Keynote Speakers
 |
Advances in MEMS/NOEMS Technology –
|
| With the growing demand for energy-efficient AI applications, the rapid development of MEMS/NOEMS and self-powered sensors together with edge computing and edge AI technology at the sensor nodes has led to the new era of AI sensors. Traditional sensors and sensing systems can no longer meet the demands for real-time multimodal sensing and large-scale data processing, leading to a shift towards a new paradigm of AI Sensors and Artificial Intelligence of Things (AIoT) sensing systems with integrated computational intelligence. The convergence of sensing, computing, and actuation at the chip level is redefining the future of intelligent systems. Recent progress in MEMS technology has unlocked new pathways for embedding reconfigurability, adaptability, and intelligence directly into the physical layer of sensors. This talk explores how emerging microsystems—built at the intersection of MEMS, integrated photonics, and AI—are enabling a new generation of AI sensors optimized for edge intelligence. We highlight two synergistic platforms: guided-wave photonic circuits and free-space nanoantenna metasurfaces. The guided-wave architecture leverages electro-optic, acousto-optic, and MEMS-tunable photonic elements to construct reconfigurable photonic neural networks capable of real-time, light-speed signal processing. These systems offer a unified path toward high-bandwidth optical communication and computing on chip. On the other hand, free-space platforms integrate nanophotonic antenna arrays with emerging functional materials and MEMS structures to enable programmable, multi-modal, and adaptive sensing surfaces. These intelligent metasurfaces open new opportunities for local inference and context-aware responses, laying the foundation for next-generation optical edge devices. Self-powered wearable sensors have promoted low-power or battery-free sensing platforms for applications including human-machine interaction, soft robotics, and electronic skin (e-Skin). By embedding triboelectric, piezoelectric, electromagnetic and hybrid nanogenerators directly into the sensing structure, the same transducer now harvests ambient mechanical, vibrational or thermal energy while generating electrical signals based on stimuli for sensing, creating genuinely self-sustained sensor nodes. Breathable tribo-ferroelectric textiles, hydrogel-based nano-generators and silk-derived e-skins exemplify this dual-function strategy, delivering gesture-tracking, pressure–temperature mapping and long-term physiological monitoring without external batteries. Similarly, hybridized generators with combining nano-generators can further power high-power consumption modules and realize AI-enhanced self-powered sensing at the same time, enabling fully autonomous wearables and industrial condition-monitoring systems. By embedding in-sensor computing capabilities within reconfigurable photonic hardware, these platforms reduce system latency, power consumption, and data transfer burdens. Using optical edge computing platform advanced by Si Photonics sensors, optical edge AI sensors have become an attractive approach for enabling intelligent, energy-efficient, adaptive sensing networks. Beyond Si photonics, edge-AI hardware is rapidly converging on memristor and ferroelectric-transistor crossbar accelerators, reservoir- and convolution-type photonic processors, flexible neuromorphic synapses, and spiking-neuron arrays that fuse sensing, memory and computation on the same chip or fabric. This presentation outlines the architectural vision, key enabling technologies, and potential impact of hybrid MEMS/NOEMS in shaping the next era of intelligent edge microsystems. Biography |
|
