基于混合增强智能的人机混合决策策略研究

doi:10.3901/JME.2025.10.288

摘要/Abstract

摘要： 为克服驾驶人对自动驾驶系统可接受度弱问题，提高系统在真实道路上的安全性与适应性，提出基于混合增强智能的人机混合决策策略。混合增强智能理论可处理具备“黑-白盒”属性的人机系统交互问题，并解耦混合决策关系。通过驾驶权分配机制与人机决策知识库，实现人机混合决策并充分发挥人机系统“1+1＞2”的决策优势。建立考虑驾驶能力与人机轨迹相似度的驾驶权分配机制，分别通过主客观相结合方法与反向传播(Back propagation，BP)神经网络，对驾驶能力进行分类与辨识。建立基于长短期记忆网络(Long short-term memory，LSTM)的自动驾驶子系统决策逻辑，并对其决策性能进行评估。将包含安全性和舒适性的综合评价函数作为人机共驾系统决策评价准则，并通过主观问卷评价驾驶人对系统的可接受度。采用具备人在回路功能的高精度智能驾驶模拟平台，进行试验数据采集与系统性能测试。结果表明，在双车道和环岛场景中施加不同程度驾驶人激励时，所提出的人机混合决策策略性能优于仅驾驶人、全自动驾驶系统以及基于博弈框架的人机共驾系统决策策略性能。

关键词: 车辆工程, 人机共驾, 人机混合决策, 驾驶能力, 混合增强智能

Abstract: In order to overcome weak driver acceptability automated vehicles and improve system safety and adaptability on the real road, a human-machine hybrid decision-making strategy basing on the hybrid-augmented intelligence theory is proposed. The hybrid-augmented intelligence theory aims at dealing with the interaction issue between human and automated systems with “black and white box” property, and decouple their hybrid decision-making relationship. Basing on the driving authority allocation mechanism and human-machine decision-making knowledge base, the proposed strategy can be achieved and its advantages on decision-making of “1+1>2” between human and machine can be fully utilized as well. The driving authority allocation mechanism considering driving ability and human-machine trajectory similarity is established and the driving ability is classified and identified basing on the combination of subjective and objective methods and BP neural network. The decision-making logic of the automated vehicle subsystem is modeled basing on the LSTM network and its corresponding performance is evaluated. The comprehensive evaluation function consisting of the driving safety and comfort is proposed as the decision-making evaluation criterion for the shared control, and drivers’ acceptability is evaluated basing on subjective questionnaires. Using a high-precision intelligent driving simulation platform with human in the loop function for experimental data collection and system performance testing. The results show that when different levels of driver incentives are applied in the two lane and roundabout scenarios, the performance of the proposed human-machine hybrid decision-making strategy is better than that of the driver only, the full auto drive system and the shared control system based on the game framework.

Key words: vehicle engineering, shared control, human-machine hybrid decision-making, driving ability, hybrid-augmented intelligence

中图分类号:

U461

马文霄, 孙博华, 赵帅, 代凯, 赵航, 吴坚. 基于混合增强智能的人机混合决策策略研究[J]. 机械工程学报, 2025, 61(10): 288-304.

MA Wenxiao, SUN Bohua, ZHAO Shuai, DAI Kai, ZHAO Hang, WU Jian. Research on the Human-machine Hybrid Decision-making Strategy Basing on the Hybrid-augmented Intelligence[J]. Journal of Mechanical Engineering, 2025, 61(10): 288-304.

参考文献

[1] HUANG C，LV C，HANG P. Human machine adaptive shared control for safe driving under automation degradation[J]. IEEE Intelligent Transportation Systems Magazine，2021，14(2)：53-66.
[2] MUHAMMAD K，ULLAH A，LLORET J. Deep learning for safe autonomous driving：Current challenges and future directions[J]. IEEE Transactions on Intelligent Transportation Systems，2021，22(7)：4316-4336.
[3] JIANG B，LI X，ZENG Y. Human-machine cooperative trajectory planning for semi-autonomous driving based on the understanding of behavioral semantics[J]. Electronics，2021，10(8)：946-968.
[4] BENLOUCIF A，NGUYEN A T，SENTOUH C. Cooperative trajectory planning for haptic shared control between driver and automation in highway driving[J]. IEEE Transactions on Industrial Electronics，2019，66(12)：9846-9857.
[5] 宗长富，代昌华，张东. 智能汽车的人机共驾技术研究现状和发展趋势[J]. 中国公路学报，2021，34(6)：214-237.ZONG Changfu，DAI Changhua，ZHANG Dong. Human-machine interaction technology of intelligent vehicles: current development trends and future directions[J]. China Journal of Highway and Transport，2021，34(6)：214-237.
[6] 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.
[7] 李克强，戴一凡，李升波，等. 智能网联汽车(ICV)技术的发展现状及趋势[J]. 汽车安全与节能学报, 2017, 8(1)：1-14. LI Keqiang，DAI Yifan，LI Shengbo，et al. State-of-the-art and technical trends of intelligent and connected vehicles[J]. Journal of Automotive Safety and Energy，2017，8(1)：1-14.
[8] 郭柏苍，王胤霖，谢宪毅，等. 基于人-车风险状态的人机共驾控制权决策方法[J]. 中国公路学报，2022，35(3)：153-165.GUO Bocang，WANG Yinlin，XIE Xianyi，et al. Decision making method for control right transition of human-machine shared driving based on driver-vehicle risk state[J]. China Journal of Highway and Transport，2022，35(3)：153-165.
[9] NGUYEN A T，RATH J J，LV C，et al. Human-machine shared driving control for semi-autonomous vehicles using level of cooperativeness[J]. Sensors，2021，21(14)：4647-4666.
[10] NA X，DAVID C. Theoretical and experimental investigation of driver noncooperative-game steering control behavior[J]. IEEE/CAA Journal of Automatica Sinica，2021，8(1)：189-205.
[11] LI W Y，LI Q K，LI S B 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.
[12] JI X，YANG K，NA X，et al. Shared steering torque control for lane change assistance：A stochastic game-theoretic approach[J]. IEEE Transactions on Industrial Electronics，2019，66(4)：3093-3105.
[13] NGUYEN A T，SENTOUCH C，POPIEUL J C. Driver-automation cooperative approach for shared steering control under multiple system constraints：Design and experiments[J]. IEEE Transactions on Industrial Electronics，2017，64(5)：3819-3830.
[14] SENTOUCH C, NGUYEN A T，BENLOUCIF A，et al. Driver-automation cooperation oriented approach for shared control of lane keeping assist systems[J]. IEEE Transactions on Control Systems Technology，2019，27：1962-1978.
[15] FLAD M，FROHLICH L，HOHMANN S. Cooperative shared control driver assistance systems based on motion primitives and differential games[J]. IEEE Transactions on Human-Machine Systems，2017，47(5)：711-722.
[16] WANG W，NA X，CAO D，et al. Decision-making in driver-automation shared control：A review and perspectives[J]. IEEE/CAA Journal of Automatica Sinica，2020，7(5)：1289-1307.
[17] ZHENG N，LIU Z，REN P，et al. Hybrid-augmented intelligence: collaboration and cognition[J]. Frontiers of Information Technology & Electronic Engineering，2017，18(2)：153-179.
[18] HAN Z，CHEN M，ZHOU T，et al. Path planning of unmanned autonomous helicopter based on human- computer hybrid augmented intelligence[J]. Neural Plasticity，2021，2021：1-22.
[19] 姚建国，余涛，杨胜春，等. 提升电网调度中人工智能可用性的混合增强智能知识演化技术[J]. 电力系统自动化，2022，46(20)：1-12.YAO Jianguo，YU Tao，YANG Shengchun，et al. Knowledge evolution technology based on hybrid- augmented intelligence for improving practicability of artificial intelligence in power grid dispatch[J]. Automation of Electric Power Systems，2022，46(20)：1-12.
[20] 付海军，陈世超，林懿伦，等. 人在回路的混合增强智能在Sawyer的研究与验证[J]. 智能科学与技术学报，2019，1(3)：280-286.FU Haijun，CHEN Shichao，LIN Yilun，et al. Research and validation of human-in-the-loop hybrid-augmented intelligence in sawyer[J]. Chinese Journal of Intelligent Science and Technology，2019，1(3)：280-286.
[21] LIU F，ZHU T，SU W. Servo-level human-machine shared control flexible strategy based on driving ability, status and regionalized environmental risk[J]. IEEE Transactions on Transportation Electrification，2023，9(3)：4418-4436.
[22] 张杨，王晓原，刘士杰，等. 基于遗传优化算法的疲劳驾驶状态辨识方法[J]. 传感器与微系统，2023(9)：157-161.ZHANG Yang，WANG Xiaoyuan，LIU Shijie，et al. Identification method of fatigue driving state based on genetic optimization algorithm[J]. Transducer and Microsystem Technologies，2023(9)：157-161.
[23] ZEPEDA M V N D，MENG F. Dynamic clustering analysis for driving styles identification[J]. Engineering Applications of Artificial Intelligence，2021，97：101096.
[24] YU R，LONG X，QUDDUS M，et al. A bayesian tobit quantile regression approach for naturalistic longitudinal driving capability assessment[J]. Accident Analysis and Prevention，2020，147：105779.
[25] TANG F，GAO F，WANG Z. Driving capability-based transition strategy for cooperative driving：From manual to automatic[J]. IEEE Access 2020，8：139013-139022.
[26] 孙博华，邓伟文，吴坚，等. 虚拟随机车路场下驾驶人驾驶能力机理分析[J]. 机械工程学报，2020，56(16)：166-180.SUN Bohua，DENG Weiwen，WU Jian，et al. Mechanism analysis of driving capability in the virtual random vehicle-road field[J]. Journal of Mechanical Engineering，2020，56(16)：166-180.
[27] HOCHREITER S，SCHMIDHUBER J. Long short-term memory[J]. Neural Computation，1997，9(8)：1735-1780.
[28] OZKAN T，LAJUNEN T. What causes the differences in driving between young men and women? The effects of gender roles and sex on young drivers’ driving behaviour and self-assessment of skills[J]. Transportation Research Part F：Psychology & Behaviour，2006，9(4)：269-277.
[29] 陈嘉航. 自动驾驶汽车速度控制策略自动化测试与评价研究[D]. 长春：吉林大学，2022.CHEN Jiahang. Research on automatic test and evaluation of speed control strategy for autonomous vehicle[D]. Changchun：Jilin University，2022.
[30] JENNY E. Tuning for ride quality in autonomous vehicle：Application to linear quadratic path planning algorithm[D]. Uppsala：Uppsala University，2015.
[31] 朱向雷，杜志彬. 自动驾驶场景仿真与ASAM OpenX标准应用[M]. 北京：机械工业出版社，2023.ZHU Xianglei，DU Zhibin. Scenario-based autonomous driving simulation technology and application of ASAM OpenX standards[M]. Beijing：China Machine Press，2023.