LoGoPlanner: Localization Grounded Navigation Policy with Metric-aware Visual Geometry

TL;DR

The paper tackles navigation in unstructured environments by removing reliance on explicit localization modules. It introduces LoGoPlanner, which ground-truths self-state through long-horizon visual geometry learned with depth priors, reconstructs dense scene geometry, and conditions a diffusion-based planner on implicit state and geometry queries. The approach achieves strong performance and cross-embodiment generalization in both simulation and real-world tests, with substantial improvements over oracle-localization baselines. By publicly releasing code and models, the work highlights the practical impact of metric-grounded, end-to-end navigation for diverse robotic platforms.

Abstract

Trajectory planning in unstructured environments is a fundamental and challenging capability for mobile robots. Traditional modular pipelines suffer from latency and cascading errors across perception, localization, mapping, and planning modules. Recent end-to-end learning methods map raw visual observations directly to control signals or trajectories, promising greater performance and efficiency in open-world settings. However, most prior end-to-end approaches still rely on separate localization modules that depend on accurate sensor extrinsic calibration for self-state estimation, thereby limiting generalization across embodiments and environments. We introduce LoGoPlanner, a localization-grounded, end-to-end navigation framework that addresses these limitations by: (1) finetuning a long-horizon visual-geometry backbone to ground predictions with absolute metric scale, thereby providing implicit state estimation for accurate localization; (2) reconstructing surrounding scene geometry from historical observations to supply dense, fine-grained environmental awareness for reliable obstacle avoidance; and (3) conditioning the policy on implicit geometry bootstrapped by the aforementioned auxiliary tasks, thereby reducing error propagation.We evaluate LoGoPlanner in both simulation and real-world settings, where its fully end-to-end design reduces cumulative error while metric-aware geometry memory enhances planning consistency and obstacle avoidance, leading to more than a 27.3\% improvement over oracle-localization baselines and strong generalization across embodiments and environments. The code and models have been made publicly available on the \href{https://steinate.github.io/logoplanner.github.io/}{project page}.

Figures (5)

• Figure 1: (a) Traditional modular planners decompose tasks into modules, introducing cascading errors. (b) Existing end-to-end frameworks directly map observations to control signals but still rely on explicit localization modules. (c) LoGoPlanner integrates implicit state estimation and metric-aware geometry perception into policy for fully end-to-end planning.
• Figure 2: Architecture overview. Our method injects scale priors into the image patches that are encoded by ViT oquab2023dinov2, and finetunes the video geometry model to metric scale prediction. We adopt a query-based design in which ego state representation and environment geometry are implicitly aggregated through task-specific queries. A diffusion policy head is detached to generate feasible and collision-free trajectories.
• Figure 3: Home scenes are characterized by narrow passages and cluttered semantic layouts, while commercial scenes cover representative categories such as hospitals, supermarkets, restaurants, schools, libraries, and offices.
• Figure 4: Visualization of LoGoPlanner in real-world scenarios on different robot platforms. The green curves are the planned trajectories of LoGoPlanner. Blue and grey clouds are the obstacles of the current frame and the previous frame respectively.
• Figure 5: Visualization of reconstruction results: the first row shows the scene point cloud of the ground truth, and the second row shows the predicted scene point cloud. The point cloud at the metric scale is predicted with the chassis of the last frame as the coordinate origin.