A Flight-Fault-Aware Path Planning Strategy for VTOL Intelligent Air-Ground Vehicle Using Game Learning Approach

Jing Zhao, Chao Yang, Guosheng Liu, Weida Wang, Tianqi Qie, Changle Xiang, Hui Liu

科研成果: 期刊稿件文章同行评审

摘要

The VTOL intelligent air-ground vehicle can complete high mobility tasks in complex terrains by switching air-ground modes. During the tasks, path planning plays an important role in achieving the autonomous operation of the vehicle. The path planning process faces the following challenge. Under inevitable flight faults, reasonable mode switching decisions are required to obtain different short and energy-efficient multi-mode paths. To address this, a flight-fault-aware path planning strategy using game learning approach is proposed. Firstly, by constructing a two-layer game framework, Nash equilibrium solutions are solved for air-ground mode switching. The above framework includes the distance layer and energy consumption layer. Secondly, considering the multi-mode movement capability, a newly designed reward function expands the passable areas. Based on the above, an improved update rule constantly updates a new Q table to obtain a short and energy-efficient multi-mode path. Thirdly, under different faults, optimal flight conditions are solved to regain Nash equilibrium solutions for new multi-mode paths. Finally, the proposed strategy is verified in maps of different sizes. Under each fault, this strategy provides short and energy-efficient multi-mode paths of the evader and pursuer. The former focuses on shorter path distance, while the latter focuses on less energy consumption. They are selected for different task requirements.

源语言英语
页(从-至)1-27
页数27
期刊IEEE Transactions on Intelligent Vehicles
DOI
出版状态已接受/待刊 - 2024

指纹

探究 'A Flight-Fault-Aware Path Planning Strategy for VTOL Intelligent Air-Ground Vehicle Using Game Learning Approach' 的科研主题。它们共同构成独一无二的指纹。

引用此

Zhao, J., Yang, C., Liu, G., Wang, W., Qie, T., Xiang, C., & Liu, H. (已接受/印刷中). A Flight-Fault-Aware Path Planning Strategy for VTOL Intelligent Air-Ground Vehicle Using Game Learning Approach. IEEE Transactions on Intelligent Vehicles, 1-27. https://doi.org/10.1109/TIV.2024.3420744