Robust Visual Localization in Compute-Constrained Environments by Salient Edge Rendering and Weighted Hamming Similarity

2025-09-29

AI Summary

Mars sample return missions demand highly robust and accurate 6-DoF pose estimation under severe computational constraints and low-clearance grasping conditions. Method: We propose a lightweight pose estimation algorithm based on salient-edge rendering and weighted Hamming similarity matching. Our approach introduces an edge-domain-specific renderer that generates edge templates solely from low-fidelity, texture-free 3D models, and defines a weighted Hamming distance metric to enhance robustness against illumination variations and sensor noise. Contribution/Results: Trained on synthetic data and fine-tuned across domains—including Earth-based testbeds and real Martian imagery—the method achieves superior accuracy, real-time performance (≥30 FPS on embedded GPUs), and cross-domain generalization compared to state-of-the-art methods. It provides an efficient, reliable visual localization solution for resource-constrained deep-space exploration scenarios.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

We consider the problem of vision-based 6-DoF object pose estimation in the context of the notional Mars Sample Return campaign, in which a robotic arm would need to localize multiple objects of interest for low-clearance pickup and insertion, under severely constrained hardware. We propose a novel localization algorithm leveraging a custom renderer together with a new template matching metric tailored to the edge domain to achieve robust pose estimation using only low-fidelity, textureless 3D models as inputs. Extensive evaluations on synthetic datasets as well as from physical testbeds on Earth and in situ Mars imagery shows that our method consistently beats the state of the art in compute and memory-constrained localization, both in terms of robustness and accuracy, in turn enabling new possibilities for cheap and reliable localization on general-purpose hardware.

Problem

• Achieving robust visual localization under severe hardware constraints
• Enabling 6-DoF object pose estimation with low-fidelity 3D models
• Developing efficient localization for compute-constrained robotic operations

Innovation

• Salient edge rendering for visual localization
• Weighted Hamming similarity metric for matching
• Textureless 3D models enable robust pose estimation