Simulation and Real Robot Experiment Data

该数据集包含了从模拟实验和真实机器人实验中收集的数据，用于评估SPINE规划器相较于其他竞争性基准方案的性能表现。数据涵盖了完成任务的各项指标，如所需时间、行进距离、LLM API调用次数以及必要的用户交互。在模拟实验中，数据集涵盖了超过40,000平方米的面积，并在真实世界环境中包含了不同规模的任务。该数据集的任务是评估在不完全规格条件下，SPINE完成任务效率与显式任务指派及地图绘制后规划方法相比的优劣。

This dataset contains data collected from both simulation experiments and real-world robotic experiments, which is used to evaluate the performance of the SPINE planner against other competitive baseline methods. The data covers various metrics for task completion, including elapsed time, travel distance, the number of LLM API calls, and necessary user interactions. For simulation experiments, the dataset covers an area of over 40,000 square meters, and it includes tasks of varying scales in real-world environments. The purpose of this dataset is to assess the performance advantages of SPINE in task completion efficiency compared to methods that rely on explicit task assignment and post-mapping planning under incomplete task specification conditions.