Occ-LLM: Enhancing Autonomous Driving with Occupancy-Based Large Language Models

TL;DR

This paper introduces Occ-LLM, a novel occupancy-based large language model for autonomous driving, which improves scene understanding and planning by effectively encoding occupancy data and separating dynamic from static elements.

Contribution

The study pioneers the integration of occupancy representations with LLMs and proposes MS-VAE for better dynamic-static scene separation, enhancing autonomous driving tasks.

Findings

Achieved 6% IoU improvement in 4D occupancy forecasting.

Surpassed state-of-the-art methods in scene question answering.

Enhanced dynamic trajectory prediction accuracy.

Abstract

Large Language Models (LLMs) have made substantial advancements in the field of robotic and autonomous driving. This study presents the first Occupancy-based Large Language Model (Occ-LLM), which represents a pioneering effort to integrate LLMs with an important representation. To effectively encode occupancy as input for the LLM and address the category imbalances associated with occupancy, we propose Motion Separation Variational Autoencoder (MS-VAE). This innovative approach utilizes prior knowledge to distinguish dynamic objects from static scenes before inputting them into a tailored Variational Autoencoder (VAE). This separation enhances the model's capacity to concentrate on dynamic trajectories while effectively reconstructing static scenes. The efficacy of Occ-LLM has been validated across key tasks, including 4D occupancy forecasting, self-ego planning, and occupancy-based scene question answering. Comprehensive evaluations demonstrate that Occ-LLM significantly surpasses existing state-of-the-art methodologies, achieving gains of about 6\% in Intersection over Union (IoU) and 4\% in mean Intersection over Union (mIoU) for the task of 4D occupancy forecasting. These findings highlight the transformative potential of Occ-LLM in reshaping current paradigms within robotic and autonomous driving.