神经形态工程学
计算机科学
可穿戴计算机
数码产品
冯·诺依曼建筑
人工智能
嵌入式系统
传感器融合
可穿戴技术
计算机体系结构
人工神经网络
人机交互
工程类
电气工程
操作系统
作者
Feng Wen,Chan Wang,Chengkuo Lee
出处
期刊:Nano Research
[Springer Nature]
日期:2023-07-28
卷期号:16 (9): 11801-11821
被引量:5
标识
DOI:10.1007/s12274-023-5879-4
摘要
Wearable and flexible electronics are shaping our life with their unique advantages of light weight, good compliance, and desirable comfortability. With marching into the era of Internet of Things (IoT), numerous sensor nodes are distributed throughout networks to capture, process, and transmit diverse sensory information, which gives rise to the demand on self-powered sensors to reduce the power consumption. Meanwhile, the rapid development of artificial intelligence (AI) and fifth-generation (5G) technologies provides an opportunity to enable smart-decision making and instantaneous data transmission in IoT systems. Due to continuously increased sensor and dataset number, conventional computing based on von Neumann architecture cannot meet the needs of brain-like high-efficient sensing and computing applications anymore. Neuromorphic electronics, drawing inspiration from the human brain, provide an alternative approach for efficient and low-power-consumption information processing. Hence, this review presents the general technology roadmap of self-powered sensors with detail discussion on their diversified applications in healthcare, human machine interactions, smart homes, etc. Via leveraging AI and virtual reality/augmented reality (VR/AR) techniques, the development of single sensors to intelligent integrated systems is reviewed in terms of step-by-step system integration and algorithm improvement. In order to realize efficient sensing and computing, brain-inspired neuromorphic electronics are next briefly discussed. Last, it concludes and highlights both challenges and opportunities from the aspects of materials, minimization, integration, multimodal information fusion, and artificial sensory system.
科研通智能强力驱动
Strongly Powered by AbleSci AI