Article Details

Abstract

As the crime landscape constantly changes, police officers need to be continuously trained to adapt to new challenges. In order to allow police officers to conduct vehicle pursuit training in a safe environment, make tactical rehearsal of the pursuit process, and save the training cost, this study proposes a design scheme of police vehicle pursuit training and tactical rehearsal system based on VR. This scheme uses Blender software to build scenarios, vehicles and other models, and imports them into the Unity platform for code writing to realize the corresponding functions. Through external equipment, police officers can control vehicles and experience the immersive training. In addition, it uses ML-Agents plugin combined with Reinforcement Learning algorithm to train intelligent agents, so as to complete the tactical rehearsal. The police vehicle pursuit training and tactical rehearsal system based on VR provides a new means of pursuit training for police officers, and provides a reference for optimizing the deployment of police force, which has significant practical value.

Keywords

pursuit training; tactical rehearsal; police; Multi-Agent; Reinforcement Learning

References

1. [1] Shi Bo, Wang Fuzhu Simulation of Virtual Dynamic Scene Construction for Police Practical Training [J]. Computer Simulation, 2023, 40 (3): 1-4+134
2. [2] Shangpu Artificial Intelligence and Criminal Investigation: Historical Changes, Technological Classification, and Future Prospects [J]. Journal of People's Public Security University of China: Social Sciences Edition, 2020, 36 (6): 46-54
3. [3] Wan Yuhang, Zhu Zilu, Zhong Chunfu, etc Dynamic path planning of robotic arm based on improved PPO algorithm [J/OL]. Journal of System Simulation, 2024:1-14 (2024-05-21) https://doi.org/10.16182/j.issn1004731x.joss.24-0122.
4. [4] Liu Chunhui Research on Multi Agent Collaborative Encirclement Method Based on Deep Reinforcement Learning [D]. Dalian: Dalian Maritime University, 2023
5. [5] Zhang Zewei Research on Multi Agent Attack and Defense Strategies Based on Deep Reinforcement Learning [D]. Chongqing: Chongqing University of Technology, 2023
6. [6] Li Jihao Research on Virtual Assembly Positioning Constraints and Intelligent Path Planning [D]. Xi'an: Xi'an University of Technology, 2023
7. [7] Peng Di, Chen Huangzi Design and Teaching Implementation of a Virtual System for Counter Terrorism Tactical Training for Practical Use [J]. Laboratory Research and Exploration, 2023, 42 (4): 205-209+231
8. [8] Lei Xiaoli, Sun Liang, Zhao Lei Design of a Visual Radar Simulation Software Based on Unity3D [J]. Journal of Aerospace Early Warning Research, 2024, 38 (2): 132-137
9. [9] Xu Jingqing, Cui Ping, Wen Jian, etc Design of a Virtual Training System for Terminal Guided Missile Shooting Preparation Based on 3DSMAX and Unity3D [J]. Journal of Weapon Equipment Engineering, 2021, 42 (7): 30-34
10. [10] Zhang Xiangrui, Tan Tai, Li Hui, etc Unmanned Aerial Vehicle Air Combat Maneuvering Decision Method Based on Deep Reinforcement Learning [J/OL]. Computer Engineering, 2024:1-15 (2024-08-23) https://doi.org/10.19678/j.issn.1000-3428.0069621.
11. [11] LAI J，CHEN X，ZHANG X. Training an Agent for Third-person Shooter Game Using Unity ML-Agents [C]//2019 International Conference on Artificial Intelligence and Computing Science.Chengdu：ICAICS，2019：6.
12. [12] Kuang Liqun, Li Siyuan, Feng Li, etc Application of Deep Reinforcement Learning Algorithm in Intelligent Military Decision Making [J]. Computer Engineering and Applications, 2021, 57 (20): 271-278
13. [13] CAI H，LUO Y G，GAO H L，et al. A Multiphase Semistatic Training Method for Swarm Confrontation Using Multiagent Deep Reinforcement Learning [J/OL ].Computational Intelligence and Neuroscience，2023：2955442（2023-07-07）.https://doi.org/10.1155/2023/2955442.

Funding

Technical Research Program of the Ministry of Public Security (2024JSYJC06)

About the authors

You Han (2002-), female, Han ethnicity, from Baoding, Hebei Province, currently pursuing a master's degree in police simulation; Li Zhihui (1981-), female, Han ethnicity, from Huangshi, Hubei, associate professor, PhD, research direction: police simulation.