CCR_b4	25.00	13.90	40	-6

表2 AEB仿真工况场景

距离/ m	本方法平均精度/ %			YOLOv11平均精度/ %
距离/ m	5 m/s	10 m/s	15 m/s	5 m/s	10 m/s	15 m/s
50	85	81	79	82	78	76
30	88	85	80	85	82	76
20	89	87	85	86	84	81

表3 本文方法与YOLOv11精度对比

图3 CCRS工况相对距离与风险等级测试结果

图4 CCRS工况自车车速与加速度测试结果

图5 CCRm工况相对距离与风险等级测试结果

图6 CCRm工况自车车速与加速度测试结果

图7 CCRb工况相对距离与风险等级测试结果

图8 CCRb工况自车车速与加速度测试结果

图9 非标准CCRb工况2种策略控制下相对距离仿真结果对比

图10 非标准CCRb工况2种策略下自车加速度仿真结果对比

参考文献 21

[1]	中华人民共和国公安部. 全国机动车保有量达4.53亿辆/驾驶人达5.42亿人[R/OL]. (2025-01-18) https://www.mps.gov.cn/n2254314/n6409334/c9939035/content.html.
	Ministry of Public Security of the People's Republic of China. The number of motor vehicles in China has reached 453 million, and the number of drivers has reached 542 million[R/OL]. (2025-01-18) https://www.mps.gov.cn/n2254314/n6409334/c9939035/content.html. (in Chinese)
[2]	国柳鹏, 赵克刚, 梁志豪, 等. 基于深度强化学习CLPER-DDPG的车辆纵向速度规划[J]. 汽车安全与节能学报, 2024, 15(5): 702-710.
	GUO Liupeng, ZHAO Kegang, LIANG Zhihao, et al. Vehicle longitudinal speed planning based on deep reinforcement learning CLPER-DDPG[J]. J Autom Safe Energ, 2024, 15(5): 702-710. (in Chinese)
[3]	辛佳庚, 杨复钰, 张宝迪, 等. 基于融合算法的电动汽车AEB控制策略[J]. 北京交通大学学报, 2021, 45(6): 77-86, 93. doi: 10.11860/j.issn.1673-0291.20210029
	XIN Jiageng, YANG Fuyu, ZHANG Baodi, et al. AEB control strategy of electric vehicle based on fusion algorithm[J]. J Beijing Jiaotong Univ, 2021, 45(6): 77-86, 93. (in Chinese) doi: 10.11860/j.issn.1673-0291.20210029
[4]	胡朝辉, 黄顺霞, 杜展鹏, 等. 基于横向安全距离模型的主动避障算法[J]. 汽车工程, 2020, 42(5): 581-587.
	HU Zhaohui, HUANG Shunxia, DU Zhanpeng, et al. Active obstacle avoidance algorithm based on lateral safety distance model[J]. Autom Engineering, 2020, 42(5): 581-587. (in Chinese)
[5]	兰凤崇, 余蒙, 李诗成, 等. 考虑预碰撞时间的自动紧急制动系统分层控制策略研究[J]. 汽车工程, 2020, 42(2): 206-214.
	LAN Fengchong, YU Meng, LI Shicheng, et al. Research on hierarchical control strategy for automatic emergency braking system with consideration of time-to-collision[J]. Autom Engineering, 2020, 42(2): 206-214. (in Chinese)
[6]	郭祥靖, 孙攀, 邓杰, 等. 基于BP神经网络算法预测的重型半挂汽车列车AEB控制策略研究[J]. 汽车工程, 2021, 43(9): 1350-1359, 1366.
	GUO Xiangjing, SUN Pan, DENG Jie, et al. Research on AEB control strategy of a heavy tractor-semitrailer combination based on BP neural network algorithm prediction[J]. Autom Engineering, 2021, 43(9): 1350-1359, 1366. (in Chinese)
[7]	胡远志, 刘俊生, 何佳, 等. 基于激光雷达点云与图像融合的车辆目标检测方法[J]. 汽车安全与节能学报, 2019, 10(4): 451-458.
	HU Yuanzhi, LIU Junsheng, HE Jia, et al. Vehicle target detection method based on lidar point cloud and image fusion[J]. J Autom Safe Energ, 2019, 10(4): 451-458. (in Chinese)
[8]	LI Yicheng, ZHONG Wei, CAI Yingfeng, et al. A new visual sensing system for motion state estimation of lateral localization of intelligent vehicles[J]. Measurement, 2024, 237: 115212.
[9]	CUI Yaodong, CHEN Ren, CHU Wenbo, et al. Deep learning for image and point cloud fusion in autonomous driving: A review[J]. IEEE Trans Intel Transport Syst, 2022, 23(2): 722-739.
[10]	ZHANG Lei, LI Xu, TANG Kaichen, et al. FS-Net: LiDAR-camera fusion with matched scale for 3D object detection in autonomous driving[J]. IEEE Trans Intel Transport Syst, 2023, 24(11): 12154-12165.
[11]	李从庆. 基于障碍物聚类分析的物流无人车环境感知研究[J]. 佳木斯大学学报(自然科学版), 2024, 42(6): 138-141.
	LI Congqing. Research on environment perception of logistics autonomous vehicles based on obstacle clustering analysis[J]. J Jiamusi Univ (Nat Sci Edit), 2024, 42(6): 138-141. (in Chinese)
[12]	李研芳, 黄影平. 基于激光雷达和相机融合的目标检测[J]. 电子测量技术, 2021, 44(5): 112-117.
	LI Yanfang, HUANG Yingping. Target detection based on fusion of lidar and camera[J]. Elect Meas Tech, 2021, 44(5): 112-117. (in Chinese)
[13]	刘永涛, 刘传攀, 刘湘安, 等. 基于自适应采样时间MPC的自动紧急制动系统[J]. 汽车工程, 2023, 45(1): 32-41.
	LIU Yongtao, LIU Chuanpan, LIU Xiangan, et al. Automatic emergency braking system based on model predictive control with adaptive sampling time[J]. Autom Engineering, 2023, 45(1): 32-41. (in Chinese)
[14]	陆颖, 陈烨, 杨鹏, 等. 基于自适应权重MPC的AEB-P控制策略研究[J]. 河北科技大学学报, 2024, 45(4): 341-350.
	LU Ying, CEHN Ye, YANG Peng, et al. Research on AEB-P control strategy based on adaptive weight MPC[J]. J Hebei Univ Sci Tech, 2024, 45(4): 341-350. (in Chinese)
[15]	段顺昌, 白先旭, 石琴, 等. 汽车自动紧急制动系统控制策略的预期功能安全设计[J]. 汽车工程, 2022, 44(9): 1305-1317, 1338.
	DUAN Shunchang, BAI Xianxu, SHI Qin, et al. The design of the safety of the intended functionality of the control strategies for vehicle automatic emergency braking system[J]. Autom Engineering, 2022, 44(9): 1305-1317, 1338. (in Chinese)
[16]	WANG Dingran, TAN Jiasheng, WANG Hong, et al. SDS-YOLO: An improved vibratory position detection algorithm based on YOLOv11[J]. Measurement, 2025, 244: No 116518.
[17]	李彬, 李生林. 改进YOLOv11n的无人机小目标检测算法[J]. 计算机工程与应用, 2025, 61(7): 96-104. doi: 10.3778/j.issn.1002-8331.2411-0072
	LI Bin, LI Shenglin. Improved YOLOv11n small object detection algorithm in UAV view[J]. Comput Engi Appl, 2025, 61(7): 96-104. (in Chinese)
[18]	CHEN Jianhong, LIN Guohan, Yelamandala C, et al. High-accuracy mapping design based on multi-view images and 3d lidar point clouds[C]// IEEE Int’l Conf Consum Electr (ICCE), Kuala Lumpur, Malaysia: IEEE Press, 2020: 1-2.
[19]	张晨, 刘畅, 赵津, 等. 基于多尺度注意力机制的实时激光雷达点云语义的分割[J]. 汽车安全与节能学报, 2024, 15(4): 591-601.
	ZHANG Chen, LIU Chang, ZHAO Jin, et al. Semantic segmentation of real-time LiDAR point clouds based on multi-scale self-attention[J]. J Autom Safe Energ, 2024, 15(4): 591-601. (in Chinese)
[20]	张炳力, 潘泽昊, 姜俊昭, 等. 基于交叉注意力机制的多模态感知融合方法[J]. 中国公路学报, 2024, 37(3): 181-193. doi: 10.19721/j.cnki.1001-7372.2024.03.009
	ZHANG Bingli, PAN Zehao, JIANG junzhao, et al. Multi-modal perception fusion method based on cross attention[J]. China J Highw Transport, 2024, 37(3): 181-193. (in Chinese)
[21]	WANG Yuzhong, JIN Qibing, ZHANG Ridong. Improved fuzzy PID controller design using predictive functional control structure[J]. ISA Trans, 2017, 71: 354-363.