FAR-AVIO: Fast and Robust Schur-Complement Based Acoustic-Visual-Inertial Fusion Odometry with Sensor Calibration

2025-12-23

AI Summary

Underwater weak excitation motion and turbid environments degrade observability and cause frequent tracking failures in visual-inertial odometry (VIO). To address these challenges, this paper proposes a tightly coupled acoustic-visual-inertial fusion framework for real-time, robust pose estimation on resource-constrained underwater robots. Key contributions include: (1) the first Schur-complement-embedded extended Kalman filter (EKF), enabling constant-time joint state optimization; (2) AWARE—an online sensor health assessment and adaptive weighting mechanism; and (3) a calibration-free, online joint extrinsic calibration method for DVL–IMU alignment. Evaluated in both simulation and real-world underwater experiments, the framework achieves 32% higher localization accuracy than state-of-the-art underwater SLAM systems, with per-frame processing time under 15 ms. The implementation is open-sourced and successfully deployed on low-power embedded platforms.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Hardware-aware ML

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Underwater environments impose severe challenges to visual-inertial odometry systems, as strong light attenuation, marine snow and turbidity, together with weakly exciting motions, degrade inertial observability and cause frequent tracking failures over long-term operation. While tightly coupled acoustic-visual-inertial fusion, typically implemented through an acoustic Doppler Velocity Log (DVL) integrated with visual-inertial measurements, can provide accurate state estimation, the associated graph-based optimization is often computationally prohibitive for real-time deployment on resource-constrained platforms. Here we present FAR-AVIO, a Schur-Complement based, tightly coupled acoustic-visual-inertial odometry framework tailored for underwater robots. FAR-AVIO embeds a Schur complement formulation into an Extended Kalman Filter(EKF), enabling joint pose-landmark optimization for accuracy while maintaining constant-time updates by efficiently marginalizing landmark states. On top of this backbone, we introduce Adaptive Weight Adjustment and Reliability Evaluation(AWARE), an online sensor health module that continuously assesses the reliability of visual, inertial and DVL measurements and adaptively regulates their sigma weights, and we develop an efficient online calibration scheme that jointly estimates DVL-IMU extrinsics, without dedicated calibration manoeuvres. Numerical simulations and real-world underwater experiments consistently show that FAR-AVIO outperforms state-of-the-art underwater SLAM baselines in both localization accuracy and computational efficiency, enabling robust operation on low-power embedded platforms. Our implementation has been released as open source software at https://far-vido.gitbook.io/far-vido-docs.

Problem

• Underwater visual-inertial odometry suffers from poor visibility and motion limitations
• Acoustic-visual-inertial fusion is computationally heavy for real-time robot deployment
• Sensor calibration and reliability assessment require manual intervention underwater

Innovation

• Schur complement EKF for constant-time pose-landmark optimization
• Adaptive sensor weight adjustment via online reliability evaluation
• Online DVL-IMU extrinsic calibration without dedicated maneuvers