NavGPT: Leveraging LLM for Route Classification Integrating Pollution and Crime Analytics

Large language models (LLMs) such as OpenAI's ChatGPT, Meta's LLaMA, and Google's Gemini have significantly transformed the field of natural language processing (NLP). These advancements have moved NLP from rule-based systems to models that can understand context and generate coherent narratives. One promising application for LLMs is in enhancing navigation systems. Traditional navigation systems primarily focus on optimizing routes based on distance and time, often overlooking factors such as air quality and crime rates, which directly impact user well-being. This paper introduces NavGPT, an innovative navigation framework that integrates real-time environmental and safety data into route planning. NavGPT fine-tunes LLMs to classify routes based on quantitative metrics like the Air Quality Index (AQI) and Crime Severity Index (CSI), allowing users to make informed travel decisions. By combining geospatial data with real-time inputs, NavGPT provides color-coded route classifications that prioritize safety and health considerations. This approach offers a more comprehensive solution compared to traditional navigation systems. The study demonstrates how LLMs, when fine-tuned for specialized tasks, can significantly enhance urban mobility systems. NavGPT creates a platform that not only optimizes travel efficiency but also promotes healthier and safer routes.