LA-RCS User Requests Benchmark

该数据集是一个包含20个用户请求的基准测试，覆盖了四个领域：目标检测、命令执行、障碍物导航和情境感知。它旨在评估LA-RCS系统的性能。此外，该基准测试还包含了四个领域内用户请求的详细成功与失败率，并针对不同代理配置（GPT-4-Turbo和GPT-4o）的性能指标进行了细分。该数据集规模涉及4个领域的20个用户请求，任务重点是评估机器人控制中用户请求的成功与失败率。

This dataset is a benchmark comprising 20 user requests covering four domains: object detection, command execution, obstacle navigation, and situational awareness. It aims to evaluate the performance of the LA-RCS system. Additionally, this benchmark includes detailed success and failure rates of user requests within each of the four domains, with performance metrics subdivided for different agent configurations such as GPT-4-Turbo and GPT-4o. With 20 user requests across four domains, this dataset focuses on assessing the success and failure rates of user requests in robot control.