Training data from simulation and physical experiments

<b>Description:</b><br>This dataset includes the training data and trained models used in our study on autonomous microrobot control via model-based reinforcement learning (MBRL). The data encompasses both simulation and physical experiment results used to train and evaluate an ultrasound-actuated microrobot navigating microfluidic environments.Specifically, the repository contains:<b>Training logs and reward trajectories</b> for MBRL agents across six environments: Racetrack, SPA, Multi-Output Tributary, Vascular, Mixed, and Randomized channels.<b>Trained model checkpoints</b> for each of these environments.<b>Fine-tuning results</b> from real-world experiments, including target success rates

<b>数据集描述：</b><br>本数据集收录了我们开展基于模型强化学习（Model-Based Reinforcement Learning，MBRL）的自主微型机器人控制研究时所用的训练数据与已训练模型。该数据集涵盖用于训练与评估在微流控环境中导航的超声驱动微型机器人的仿真与实体实验结果。具体而言，本数据集仓库包含以下内容：<b>训练日志与奖励轨迹</b>，对应六种环境下的基于模型强化学习智能体，涵盖赛道（Racetrack）、SPA、多输出支流（Multi-Output Tributary）、血管（Vascular）、混合场景（Mixed）以及随机通道（Randomized channels）；<b>各对应环境的已训练模型检查点</b>；<b>实体实验微调结果</b>，涵盖目标成功率。