Enhancing Autonomous Driving Safety with Collision Scenario Integration
This work addresses safety challenges in autonomous driving by providing a scalable solution to leverage collision data, though it is incremental as it builds on existing planning methods.
The paper tackled the problem of autonomous vehicle safety by proposing SafeFusion, a training framework that integrates collision data and safety metrics to improve planning, resulting in a 56% performance improvement in collision-prone scenarios over previous state-of-the-art planners.
Autonomous vehicle safety is crucial for the successful deployment of self-driving cars. However, most existing planning methods rely heavily on imitation learning, which limits their ability to leverage collision data effectively. Moreover, collecting collision or near-collision data is inherently challenging, as it involves risks and raises ethical and practical concerns. In this paper, we propose SafeFusion, a training framework to learn from collision data. Instead of over-relying on imitation learning, SafeFusion integrates safety-oriented metrics during training to enable collision avoidance learning. In addition, to address the scarcity of collision data, we propose CollisionGen, a scalable data generation pipeline to generate diverse, high-quality scenarios using natural language prompts, generative models, and rule-based filtering. Experimental results show that our approach improves planning performance in collision-prone scenarios by 56\% over previous state-of-the-art planners while maintaining effectiveness in regular driving situations. Our work provides a scalable and effective solution for advancing the safety of autonomous driving systems.