Problem
Research questions and friction points this paper is trying to address.
Fault Detection
Mobile Robots
Edge Computing
Time-Series Anomaly Detection
Online Adaptation
Continuous Online Fault Detection for Mobile Robots via Adaptive Edge Models
🤖 AI Summary
This study addresses the challenge of deploying deep temporal anomaly detection models on resource-constrained edge devices for mobile robots in real time. We propose a teacher-student distillation framework with online adaptation, wherein a TSPulse teacher model generates pseudo-labels to train a lightweight MiniRocket student model. A recursive least squares estimator enables low-latency online updates, while an uncertainty-guided active learning strategy effectively mitigates catastrophic forgetting and reduces manual annotation overhead. Experimental results demonstrate that the system achieves a CPU inference latency of only 4.30 milliseconds and improves the VUS-PR metric from 0.26 to 0.75 under real-world domain shifts, validating its deployment feasibility in resource-limited scenarios.
📝 Abstract
Mobile robots require robust, real-time fault detection capable of continuous adaptation on constrained edge hardware. While deep time-series models excel at unsupervised anomaly detection, their computational cost prohibits high-frequency onboard execution. This paper bridges this gap via a Teacher-Student distillation framework. An offline foundation model (TSPulse) generates pseudo-labels from unlabeled time series augmented with fault injections. A lightweight MiniRocket Student, adapted with a Recursive Least Squares estimator, approximates this complex decision boundary to execute real-time inference onboard. Evaluations on the TSB-AD benchmark and a physical mobile robot demonstrate the Student achieves a 4.30 ms CPU inference latency. During real-world domain shifts, online adaptation enables the Student to recover from unseen mechanical degradation, improving VUS-PR scores from 0.26 to 0.75 without catastrophic forgetting. Crucially, an uncertainty-guided active learning strategy minimizes operator cognitive load, requesting sparse interventions only when encountering novel fault distributions. These results validate the deployment of state-of-the-art anomaly detection on resource-constrained robotics through offline-to-online distillation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Knowledge Distillation
Online Adaptation
Edge Computing
Active Learning
Anomaly Detection
🔎 Similar Papers
No similar papers found.
