Multimodal flexible sensors for edge AI system

Abstract Wireless multimodal sensor systems that integrate sensing, data processing, and feedback are central to next-generation Internet of Things (IoT) applications in healthcare, robotics, and industrial and environmental monitoring. Flexible and stretchable sensors enable conformal contact with soft or nonplanar surfaces, allowing precise, low-noise signal acquisition. To achieve real-time feedback and low-power consumption, edge computing has emerged as a promising alternative to cloud-based analysis. Within this context, lightweight machine learning methods, such as reservoir computing, offer efficient time-series analysis under resource constraints. However, challenges remain regarding the reproducibility of data from flexible sensors and overall energy consumption. Integrating multimodal sensing, flexibility, and edge computing paves the way for advanced IoT systems that promote more convenience, comfort, and sustainability, thereby contributing to an improved quality of life. This review highlights progress in multimodal flexible sensors, machine learning, physical computing, and edge systems. Furthermore, current challenges and future directions are discussed.