仅用摄像头实现端到端自动驾驶规划，效率更高。
PRIX: Learning to Plan from Raw Pixels for End-to-End Autonomous Driving
- 直接从原始图像预测安全轨迹，无需激光雷达和BEV表示
- 在NavSim和nuScenes上达到顶尖性能，推理速度更快、模型更小
- 适合车载摄像头系统，对量产车部署更友好
尽管端到端自动驾驶模型表现优异，但其实际部署常受限于大模型尺寸、对昂贵激光雷达的依赖以及计算密集的鸟瞰图（BEV）特征表示，限制了在仅配备摄像头的大众车型上的可扩展性。为解决这一问题，我们提出PRIX（从原始像素中规划）。该新型高效架构仅使用摄像头数据，不依赖显式的BEV表示，也无需激光雷达。PRIX通过视觉特征提取器与生成式规划头，直接从原始像素输入预测安全轨迹。核心组件是上下文感知重校准变压器（CaRT），用于增强多层级视觉特征以提升规划鲁棒性。实验表明，PRIX在NavSim和nuScenes基准上达到当前最优性能，媲美更大规模的多模态扩散规划模型，同时在推理速度和模型大小上显著更优，具备实际部署潜力。代码已开源，地址：https://maxiuw.github.io/prix。
原文摘要 · Abstract (English)
While end-to-end autonomous driving models show promising results, their practical deployment is often hindered by large model sizes, a reliance on expensive LiDAR sensors and computationally intensive BEV feature representations. This limits their scalability, especially for mass-market vehicles equipped only with cameras. To address these challenges, we propose PRIX (Plan from Raw Pixels). Our novel and efficient end-to-end driving architecture operates using only camera data, without explicit BEV representation and forgoing the need for LiDAR. PRIX leverages a visual feature extractor coupled with a generative planning head to predict safe trajectories from raw pixel inputs directly. A core component of our architecture is the Context-aware Recalibration Transformer (CaRT), a novel module designed to effectively enhance multi-level visual features for more robust planning. We demonstrate through comprehensive experiments that PRIX achieves state-of-the-art performance on the NavSim and nuScenes benchmarks, matching the capabilities of larger, multimodal diffusion planners while being significantly more efficient in terms of inference speed and model size, making it a practical solution for real-world deployment. Our work is open-source and the code will be at https://maxiuw.github.io/prix.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。