S2Planner：用于端到端自动驾驶的多尺度语义规划器

中文摘要 AI 辅助

我们提出了S2Planner，一种轨迹规划器，它结合了三个前视摄像头、自车运动历史和当前驾驶指令。微调的DINOv3骨干网络和空间调适适配器生成多尺度图像特征；然后，一个从粗到细的解码器利用轨迹自注意力和相机投影的交叉注意力来细化候选路径点。其贡献在于将基于自车条件的轨迹初始化与迭代的、几何引导的多尺度图像特征采样相结合，而非提出新的视觉骨干网络或注意力算子。在NAVSIM v1非反应式评估中，先前报告的navtest运行获得了88.03 PDMS。由于该运行是基于navtest性能选择的，该数字具有探索性，不能解释为无偏的测试估计。需要在未曝光数据上进行验证集选择的评估、重复运行以及计算测量，以确定泛化性和效率。

英文摘要

We present S2Planner, a trajectory planner that combines three front-facing cameras with ego-motion history and the current driving command. A fine-tuned DINOv3 backbone and a Spatial Tuning Adapter produce multi-scale image features; a coarse-to-fine decoder then uses trajectory self-attention and camera-projected cross-attention to refine candidate waypoints. The contribution is the integration of ego-conditioned trajectory initialization with iterative, geometry-guided sampling of multi-scale image features, rather than a new visual backbone or attention operator. On the NAVSIM v1 non-reactive evaluation, the previously reported navtest run obtained 88.03 PDMS. Because that run was selected using navtest performance, this number is exploratory and cannot be interpreted as an unbiased test estimate. Validation-selected evaluation on unexposed data, repeated runs, and computational measurements are needed to establish generalization and efficiency.

发表机构

• Technical University of Munich（慕尼黑工业大学）
• Huaibei Normal University（淮北师范大学）

机构由 AI 辅助整理，请以论文原文为准。