Efficient Driving Behavior Narration and Reasoning on Edge Device Using Large Language Models

TL;DR

This paper presents a framework deploying large language models on edge devices for autonomous driving, enabling efficient scene narration and reasoning with improved performance and response speed.

Contribution

It introduces a novel edge-based LLM deployment framework with a multi-modal prompt strategy for driving behavior narration and reasoning.

Findings

LLMs on edge devices achieve satisfactory response speeds

The multi-modal prompt strategy enhances narration and reasoning performance

Experiments on OpenDV-Youtube dataset show significant performance improvements

Abstract

Deep learning architectures with powerful reasoning capabilities have driven significant advancements in autonomous driving technology. Large language models (LLMs) applied in this field can describe driving scenes and behaviors with a level of accuracy similar to human perception, particularly in visual tasks. Meanwhile, the rapid development of edge computing, with its advantage of proximity to data sources, has made edge devices increasingly important in autonomous driving. Edge devices process data locally, reducing transmission delays and bandwidth usage, and achieving faster response times. In this work, we propose a driving behavior narration and reasoning framework that applies LLMs to edge devices. The framework consists of multiple roadside units, with LLMs deployed on each unit. These roadside units collect road data and communicate via 5G NSR/NR networks. Our experiments show that LLMs deployed on edge devices can achieve satisfactory response speeds. Additionally, we propose a prompt strategy to enhance the narration and reasoning performance of the system. This strategy integrates multi-modal information, including environmental, agent, and motion data. Experiments conducted on the OpenDV-Youtube dataset demonstrate that our approach significantly improves performance across both tasks.