Binary Tracking for Spatial QA and Navigation with Open Vision-Language Models

2026-06-15

AI Summary

This work addresses the challenges of network instability, high latency, and substantial deployment costs faced by service robots relying on closed-source models for spatial question answering during long-range egocentric navigation. To overcome these limitations, we propose BinTrack—a fully open-source spatial localization agent that integrates an open-source vision-language model with trajectory temporal modeling. BinTrack innovatively introduces a binary search mechanism between two recognized anchor points and incorporates retrieval-augmented strategies to significantly enhance both inference efficiency and localization accuracy. Experimental results demonstrate that BinTrack achieves up to a 22.8% absolute improvement in accuracy on the global category of the SpaceLocQA benchmark and accelerates inference by over 1.5×, matching the performance of GPT-4o. Additionally, we release GangnamLoop, a real-world outdoor multi-trajectory dataset, to facilitate edge-deployable spatial reasoning research.

Abstract

This work addresses spatial question answering for service robots traversing long egocentric routes. Given a query such as "where can I find a dry cleaner on the way back home?", the system returns a metric coordinate that downstream navigation components can act on. Prior Spatial Question Answering approaches leverage retrieval-augmented agents built on closed-source models such as GPT-4o for path exploration. However, robots operating in the real world often cannot reliably depend on online closed-source models due to network instability, communication latency, and deployment cost. It creates a need for open-source based Spatial Question Answering approaches that can run onboard the robot, yet prior research in this direction remains limited. This work proposes BinTrack, a simple yet effective, fully open-source spatial-localization agent that leverages the temporal ordering of a robot's trajectory. BinTrack performs a binary search over the trajectory segments between two anchor landmarks identified from a query. It improves overall accuracy by up to 22.8% over other open-source implementations and even matches the reported closed-source model result on the global category of the SpaceLocQA benchmark, the most challenging setting that has so far required strong reasoning agents such as GPT-4o. Furthermore, its optimized inference strategy consistently yields more than a 1.5x inference speedup over previous approaches. Finally, this work releases GangnamLoop, a novel and practical multi-trip outdoor benchmark collected by deploying a real quadruped robot on public streets with the anonymization policy. It revisits the same locations under different outdoor conditions and pairs the robot's low viewpoint with the human owner's. The source codes and datasets are publicly available at https://github.com/ndb796/BinaryTracking

Problem

• Spatial Question Answering
• Service Robots
• Open Vision-Language Models
• Onboard Navigation
• Egocentric Trajectory

Innovation

• Binary Tracking
• Open Vision-Language Models
• Spatial Question Answering
• Onboard Robot Localization
• Trajectory-based Search