Chips That Feel

Beschreibung

Electronic skin (e-skin) is an emerging technology that can be applied in the fields of medical prosthetics and robotics: e-skin provides the sense of touch to mechanical systems. This book bridges the gap between existing e-skin technologies and the remarkable capabilities of human tactile perception. It presents end-to-end design strategies for next-generation fully-integrated e-skin systems that mirror essential properties of human skin, such as being multi-modal, multi-scale, and performing local processing. Building on a pragmatic, technology-aware neuromorphic design approach, that draws inspiration from biology without blindly replicating it, the book shows the system-level benefits of event-based approaches through temporal compression of tactile data via spikes, and it explains how spike-based representations support efficient neuromorphic processing with spiking neural networks, relaxing readout design constraints. Two validated e-skin chips, called INTUITIVE and DERMIS, are presented. They form a pathway toward e-skin systems with human-like tactile perception and embedded intelligence, combining high-resolution texture sensing with low-latency, large-area slip detection to bring electronic skin closer to the capabilities of the human skin.

• Offers pioneering, detailed technical discussions on fully-integrated e-skin designs;
• Proposes a pragmatic, technology-aware bioinspired approach to e-skin design, implementing what supports the system;
• Demonstrates how to treat e-skin design with a system-level, end-to-end view.

Electronic skin (e-skin) is an emerging technology that can be applied in the fields of medical prosthetics and robotics: e-skin provides the sense of touch to mechanical systems. This book bridges the gap between existing e-skin technologies and the remarkable capabilities of human tactile perception. It presents end-to-end design strategies for next-generation fully-integrated e-skin systems that mirror essential properties of human skin, such as being multi-modal, multi-scale, and performing local processing. Building on a pragmatic, technology-aware neuromorphic design approach, that draws inspiration from biology without blindly replicating it, the book shows the system-level benefits of event-based approaches through temporal compression of tactile data via spikes, and it explains how spike-based representations support efficient neuromorphic processing with spiking neural networks, relaxing readout design constraints. Two validated e-skin chips, called INTUITIVE and DERMIS, are presented. They form a pathway toward e-skin systems with human-like tactile perception and embedded intelligence, combining high-resolution texture sensing with low-latency, large-area slip detection to bring electronic skin closer to the capabilities of the human skin.

Autor*in

Mark Daniel Alea

Themen in »Chips That Feel«

Electronic Skin Neuromorphic Skin Tactile Sensors Taxel Readout Chip Embedded Receptive Field Processing

Stimmen zu »Chips That Feel«

Details

Link teilen