DeepGame-TP: Integrating Dynamic Game Theory and Deep Learning for Trajectory Planning

被引：0

作者：

Lucente, Giovanni ^{[1
,2
]}

Maarssoe, Mikkel Skov ^{[1
]}

Konthala, Sanath Himasekhar ^{[1
]}

Abulehia, Anas ^{[1
]}

Dariani, Reza ^{[1
]}

Schindler, Julian ^{[1
]}

机构：

[1] German Aerosp Ctr DLR, Inst Transportat Syst, D-38108 Braunschweig, Germany

[2] Tech Univ Berlin, Fak Verkehrs und Maschinensyst, TU Berlin, D-10623 Berlin, Germany

来源：

IEEE OPEN JOURNAL OF INTELLIGENT TRANSPORTATION SYSTEMS | 2024年 / 5卷

关键词：

Trajectory; Trajectory planning; Planning; Training; Deep learning; Safety; Real-time systems; Nash equilibrium; Imitation learning; Games; Dynamic game; deep learning; generalized Nash equilibrium; LSTM; trajectory planning; PREDICTION;

D O I：

10.1109/OJITS.2024.3515270

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Trajectory planning for automated vehicles in traffic has been a challenging task and a hot topic in recent research. The need for flexibility, transparency, interpretability and predictability poses challenges in deploying data-driven approaches in this safety-critical application. This paper proposes DeepGame-TP, a game-theoretical trajectory planner that uses deep learning to model each agent's cost function and adjust it based on observed behavior. In particular, a LSTM network predicts each agent's desired speed, forming a penalizing term that reflects aggressiveness in the cost function. Experiments demonstrated significant advantages of this innovative framework, highlighting the adaptability of DeepGame-TP in intersection, overtaking, car following and merging scenarios. It effectively avoids dangerous situations that could arise from incorrect cost function estimates. The approach is suitable for real-time applications, solving the Generalized Nash Equilibrium Problem (GNEP) in scenarios with up to four vehicles in under 100 milliseconds on average.

引用

页码：873 / 888

页数：16

共 41 条

[1] Rauker T., Ho A., Casper S., Hadfield-Menell D., Toward transparent ai: A survey on interpreting the inner structures of deep neural networks
[2] Next generation simulation (NGSIM) vehicle trajectories and supporting data, (2016)
[3] Betz J., Et al., Autonomous vehicles on the edge: A survey on autonomous vehicle racing, IEEE Open J. Intell. Transp. Syst., 3, pp. 458-488, (2022)
[4] Wang Z., Guo J., Hu Z., Zhang H., Zhang J., Pu J., Lane transformer: A high-efficiency trajectory prediction model, IEEE Open J. Intell. Transp. Syst., 4, pp. 2-13
[5] Trauth R., Moller K., Betz J., Toward safer autonomous vehicles: Occlusion-aware trajectory planning to minimize risky behavior, IEEE Open J. Intell. Transp. Syst., 4, pp. 929-942
[6] Tselentis D.I., Papadimitriou E., Driver profile and driving pattern recognition for road safety assessment: Main challenges and future directions, IEEE Open J. Intell. Transp. Syst., 4, pp. 83-100
[7] Vasile L., Dinkha N., Seitz B., Dasch C., Schramm D., Comfort and safety in conditional automated driving in dependence on personal driving behavior, IEEE Open J. Intell. Transp. Syst., 4, pp. 772-784
[8] Williams K.R., Et al., Trajectory planning with deep reinforcement learning in high-level action spaces, IEEE Trans. Aerosp. Electron. Syst., 59, 3, pp. 2513-2529, (2023)
[9] Zhang E., Zhang R., Masoud N., Predictive trajectory planning for autonomous vehicles at intersections using reinforcement learning, Transp. Res. Part C, Emerg. Technol., 149
[10] Wang Z., Tu J., Chen C., Reinforcement learning based trajectory planning for autonomous vehicles, Proc. China Autom. Congr. (CAC), 2021, pp. 7995-8000

← 1 2 3 4 5 →