基于人机博弈的驾驶权接管控制策略

doi:10.19562/j.chinasae.qcgc.2025.08.007

摘要/Abstract

摘要：

当前L3级别有条件自动驾驶能够在部分特定情况下独立执行驾驶任务，从而允许人类驾驶员参与驾驶无关任务，但当系统发出接管请求，如何在驾驶员状态恢复过程中实现从自动到手动驾驶的平稳过渡是当前自动驾驶应用的关键问题。本文面向L3自动驾驶中人类驾驶员接管控制权场景，根据系统发出接管请求后驾驶员的接管行为阶段，构建了驾驶权过渡过程中的人机非零和微分博弈框架，开发了基于单评价网络的事件触发自适应动态规划控制算法，通过求解人机双方耦合控制目标下的驾驶策略以及设计基于指数-三角优化算法的柔性控制权限转移，实现了安全、稳定的自动-手动驾驶模式切换。结果表明采用该驾驶员接管策略相比于刚性控制权切换方案时，系统能实现更理想的避障表现以及更稳定的车辆运动。

关键词: 自动驾驶, 接管控制, 非零和微分博弈, 指数-三角优化算法

Abstract:

Level 3 conditional autonomous driving systems can independently execute driving tasks in certain specific situation， allowing human drivers to get involved in tasks unrelated to driving. However， when the system sends out a takeover request， achieving a smooth transition from automatic to manual vehicle control during the driver's state recovery process is a key concern in current autonomous driving application. In this paper， for the takeover process of human driver in L3 autonomous driving， a human-machine non-zero-sum differential game framework is constructed for the authority transition based on the takeover behavior of human driver upon the system's takeover request and an event-triggered adaptive dynamic programming control algorithm is developed based on critic-only structure. By solving the driving strategies under the coupled control objectives of both players and designing exponential-trigonometric optimization algorithm-based flexible authority transfer strategy， a safe and stable switch between automatic and manual driving is achieved. The results show that the proposed takeover strategy has better obstacle avoidance performance and more stable vehicle motion than rigid takeover method.

Key words: autonomous driving, takeover control, non-zero differential game, exponential-trigonometric optimization algorithm

石英魁,胡川,邹建中,张希. 基于人机博弈的驾驶权接管控制策略[J]. 汽车工程, 2025, 47(8): 1501-1512.

Yingkui Shi,Chuan Hu,Jianzhong Zou,Xi Zhang. Takeover Control Strategy Based on Human-Machine Game[J]. Automotive Engineering, 2025, 47(8): 1501-1512.

图/表 19

图1

图2

图3

图4

图5

图6

图7

图8

表1

仿真参数"

参数	数值	参数	数值
$m$ /kg	1 412	$C f$ /（N·m^-1）	-112 600
$l f$ /m	1.015	$C r$ /（N·m^-1）	-94 548
$l r$ /m	1.895	$I z$ /（kg·m²）	1 536.7

表1

图9

图10

图11

图12

图13

图14

图15

图16

图17

图18

参考文献 28

[1]	ESKANDARIAN A， WU C， SUN C. Research advances and cha-llenges of autonomous and connected ground vehicles［J］. IEEE Transactions on Intelligent Transportation Systems， 2021， 22（2）： 683-711.
[2]	李梦凡，冯忠祥，张卫华，等. 面向人机共驾模式下驾驶人接管过程的视觉转移特性研究［J］. 汽车工程， 2024， 46（5）： 795-804.
	LI M F， FENG Z X， ZHANG W H， et al. Study on driver’s transfer characteristics during the takeover process of human-computer co-driving mode［J］. Automotive Engineering， 2024， 46（5）： 795-804.
[3]	FANG Z， WANG J， WANG Z， et al. Human-machine shared control for path following considering driver fatigue characteristics［J］. IEEE Transactions on Intelligent Transportation Systems， 2024， 25（7）： 7250-7264.
[4]	曾嘉宾，李靖宇，尤志栋. 自动驾驶人机共驾环境下驾驶人接管能力研究进展［J］. 交通科技与经济， 2022， 24（5）： 30-37.
	ZENG J B， LI J Y， YOU Z D. A review of progress on drivers’ take-over ability in automated driving human-machine codriving environment［J］. Technology & Economy in Areas of Communications， 2022， 24（5）： 30-37.
[5]	CHEN H， ZHAO X， LI Z， et al. Construction and analysis of driver takeover behavior modes based on situation awareness theory［J］. IEEE Transactions on Intelligent Vehicles， 2024， 9（2）： 4040-4054.
[6]	郭烈，胥林立，秦增科，等. 自动驾驶接管影响因素分析与研究进展［J］. 交通运输系统工程与信息， 2022， 22（2）： 72-90.
	GUO L， XU L L， QIN Z K， et al. Analysis and overview of influencing factors on autonomous driving takeover［J］. Journal of Transp-ortation Systems Engineering and Information Technology， 2022， 22（2）： 72-90.
[7]	孔召权. 面向人机共驾的冲突预测及协同控制策略研究［D］. 成都：电子科技大学， 2024.
	KONG Z Q. Research on conflict prediction and collaborative control strategy for driver-automation co-driving［D］. Chengdu： University of Electronic Science and Technology of China， 2024.
[8]	HU L， CAI H， HUANG J， et al. The challenges of driving mode switching in automated vehicles： a review［J］. IEEE Transactions on Vehicular Technology， 2024， 73（2）： 1777-1791.
[9]	刘莹. 考虑接管场景的自动-人工驾驶切换过程协同控制方法［D］. 长春：吉林大学， 2024.
	LIU Y. Collaborative control for automated-manual driving switching process considering takeover scenarios［D］. Changchun： Jilin University， 2024.
[10]	吴超仲，吴浩然，吕能超. 基于间接共享控制的智能车协同接管方法［J］. 中国公路学报， 2022， 35（3）： 101-114.
	WU C Z， WU H R， LV N C. Cooperative takeover method for automated vehicles based on indirect shared control［J］. China Journal of Highway and Transport， 2022， 35（3）： 101-114.
[11]	LV C， LI Y， XING Y， et al. Human-machine collaboration for automated driving using an intelligent two-phase haptic interface［J］. Advanced Intelligent Systems， 2021， 3（4）： 2000229.
[12]	ZHANG Z， WANG C， ZHAO W， et al. Driving authority allocation strategy based on driving authority real-time allocation domain［J］. IEEE Transactions on Intelligent Transportation Systems， 2022， 23（7）： 8528-8543.
[13]	ZHAO Y， XIE Z， XU C， et al. A novel haptic takeover method based on human-machine collaboration states［C］. 2024 14th Asian Control Conference （ASCC）， 2024： 362-367.
[14]	LI W， LI Q， LI S E， et al. Indirect shared control through non-zero sum differential game for cooperative automated driving［J］. IEEE Transactions on Intelligent Transportation Systems， 2022， 23（9）： 15980-15992.
[15]	MA B， LIU Y， NA X， et al. A shared steering controller design based on steer-by-wire system considering human-machine goal consis-tency［J］. Journal of the Franklin Institute， 2019， 356（8）： 4397-4419.
[16]	LI M， SONG X， CAO H， et al. Shared control with a novel dynamic authority allocation strategy based on game theory and driving safety field［J］. Mechanical Systems and Signal Processing， 2019， 124： 199-216.
[17]	SWAIN S K， RATH J J， VELUVOLU K C. Cooperative game theory based robust shared lateral control for driver vehicle interactions［J］. Control Engineering Practice， 2023， 141： 105678.
[18]	SUN W， WANG X， ZHANG C. A model-free control strategy for vehicle lateral stability with adaptive dynamic programming［J］. IEEE Transactions on Industrial Electronics， 2019， 67（12）： 10693-10701.
[19]	JIANG Y， JIANG Z P. Robust adaptive dynamic programming［M］. Hoboken， New Jersey： Wiley ： IEEE Press， 2017.
[20]	ZHANG K， SU R， ZHANG H G， et al. Adaptive resilient event-triggered control design of autonomous vehicles with an iterative single critic learning framework［J］. IEEE Transactions on Neural Networks and Learning Systems， 2021， 32（12）： 5502-5511.
[21]	WANG Y， DUAN G， LI P. Event-triggered adaptive sliding mode control of uncertain nonlinear systems based on fully actuated system approach［J］. IEEE Transactions on Circuits and Systems II： Express Briefs， 2024， 71（5）： 2749-2753.
[22]	江浩斌，袁培淳，杨昆，等. 基于Simscape的APA-EPS系统建模及人机共驾转向控制策略研究［J］. 汽车工程， 2023， 45（12）： 2209-2221.
	JIANG H B， YUAN P C， YANG K， et al. Simscape-based APA-EPS system modeling and human-machine co-driving steering control strategies research［J］. Automotive Engineering， 2023， 45（12）： 2209-2221.
[23]	LUO B， YANG Y， LIU D， et al. Event-triggered optimal control with performance guarantees using adaptive dynamic programming［J］. IEEE Transactions on Neural Networks and Learning Systems， 2020， 31（1）： 76-88.
[24]	XUE S， LUO B， LIU D. Event-triggered adaptive dynamic programming for unmatched uncertain nonlinear continuous-time systems［J］. IEEE Transactions on Neural Networks and Learning Systems， 2021， 32（7）： 2939-2951.
[25]	ZHAO B， LIU D， LUO C. Reinforcement learning-based optimal stabilization for unknown nonlinear systems subject to inputs with uncertain constraints［J］. IEEE Transactions on Neural Networks and Learning Systems， 2020， 31（10）： 4330-4340.
[26]	ZHANG Y， ZHAO B， LIU D， et al. Event-triggered control of discrete-time zero-sum games via deterministic policy gradient adaptive dynamic programming［J］. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 2022， 52（8）： 4823-4835.
[27]	YANG X， HE H. Adaptive critic designs for event-triggered robust control of nonlinear systems with unknown dynamics［J］. IEEE Transactions on Cybernetics， 2019， 49（6）： 2255-2267.
[28]	ZHANG Y， ZHAO B， LIU D， et al. Event-triggered optimal tracking control of multiplayer unknown nonlinear systems via adaptive critic designs［J］. International Journal of Robust and Nonlinear Control， 2022， 32（1）： 29-51.