IMU-Based Motion Trajectory Reconstruction and Recognition with Dynamic Calibration on Embedded System Platform

Motion trajectory reconstruction and recognition with wireless devices are crucial in applications such as Virtual Reality (VR), sports performance monitoring, and intelligent medical systems. Traditional camera-based systems, though accurate, suffer from limited field of view, latency issues, and high costs. In contrast, wearable devices like Inertial Measurement Unit (IMU) sensors offer a compact and cost-effective alternative. Despite their advantages, low-cost IMUs face challenges like scaling errors, sensor axis misalignment, and nonzero biases. This paper proposes a novel Embedded System Platform (ESP) design for IMU-based trajectory reconstruction and recognition, incorporating dynamic calibration to address these challenges. Experimental results demonstrate the effectiveness of the proposed approach in enhancing motion trajectory reconstruction and recognition accuracy.