Journal
ENGINEERING COMPUTATIONS
Volume 39, Issue 3, Pages 993-1019Publisher
EMERALD GROUP PUBLISHING LTD
DOI: 10.1108/EC-09-2020-0500
Keywords
Reinforcement learning; Path planning; Intelligent driving; Shortest path algorithm
Categories
Funding
- National Natural Science Foundation of China [61571328]
- Tianjin Key Natural Science Foundation Project [18JCZDJC96800]
- Tianjin Science and Technology Major Project [15ZXDSGX 00050]
- Tianjin Science and Technology Innovation Team Fund Projects [TD12-5016, TD13-5025, TD2015-23]
- Tianjin Science and Technology Service Industry Major Science and Technology Project [16ZXFWGX00010, 17YFZCGX00360]
Ask authors/readers for more resources
The hybrid path planning algorithm in this paper combines optimized reinforcement learning and improved particle swarm optimization to achieve efficient path planning results. By optimizing RL hyperparameters, designing a pre-set operation for PSO, and proposing a correction variable, the algorithm selects the optimal path effectively.
Purpose To solve the path planning problem of the intelligent driving vehicular, this paper designs a hybrid path planning algorithm based on optimized reinforcement learning (RL) and improved particle swarm optimization (PSO). Design/methodology/approach First, the authors optimized the hyper-parameters of RL to make it converge quickly and learn more efficiently. Then the authors designed a pre-set operation for PSO to reduce the calculation of invalid particles. Finally, the authors proposed a correction variable that can be obtained from the cumulative reward of RL; this revises the fitness of the individual optimal particle and global optimal position of PSO to achieve an efficient path planning result. The authors also designed a selection parameter system to help to select the optimal path. Findings Simulation analysis and experimental test results proved that the proposed algorithm has advantages in terms of practicability and efficiency. This research also foreshadows the research prospects of RL in path planning, which is also the authors' next research direction. Originality/value The authors designed a pre-set operation to reduce the participation of invalid particles in the calculation in PSO. And then, the authors designed a method to optimize hyper-parameters to improve learning efficiency of RL. And then they used RL trained PSO to plan path. The authors also proposed an optimal path evaluation system. This research also foreshadows the research prospects of RL in path planning, which is also the authors' next research direction.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available