基于并行行为树架构的自动驾驶汽车行为控制技术研究

doi:10.16182/j.issn1004731x.joss.24-0126

摘要/Abstract

摘要：

针对传统串行行为树在自动驾驶汽车控制中高碰撞率和低效率问题，探讨了基于改进并行行为树架构的解决方案以实现安全行为控制。提出了动态路况下的安全行为控制策略，并构建了观察、决策、移动的行为模型及其时序约束关系；提出了一种改进的并行行为树控制架构，通过并行控制节点实现行为的并行执行与实时交互，提升决策控制的实时性。结果表明：与传统串行行为树相比，并行行为树架构将碰撞率降低了23%，平均速度提高了5.4%，响应时间减少了56.6%；与基于模型预测控制的架构相比，碰撞率降低了2%，平均速度提升了1.6%，响应时间减少了15.7%

关键词: 自动驾驶汽车, 行为树控制架构, 并行控制节点, 安全行为控制策略

Abstract:

Aiming at the problem of high collision rate and low efficiency of traditional serial behavior tree in autonomous vehicle control, a solution based on improved parallel behavior tree architecture is discussed to achieve safe behavior control. A safety behavior control strategy under dynamic road conditions is proposed, and behavior models for observation, decision-making, and movement are constructed, as well as their temporal constraint relationships; an improved parallel behavior tree control architecture is proposed, which achieves parallel execution and real-time interaction of behaviors through parallel control nodes, improving the real-time performance of decision control. The results show that compared with traditional serial behavior trees, the parallel behavior tree architecture reduces collision rates by 23%, increases average speed by 5.4%, and reduces response time by 56.6%; compared with the architecture based on MPC, the collision rate is reduced by 2%, the average speed increased by 1.6%, and the response time is reduced by 15.7%

Key words: autonomous vehicles, behavior tree control architecture, parallel control nodes, safe behavior control strategy

中图分类号:

TP391

袁健钞,杨硕,张琪等 . 基于并行行为树架构的自动驾驶汽车行为控制技术研究[J]. 系统仿真学报, 2025, 37(6): 1542-1554.

Yuan Jianchao,Yang Shuo,Zhang Qi,et al . Research on Behavior Control Techniques for Autonomous Vehicles Based on Parallel Behavior Tree Architecture[J]. Journal of System Simulation, 2025, 37(6): 1542-1554.

图/表 16

图1

图2

图 3

图 4

图5

图 6

图7

图 8

图 9

图10

图11

图12

图13

图 14

表1

表2

参考文献 27

1	张国辉, 王璇, 张雅楠, 等. 实际环境中多无人车协同路径规划模型研究[J]. 系统仿真学报, 2023, 35(2): 408-422.
	Zhang Guohui, Wang Xuan, Zhang Yanan, et al. Research on Cooperative Path Planning Model of Multiple Unmanned Vehicles in Real Environment[J]. Journal of System Simulation, 2023, 35(2): 408-422.
2	王若萱, 吴建平, 徐辉. 自动驾驶汽车感知系统仿真的研究及应用综述[J]. 系统仿真学报, 2022, 34(12): 2507-2521.
	Wang Ruoxuan, Wu Jianping, Xu Hui. Overview of Research and Application on Autonomous Vehicle Oriented Perception System Simulation[J]. Journal of System Simulation, 2022, 34(12): 2507-2521.
3	Jamal Mais, Panov Aleksandr. Adaptive Maneuver Planning for Autonomous Vehicles Using Behavior Tree on Apollo Platform[C]//Artificial Intelligence XXXVIII. Cham: Springer International Publishing, 2021: 327-340.
4	Hatcher J. A Reflective, 3-dimensional Behavior Tree Approach to Vehicle Autonomy[D]. Tallahassee: Florida State University, 2015.
5	Zhang Qi, Yao Jian, Yin Quanjun, et al. Learning Behavior Trees for Autonomous Agents with Hybrid Constraints Evolution[J]. Applied Sciences, 2018, 8(7): 1077.
6	Cvetkov Kiril. Path-planning[EB/OL]. [2023-10-10]. .
7	Choi Yoonsuk, Lee Wonwoo, Kim Jeesu, et al. A Variable-sampling Time Model Predictive Control Algorithm for Improving Path-tracking Performance of a Vehicle[J]. Sensors, 2021, 21(20): 6845.
8	Honda Kohei, Okuda Hiroyuki, Suzuki Tatsuya, et al. Connection of Nonlinear Model Predictive Controllers for Smooth Task Switching in Autonomous Driving[J]. Asian Journal of Control, 2023, 25(3): 1805-1822.
9	Li Haoran, Zhang Tingyang, Zheng Sifa, et al. Distributed MPC for Multi-vehicle Cooperative Control Considering the Surrounding Vehicle Personality[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(3): 2814-2826.
10	Woo Young Choi, Lee Seung-Hi, Chung Choo Chung. Horizonwise Model-predictive Control with Application to Autonomous Driving Vehicle[J]. IEEE Transactions on Industrial Informatics, 2022, 18(10): 6940-6949.
11	Nair S H, Lee H, Joa E, et al. Predictive Control for Autonomous Driving with Uncertain, Multimodal Predictions[J/OL]. IEEE Transactions on Control Systems Technology. [2023-11-09]. .
12	吴永刚, 梁华为. 基于自适应模型预测的无人车跟踪控制方法研究[J]. 制造业自动化, 2023, 45(2): 198-202.
	Wu Yonggang, Liang Huawei. Research on Obstacles Avoidance Control Method of Unmanned Ground Vehicles Based on Adaptive Model Prediction[J]. Manufacturing Automation, 2023, 45(2): 198-202.
13	Wu Yuanqing, Liao Siqin, Liu Xiang, et al. Deep Reinforcement Learning on Autonomous Driving Policy with Auxiliary Critic Network[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(7): 3680-3690.
14	Fu Yiwei, Jha D K, Zhang Zeyu, et al. Neural Network-based Learning from Demonstration of an Autonomous Ground Robot[J]. Machines, 2019, 7(2): 24.
15	Meyer Eivind, Brenner Maurice, Zhang Bowen, et al. Geometric Deep Learning for Autonomous Driving: Unlocking the Power of Graph Neural Networks with Common Road-Geometric[EB/OL]. (2023-04-24) [2023-11-20]. .
16	Kwon Yonghun, Kim Woojae, Jung Inbum. Neural Network Models for Driving Control of Indoor Autonomous Vehicles in Mobile Edge Computing[J]. Sensors, 2023, 23(5): 2575.
17	Claudiu Radu Pozna, Antonya Csaba. Proposal of an Autonomous Vehicle Control Architecture[C]//2021 IEEE 25th International Conference on Intelligent Engineering Systems (INES). Piscataway: IEEE, 2021: 37-42.
18	Xia Yingji, Geng Maosi, Chen Yong, et al. Understanding Common Human Driving Semantics for Autonomous Vehicles[J]. Patterns, 2023, 4(7): 100730.
19	Fu Mengyin, Zhang Ting, Song Wenjie, et al. Trajectory Prediction-based Local Spatio-temporal Navigation Map for Autonomous Driving in Dynamic Highway Environments[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(7): 6418-6429.
20	Chao Qianwen, Bi Huikun, Li Weizi, et al. A Survey on Visual Traffic Simulation: Models, Evaluations, and Applications in Autonomous Driving[J]. Computer Graphics Forum, 2020, 39(1): 287-308.
21	Apollo. Apollo Simulation[EB/OL]. [2018-09-03]. .
22	Best A, Narang S, Pasqualin L, et al. AutonoVi-sim: Autonomous Vehicle Simulation Platform with Weather, Sensing, and Traffic Control[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Piscataway: IEEE, 2018: 1161-11618.
23	Dosovitskiy Alexey, Ros German, Codevilla Felipe, et al. CARLA: An Open Urban Driving Simulator[C]//Proceedings of the 1st Annual Conference on Robot Learning. Chia Laguna Resort: PMLR, 2017: 1-16.
24	尹智帅, 何嘉雄, 聂琳真, 等. 基于优化算法的自动驾驶车辆纵向自适应控制[J]. 系统仿真学报, 2021, 33(2): 409-420.
	Yin Zhishuai, He Jiaxiong, Nie Linzhen, et al. Longitudinal Adaptive Control of Autonomous Vehicles Base on Optimization Algorithm[J]. Journal of System Simulation, 2021, 33(2): 409-420.
25	Li Yanfeng, Guan H, Jia Xin, et al. Decision-making Model for Dynamic Scenario Vehicles in Autonomous Driving Simulations[J]. Applied Sciences, 2023, 13(14): 8515.
26	Feng Shuo, Yan Xintao, Sun Haowei, et al. Intelligent Driving Intelligence Test for Autonomous Vehicles with Naturalistic and Adversarial Environment[J]. Nature Communications, 2021, 12(1): 748.
27	Colledanchise Michele, Ögren Petter. Behavior Trees in Robotics and AI: An Introduction[M]. Boca Raton: CRC Press, 2018.

参数名称	参数大小	参数类型	作用
速度	0	初始状态参数	为主车设定初始速度
加速度	0	初始状态参数	为主车设定初始加速度
加加速度	0	初始状态参数	为主车设定初始加加速度
所在车道	1	初始状态参数	为主车设定初始车道
最大速度/(km/h)	80.47	边界条件参数	控制主车的最大速度
最小速度/(km/h)	64.37	边界条件参数	控制主车的最小速度
最大横向/纵向加速度/(m/s²)	10	边界条件参数	控制主车的最大加速度
安全行驶时长	—	时间参数	记录最长安全行驶时长
当前行驶时长	—	时间参数	记录当前行驶时长
仿真步长/s	1	时间参数	描述仿真间隔
仿真总时长	5 min 30 s	时间参数	控制一次实验的总时长

结构	碰撞概率/%	总平均速度/(km/h)	总平均纵向加速度/(m/s²)	总平均横向加速度/(m/s²)	总平均响应时间/s
传统行为树	30(9+21)	71.78	5.57	4.59	1.98
MPC控制	9(5+4)	74.43	4.98	3.16	1.02
并行行为树	7(4+3)	75.66	4.42	2.89	0.86

[1]	王琪玮, 张琪, 杨硕, 彭勇. 动态任务环境下智能体健壮行为树控制架构设计[J]. 系统仿真学报, 2025, 37(1): 54-65.
[2]	王若萱, 吴建平, 徐辉. 自动驾驶汽车感知系统仿真的研究及应用综述[J]. 系统仿真学报, 2022, 34(12): 2507-2521.