MOJITO: Modal Joint Learning for Unified End-to-End Autonomous Driving
End-to-end autonomous driving systems commonly follow a cascaded two-stage pipeline where a perception stage compresses multi-modal sensor inputs into a compact context and a downstream planner predicts trajectories conditioned on this context. We argue that this one-way perception-to-planning interface forces sensor inputs into a compact representation, losing the fine-grained details critical for planning. Moreover, by constraining the planner to this compressed context, it is difficult to leverage the rich representations offered by modern vision foundation models. To address these issues, we propose MOJITO, a unified sensor-to-action framework for end-to-end autonomous driving built on modal joint learning. MOJITO removes the cascaded interface and instead performs block-wise Modal Joint Attention that simultaneously updates action, image, and LiDAR features, allowing the planner to directly access multi-modal features during action generation. MOJITO achieves 88.9 PDMS on the NAVSIM v1 dataset and 88.4 EPDMS on the more challenging NAVSIM v2 dataset, setting a new state-of-the-art. Extensive experiments further demonstrate strong scalability, instruction following, and diverse trajectory generation. Code and models are available at https://github.com/mumucc01/MOJITO.
Code (0)
등록된 구현이 없습니다.
Tasks
Instruction FollowingAutonomous DrivingSimilar Papers 제목 키워드 기반
Toward Unified Multimodal Representation Learning for Autonomous Driving
Contrastive Language-Image Pre-training (CLIP) has shown impressive performance in aligning visual and textual representations. Recent studies have extended this paradigm to 3D vision to improve scene understanding for a…
Representation LearningContrastive LearningScene UnderstandingAutonomous DrivingMaskFuser: Masked Fusion of Joint Multi-Modal Tokenization for End-to-End Autonomous Driving
Current multi-modality driving frameworks normally fuse representation by utilizing attention between single-modality branches. However, the existing networks still suppress the driving performance as the Image and LiDAR…
Autonomous DrivingA Collaborative Multi-Modality Interaction for VLA-based End-to-End Autonomous Driving
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for end-to-end autonomous driving by jointly integrating perception, reasoning, and decision making within a unified multimodal framework. However, …
Visual Question AnsweringTrajectory PlanningAutonomous DrivingDecision MakingUniDriveDreamer: A Single-Stage Multimodal World Model for Autonomous Driving
World models have demonstrated significant promise for data synthesis in autonomous driving. However, existing methods predominantly concentrate on single-modality generation, typically focusing on either multi-camera vi…
Autonomous DrivingOmniGen: Unified Multimodal Sensor Generation for Autonomous Driving
Autonomous driving has seen remarkable advancements, largely driven by extensive real-world data collection. However, acquiring diverse and corner-case data remains costly and inefficient. Generative models have emerged …
Autonomous Driving