|
|
自动驾驶车辆多风格自适应跟驰决策
|
Abstract:
本研究提出了一种适用于自动驾驶车辆的多风格自适应跟驰决策框架，结合深度强化学习(DRL)与改进的粒子群优化算法，实现了对不同驾驶风格的精确控制。该框架通过DRL算法设计自适应的跟驰速度控制策略，并基于驾驶安全性、效率、舒适性和紧急制动等因素构建了复杂的奖励机制。为进一步优化跟驰性能，本文在传统粒子群优化算法中引入了莱维扰动，以精确计算奖励函数中的最优权重组合，确保算法在不同驾驶场景下均能灵活应对。实验结果显示，该算法在多种复杂交通场景中具有鲁棒性，提供了更安全、高效且舒适的驾驶体验。
In this study, a multi-style adaptive following decision framework for self-driving vehicles is proposed, which combines deep reinforcement learning (DRL) with an improved particle swarm optimization algorithm to achieve accurate control of different driving styles. The framework designs an adaptive following speed control strategy through the DRL algorithm. In order to further optimize the following speed performance, this paper introduces the Lévy perturbation into the tradi-tional particle swarm optimization algorithm to accurately calculate the optimal weight combinations in the reward function, ensuring that the algorithm can flexibly cope with different driving scenarios. Experimental results show that the algorithm is robust in multiple complex traffic scenarios, providing a safer, more efficient and comfortable driving experience.
| [1] | Xu, Z., Li, X., Zhao, X., Zhang, M.H. and Wang, Z. (2017) DSRC versus 4G-LTE for Connected Vehicle Applications: A Study on Field Experiments of Vehicular Communication Performance. Journal of Advanced Transportation, 2017, 1-10. https://doi.org/10.1155/2017/2750452 |
| [2] | Gipps, P.G. (1981) A Behavioural Car-Following Model for Computer Simulation. Transportation Research Part B: Methodological, 15, 105-111. https://doi.org/10.1016/0191-2615(81)90037-0 |
| [3] | Treiber, M., Hennecke, A. and Helbing, D. (2000) Congested Traffic States in Empirical Observations and Microscopic Simulations. Physical Review E, 62, 1805-1824. https://doi.org/10.1103/physreve.62.1805 |
| [4] | Zhu, M., Wang, Y., Pu, Z., Hu, J., Wang, X. and Ke, R. (2020) Safe, Efficient, and Comfortable Velocity Control Based on Reinforcement Learning for Autonomous Driving. Transportation Research Part C: Emerging Technologies, 117, 102662. https://doi.org/10.1016/j.trc.2020.102662 |
| [5] | Gong, Y., Abdel-Aty, M., Yuan, J. and Cai, Q. (2020) Multi-objective Reinforcement Learning Approach for Improving Safety at Intersections with Adaptive Traffic Signal Control. Accident Analysis & Prevention, 144, 105655. https://doi.org/10.1016/j.aap.2020.105655 |
| [6] | Zhou, M., Yu, Y. and Qu, X. (2020) Development of an Efficient Driving Strategy for Connected and Automated Vehicles at Signalized Intersections: A Reinforcement Learning Approach. IEEE Transactions on Intelligent Transportation Systems, 21, 433-443. https://doi.org/10.1109/tits.2019.2942014 |
| [7] | U.S. Department of Transportation Federal Highway Administration (2016) Next Generation Simulation (NGSIM) Vehicle Trajectories and Supporting Data. |
| [8] | Pu, Z., Li, Z., Jiang, Y. and Wang, Y. (2021) Full Bayesian Before-After Analysis of Safety Effects of Variable Speed Limit System. IEEE Transactions on Intelligent Transportation Systems, 22, 964-976. https://doi.org/10.1109/tits.2019.2961699 |
| [9] | Zhang, G., Wang, Y., Wei, H. and Chen, Y. (2007) Examining Headway Distribution Models with Urban Freeway Loop Event Data. Transportation Research Record: Journal of the Transportation Research Board, 1999, 141-149. https://doi.org/10.3141/1999-15 |
