Custom-built Simulation Environment
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/raaslab/rl_multi_agent
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是在一个专门构建的环境中用于模拟基于可视性的持续监控（VPM）问题。训练和测试是在一台配置有Ubuntu 16.04 LTS操作系统、Intel Xeon Silver CPU和GeForce RTX 2080 Ti GPU的计算机上进行的。每个训练剧集包含1000个步骤，而测试剧集包含2000个步骤。该数据集的规模为50x50的地图，具有25的视野范围。其任务是评估多智能体强化学习方法在解决VPM问题上的表现。
This dataset is designed to simulate the visibility-based persistent monitoring (VPM) problem in a specially constructed environment. Training and testing were conducted on a computer equipped with Ubuntu 16.04 LTS operating system, Intel Xeon Silver CPU, and GeForce RTX 2080 Ti GPU. Each training episode consists of 1000 steps, while each test episode contains 2000 steps. This dataset utilizes a 50×50 map with a field of view of 25. Its core task is to evaluate the performance of multi-agent reinforcement learning methods when solving the VPM problem.
提供机构：
Authors of the paper
