How Can LLMs and Knowledge Graphs Contribute to Robot Safety? A Few-Shot Learning Approach
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912158475878400 |
|---|---|
| author | Althobaiti, Abdulrahman Ayala, Angel Gao, JingYing Almutairi, Ali Deghat, Mohammad Razzak, Imran Cruz, Francisco |
| author_facet | Althobaiti, Abdulrahman Ayala, Angel Gao, JingYing Almutairi, Ali Deghat, Mohammad Razzak, Imran Cruz, Francisco |
| contents | Large Language Models (LLMs) are transforming the robotics domain by enabling robots to comprehend and execute natural language instructions. The cornerstone benefits of LLM include processing textual data from technical manuals, instructions, academic papers, and user queries based on the knowledge provided. However, deploying LLM-generated code in robotic systems without safety verification poses significant risks. This paper outlines a safety layer that verifies the code generated by ChatGPT before executing it to control a drone in a simulated environment. The safety layer consists of a fine-tuned GPT-4o model using Few-Shot learning, supported by knowledge graph prompting (KGP). Our approach improves the safety and compliance of robotic actions, ensuring that they adhere to the regulations of drone operations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_11387 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | How Can LLMs and Knowledge Graphs Contribute to Robot Safety? A Few-Shot Learning Approach Althobaiti, Abdulrahman Ayala, Angel Gao, JingYing Almutairi, Ali Deghat, Mohammad Razzak, Imran Cruz, Francisco Robotics Artificial Intelligence Large Language Models (LLMs) are transforming the robotics domain by enabling robots to comprehend and execute natural language instructions. The cornerstone benefits of LLM include processing textual data from technical manuals, instructions, academic papers, and user queries based on the knowledge provided. However, deploying LLM-generated code in robotic systems without safety verification poses significant risks. This paper outlines a safety layer that verifies the code generated by ChatGPT before executing it to control a drone in a simulated environment. The safety layer consists of a fine-tuned GPT-4o model using Few-Shot learning, supported by knowledge graph prompting (KGP). Our approach improves the safety and compliance of robotic actions, ensuring that they adhere to the regulations of drone operations. |
| title | How Can LLMs and Knowledge Graphs Contribute to Robot Safety? A Few-Shot Learning Approach |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2412.11387 |