Uncertainty-Guided Proactive Adaptation for Visual-Inertial SLAM
Ehsan Ullah Khan, Gon-Woo Kim
AI summary
Problem
Visual-inertial SLAM systems frequently fail in feature-poor environments because existing methods detect degradation reactively, leaving insufficient time for corrective action.
Approach
The framework uses an LSTM to predict feature counts 1–2 seconds ahead, combined with Monte Carlo Dropout to quantify prediction confidence, enabling adaptive sensor fusion weights based on uncertainty thresholds.
Key results
- LSTM achieves lowest prediction error (26.77 MAE) across 15,233 sequences
- Uncertainty stratification improves adaptation decision precision to 82.4% and recall to 86.3%
- Low-uncertainty predictions achieve 12.30 MAE error with 95.8% precision
- Real-time inference (~50ms) enables practical integration with 20Hz SLAM systems
Why it matters
Enables autonomous robots to proactively prevent localization failures in GPS-denied or textureless environments.
Abstract
Visual–inertial SLAM systems often fail in feature-poor environments such as corridors and textureless walls, leading to catastrophic tracking loss. Existing methods detect degradation reactively after failure occurs, leaving little opportunity for corrective action. We propose a proactive framework that predicts feature degradation 1–2 seconds in advance and adapts sensor fusion weights through uncertainty- guided decisions. Through a systematic comparison of eight temporal architectures across 15,233 sequences, including real robot data, we identify LSTM as the most robust predictor (26.77 MAE). We incorporate uncertainty estimation using Monte Carlo Dropout to enable confidence-aware adaptation thresholds that prevent false adjustments. Our approach pro- vides a foundation for proactive SLAM failure prevention through principled sensor fusion and real-time system adap- tation.