Imagine-2-Drive: Leveraging High-Fidelity World Models via Multi-Modal Diffusion Policies

TL;DR

Imagine-2-Drive introduces a high-fidelity world model combined with a multi-modal diffusion policy for autonomous driving, significantly enhancing policy robustness and performance with minimal online interactions.

Contribution

The paper presents a novel framework integrating a diffusion-based world model with a multi-modal diffusion policy for improved autonomous driving.

Findings

Outperforms prior world model baselines in CARLA benchmarks.

Improves Route Completion by 15% and Success Rate by 20%.

Mitigates error accumulation with a diffusion-based approach.

Abstract

World Model-based Reinforcement Learning (WMRL) enables sample efficient policy learning by reducing the need for online interactions which can potentially be costly and unsafe, especially for autonomous driving. However, existing world models often suffer from low prediction fidelity and compounding one-step errors, leading to policy degradation over long horizons. Additionally, traditional RL policies, often deterministic or single Gaussian-based, fail to capture the multi-modal nature of decision-making in complex driving scenarios. To address these challenges, we propose Imagine-2-Drive, a novel WMRL framework that integrates a high-fidelity world model with a multi-modal diffusion-based policy actor. It consists of two key components: DiffDreamer, a diffusion-based world model that generates future observations simultaneously, mitigating error accumulation, and DPA (Diffusion Policy Actor), a diffusion-based policy that models diverse and multi-modal trajectory distributions. By training DPA within DiffDreamer, our method enables robust policy learning with minimal online interactions. We evaluate our method in CARLA using standard driving benchmarks and demonstrate that it outperforms prior world model baselines, improving Route Completion and Success Rate by 15% and 20% respectively.