PyReason Simulation Environment

该数据集提供了一个模拟环境，旨在通过模拟非马尔可夫动态来促进强化学习实验，并允许智能体独立执行移动和射击等动作。该环境支持包括基础和高级场景在内的不同配置，具有多样的动作和观察空间，并旨在与现有的模拟器如AFSIM和SC2进行比较。该数据集涵盖了单智能体和多智能体场景，任务重点是强化学习智能体的训练与评估。

This dataset provides a simulated environment designed to facilitate reinforcement learning experiments by simulating non-Markovian dynamics, allowing agents to independently perform actions such as moving and shooting. The environment supports diverse configurations including basic and advanced scenarios, features varied action and observation spaces, and is intended for benchmarking against existing simulators such as AFSIM and SC2. This dataset covers single-agent and multi-agent scenarios, with a primary focus on the training and evaluation of reinforcement learning agents.