SpikeATac: A Multimodal Tactile Finger with Taxelized Dynamic Sensing for Dexterous Manipulation

2025-10-30

AI Summary

To address the challenge of simultaneously achieving dynamic contact sensing and static force control in dexterous manipulation of fragile, deformable objects, this paper introduces SpikeATac—a multimodal tactile fingertip. It integrates a 16-taxel PVDF piezoelectric film (4 kHz sampling) for millisecond-scale dynamic response (e.g., contact initiation/termination) and capacitive sensing for high-resolution static pressure mapping. We propose, for the first time, a “spike + steady-state” dual-mode tactile representation and design a tactile-reward-driven reinforcement learning framework informed by human feedback to enable end-to-end policy fine-tuning. Experiments demonstrate that SpikeATac achieves rapid response, precise force regulation, and stable interaction in high-contact-complexity tasks—including egg grasping and soft-object flipping—significantly extending the operational envelope of multifinger dexterous hands for handling delicate objects.

Technology Category

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

Application Category

• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: The sharing economy

Abstract

In this work, we introduce SpikeATac, a multimodal tactile finger combining a taxelized and highly sensitive dynamic response (PVDF) with a static transduction method (capacitive) for multimodal touch sensing. Named for its `spiky' response, SpikeATac's 16-taxel PVDF film sampled at 4 kHz provides fast, sensitive dynamic signals to the very onset and breaking of contact. We characterize the sensitivity of the different modalities, and show that SpikeATac provides the ability to stop quickly and delicately when grasping fragile, deformable objects. Beyond parallel grasping, we show that SpikeATac can be used in a learning-based framework to achieve new capabilities on a dexterous multifingered robot hand. We use a learning recipe that combines reinforcement learning from human feedback with tactile-based rewards to fine-tune the behavior of a policy to modulate force. Our hardware platform and learning pipeline together enable a difficult dexterous and contact-rich task that has not previously been achieved: in-hand manipulation of fragile objects. Videos are available at href{https://roamlab.github.io/spikeatac/}{roamlab.github.io/spikeatac}.

Problem

• Developing multimodal tactile finger for dynamic and static touch sensing
• Enabling delicate grasping of fragile deformable objects using tactile feedback
• Achieving in-hand manipulation of fragile objects through learning-based framework

Innovation

• Combines PVDF dynamic sensing with capacitive static transduction
• Uses 16-taxel film sampled at 4 kHz for fast contact detection
• Integrates reinforcement learning with tactile rewards for force modulation