Data-efficient Tactile Sensing with Electrical Impedance Tomography

2024-11-19

arXiv.org

AI Summary

To address the limited reconstruction accuracy and resolution in electrical impedance tomography (EIT)-based tactile sensing caused by scarce training data, this paper proposes a multi-view signal-domain data augmentation method operating on a single-frame EIT measurement. Leveraging only one original voltage measurement, the method synthesizes 32 equivalent, statistically independent training samples—without requiring additional hardware or measurements—effectively compensating for missing spatial positional information. The approach integrates the underlying EIT physical model, a deep neural network, and a customized signal transformation, thereby overcoming the conventional reliance on dense electrode measurements. Simulation results demonstrate a 12.3% improvement in image correlation coefficient and a 21.7% reduction in relative reconstruction error. Experimental validation confirms that merely 1/31 of the original dataset suffices to achieve comparable reconstruction quality, significantly alleviating generalization challenges under small-sample conditions.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Electrical Impedance Tomography (EIT)-inspired tactile sensors are gaining attention in robotic tactile sensing due to their cost-effectiveness, safety, and scalability with sparse electrode configurations. This paper presents a data augmentation strategy for learning-based tactile reconstruction that amplifies the original single-frame signal measurement into 32 distinct, effective signal data for training. This approach supplements uncollected conditions of position information, resulting in more accurate and high-resolution tactile reconstructions. Data augmentation for EIT significantly reduces the required EIT measurements and achieves promising performance with even limited samples. Simulation results show that the proposed method improves the correlation coefficient by over 12% and reduces the relative error by over 21% under various noise levels. Furthermore, we demonstrate that a standard deep neural network (DNN) utilizing the proposed data augmentation reduces the required data down to 1/31 while achieving a similar tactile reconstruction quality. Real-world tests further validate the approach's effectiveness on a flexible EIT-based tactile sensor. These results could help address the challenge of training tactile sensing networks with limited available measurements, improving the accuracy and applicability of EIT-based tactile sensing systems.

Problem

• Enhances tactile sensing accuracy with limited data.
• Reduces required measurements for effective tactile reconstruction.
• Improves EIT-based tactile sensor performance in real-world applications.

Innovation

• Data augmentation amplifies single-frame signals.
• Reduces required EIT measurements significantly.
• Improves tactile reconstruction accuracy and resolution.