让不同机器人的导航规划适应自身物理限制，提升真实场景中的成功率。
CrossTracer: Cross-Embodiment Navigation via VLA Model Reasoning and Trace Residuals Adapting
- 用像素空间的路径点统一语义推理与物理约束，实现跨机器人导航。
- 通过残差修正模块，根据机器人类型和可通行性自动优化路径，准确率提升28.1%。
- 无需人工标注，利用分割图自动生成训练数据，适合实际部署在轮式与足式机器人上。
视觉-语言-动作（VLA）模型为机器人导航提供了强大的语义先验，但常忽略具体实体的移动限制。一个对某机器人语义合理的路径，对另一机器人可能在物理上不可行。本文提出CrossTracer，一种基于自适应轨迹残差的分层跨实体导航框架。CrossTracer将导航计划表示为归一化的图像平面路径点，形成语义推理与物理落地之间的统一像素空间接口。首先，视觉-语言轨迹生成器（VL-Tracer）适配预训练VLA模型，从自我中心观测和灵活的目标描述中预测初始导航轨迹；其次，CE-Adapter通过预测与机器人身份相关的残差修正量，结合视觉可通行性线索和初始轨迹进行轨迹优化。为避免昂贵的人工标注，本文设计了跨实体RRT*（CE-RRT*），将全景分割转换为机器人条件下的可通行性代价图，并生成最小代价的像素空间轨迹。在NaviTrace基准测试中，CrossTracer总得分为45.68，优于最强通用基线Gemini-2.5-Pro的35.67，相对提升28.1%。真实世界部署于轮式与足式机器人进一步验证了其导航成功率与执行效率的显著提升。
原文摘要 · Abstract (English)
Vision-language-action (VLA) models provide strong semantic priors for robot navigation, but they often ignore embodiment-specific mobility constraints. A path that is semantically plausible for one robot may be physically infeasible for another. We propose CrossTracer, a hierarchical framework for cross-embodiment navigation through adaptive trace residuals. CrossTracer represents navigation plans as normalized image-plane waypoints, forming a unified pixel-space interface between semantic reasoning and physical grounding. First, Vision-Language Trace Proposer (VL-Tracer) adapts a pretrained VLA model to predict an initial navigation trace from egocentric observations and flexible goal specifications. Second, CE-Adapter refines this trace by predicting embodiment-conditioned residual corrections from visual traversability cues, robot identity, and the initial trace. To train the refinement module without costly manual annotation, Cross-Embodiment RRT* (CE-RRT*) converts panoptic segmentation into robot-conditioned traversability cost maps and generates cost-minimizing pixel-space traces. We evaluate CrossTracer on the NaviTrace benchmark, which tests whether a model can generate embodiment-consistent navigation traces from egocentric observations, language instructions, and robot embodiment types. CrossTracer achieves a total score of 45.68, outperforming the strongest evaluated general-purpose baseline, Gemini-2.5-Pro, by 10.01 points, corresponding to a 28.1% relative improvement. Real-world deployment on wheeled and legged robots further shows improved navigation success and execution efficiency.
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。