Large Language Models Powered Context-aware Motion Prediction

arXiv:2403.11057v1 Announce Type: new
Abstract: Motion prediction is among the most fundamental tasks in autonomous driving. Traditional methods of motion forecasting primarily encode vector information of maps and historical trajectory data of traffic participants, lacking a comprehensive understanding of overall traffic semantics, which in turn affects the performance of prediction tasks. In this paper, we utilized Large Language Models (LLMs) to enhance the global traffic context understanding for motion prediction tasks. We first conducted systematic prompt engineering, visualizing complex traffic …

abstract arxiv autonomous autonomous driving context cs.cv cs.ro data driving encode forecasting information language language models large language large language models maps performance prediction semantics tasks traffic trajectory type understanding vector