Adaptive Drift Compensation for Soft Sensorized Finger Using Continual Learning

2025-03-18

AI Summary

Soft robotic tactile sensing faces challenges including high nonlinearity, time-varying dynamics, and long-term signal drift in piezoelectric strain sensors, hindering safe physical interaction and accurate object property identification by soft grippers in unstructured environments. To address this, we propose an adaptive continual learning compensation framework specifically designed for piezoelectric strain sensing. Our approach uniquely integrates LSTM-based temporal modeling, memory-buffered experience replay, and Elastic Weight Consolidation (EWC) regularization—enabling dynamic drift correction while preserving initial calibration knowledge without catastrophic forgetting. Evaluated across nine reset experiments, the method consistently outperforms two baseline approaches. Ablation studies confirm the critical contribution of each component to robustness. This work establishes a scalable continual learning paradigm for ensuring long-term stability and reliability in soft tactile sensing systems.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Strain sensors are gaining popularity in soft robotics for acquiring tactile data due to their flexibility and ease of integration. Tactile sensing plays a critical role in soft grippers, enabling them to safely interact with unstructured environments and precisely detect object properties. However, a significant challenge with these systems is their high non-linearity, time-varying behavior, and long-term signal drift. In this paper, we introduce a continual learning (CL) approach to model a soft finger equipped with piezoelectric-based strain sensors for proprioception. To tackle the aforementioned challenges, we propose an adaptive CL algorithm that integrates a Long Short-Term Memory (LSTM) network with a memory buffer for rehearsal and includes a regularization term to keep the model's decision boundary close to the base signal while adapting to time-varying drift. We conduct nine different experiments, resetting the entire setup each time to demonstrate signal drift. We also benchmark our algorithm against two other methods and conduct an ablation study to assess the impact of different components on the overall performance.

Problem

• Addressing signal drift in soft finger strain sensors
• Compensating non-linearity in tactile sensing for soft grippers
• Adapting continual learning for time-varying sensor behavior

Innovation

• Continual learning for soft finger drift compensation
• LSTM network with memory buffer for rehearsal
• Regularization term to maintain signal boundaries