Fusion of inertial and optical measurements for high-frequency positioning in stereo line-scan-based 3D shape measurement

Accurate and efficient 3D shape measurement of large or complex objects remains challenging, primarily due to the limited update rate and occlusion sensitivity of existing optical positioning systems. To address this, we propose an inertial-optical fusion positioning system that integrates a high-frequency inertial measurement unit (IMU) with an optical system such as a laser tracker. By applying an extended Kalman filter and backward smoothing algorithm, the system achieves continuous 6-DoF pose estimation with an effective update rate of up to 2000 Hz, far exceeding that of standalone optical trackers. Experimental results demonstrate that when optical updates are available at 2 Hz or higher, the fused system maintains shape measurement accuracy within 0.2 mm. This fusion approach not only compensates for the drift inherent in IMUs but also enables high-speed and flexible scanning in partially occluded or dynamic environments. Overall, the results highlight the system’s ability to significantly enhance positioning frequency and robustness, making it well-suited for highspeed 3D shape measurement in complex industrial settings.