Robotic Limb Target-Finding Task

该数据集来源于一系列实验，在这些实验中，一个机器人肢体通过两种强化学习算法（PPO和AC）在与环境互动的过程中，通过伺服电机调整来定位目标。该数据集包含了近端策略优化（PPO）和离散行动者-评论家（AC）算法的训练数据，具体包括状态、行动、奖励结构以及实验结果等信息。实验规模方面，PPO算法进行了10次重复试验，而AC算法则进行了30次重复试验。任务内容是使用强化学习算法执行目标寻找任务。

This dataset is derived from a series of experiments in which a robotic limb uses two reinforcement learning algorithms, Proximal Policy Optimization (PPO) and discrete Actor-Critic (AC), to locate targets by adjusting servo motors during interactions with the environment. It contains training data generated by the PPO and discrete Actor-Critic algorithms, specifically including states, actions, reward structures, and experimental results. Regarding experimental scale, the PPO algorithm was conducted in 10 replicated trials, while the discrete Actor-Critic algorithm underwent 30 replicated trials. The core task of these experiments is a target-seeking task implemented using reinforcement learning algorithms.