Task Allocation for Heterogeneous Robots in Office Environments Using Large Language Models

In recent years, office assistant robots have become increasingly common. People are more accustomed to issuing commands using natural language, and these commands are often complex and diverse. To address the limitations of traditional algorithms, such as poor natural language understanding, modeling difficulties, and lack of robustness, we propose a general framework for heterogeneous robot task allocation based on large language models. Additionally, we have designed a dataset of natural language instructions tailored for office scenarios. Finally, we validated the effectiveness of this framework using the dataset.