Specific Dynamic Parameters: A Novel Multi-View Stereo Vision Measurement System for Vector Nozzle

As a crucial component of aero-engines, the accurate measurement of the motion parameters of the vector nozzle is of great significance for thrust vector control and flight performance improvement. This paper designs a dual-bearing rotating nozzle to simulate the motion of a real vector nozzle and a multi-view vision platform specifically for measuring the motion parameters of the vector nozzle. To address the parameter measurement problem, this paper proposes a technical solution of "3D reconstruction-point cloud segmentation-fitting measurement". First, a deep learning-based multi-view stereo vision method is used to reconstruct the nozzle model in 3D. Then, the point transformer method is used to specifically extract the point cloud of the nozzle end face region. Finally, the RANSAC method is used to fit a spatial circle to the end face point cloud for measurement purposes. This paper verifies the effectiveness of the proposed method through multiple sets of comparative experiments and evaluates the reliability and stability of the measurement system using analysis of variance in measurement system analysis. The final analysis results show that the GRR% and NDC of the measurement system are 2.81% and 7.69%, respectively, and 1.84%, 50, 18, and 76, respectively, indicating that the measurement system has high accuracy and stability.