Human State Estimation

Human state estimation aims to infer a person's location, activity, and internal state (e.g., trust, cognitive load) using various sensor modalities, such as wearable sensors, UWB radar, and visual data. Current research focuses on developing robust models, including deep neural networks (e.g., convolutional and recurrent architectures) and probabilistic frameworks (e.g., POMDPs), to handle noisy and incomplete data, often incorporating attention mechanisms to improve efficiency and accuracy. This field is crucial for enhancing human-robot interaction, improving workplace safety through risk prediction and prevention, and creating more intuitive and responsive technologies in various applications, from smart homes to autonomous vehicles.