Adaptive Relative Pose Estimation Framework with Dual Noise Tuning for Safe Approaching Maneuvers

2025-07-22

AI Summary

For high-precision, robust relative pose estimation of non-cooperative tumbling targets (e.g., ENVISAT) in active debris removal missions, this paper proposes a navigation framework integrating deep learning with adaptive filtering. Methodologically, a CNN—augmented by image preprocessing—extracts corner features, and camera projection models generate 3D measurements; a dual noise-adaptation mechanism jointly optimizes process and measurement noise covariances within an Unscented Kalman Filter (UKF), effectively mitigating dynamic modeling errors and measurement uncertainties. Evaluated in a high-fidelity ENVISAT simulation environment, the method achieves over 35% improvement in estimation accuracy compared to conventional approaches. Moreover, it maintains stable convergence under challenging conditions—including measurement outages and rapid, irregular tumbling—ensuring reliable navigation for safe proximity operations.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Motion & Tracking
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Accurate and robust relative pose estimation is crucial for enabling challenging Active Debris Removal (ADR) missions targeting tumbling derelict satellites such as ESA's ENVISAT. This work presents a complete pipeline integrating advanced computer vision techniques with adaptive nonlinear filtering to address this challenge. A Convolutional Neural Network (CNN), enhanced with image preprocessing, detects structural markers (corners) from chaser imagery, whose 2D coordinates are converted to 3D measurements using camera modeling. These measurements are fused within an Unscented Kalman Filter (UKF) framework, selected for its ability to handle nonlinear relative dynamics, to estimate the full relative pose. Key contributions include the integrated system architecture and a dual adaptive strategy within the UKF: dynamic tuning of the measurement noise covariance compensates for varying CNN measurement uncertainty, while adaptive tuning of the process noise covariance, utilizing measurement residual analysis, accounts for unmodeled dynamics or maneuvers online. This dual adaptation enhances robustness against both measurement imperfections and dynamic model uncertainties. The performance of the proposed adaptive integrated system is evaluated through high-fidelity simulations using a realistic ENVISAT model, comparing estimates against ground truth under various conditions, including measurement outages. This comprehensive approach offers an enhanced solution for robust onboard relative navigation, significantly advancing the capabilities required for safe proximity operations during ADR missions.

Problem

• Estimating relative pose for Active Debris Removal missions
• Handling measurement and dynamic model uncertainties
• Enhancing robustness in proximity operations for ADR

Innovation

• CNN with preprocessing for marker detection
• UKF framework for nonlinear pose estimation
• Dual adaptive noise tuning strategy