Pose measurement of small-size aircraft based on machine vision
作者单位：School of Mechatronic Engineering and AutomationShanghai University School of Electrical Engineering and AutomationHarbin Institute of Technology
会议名称：《第40届中国控制会议》
会议日期：2021年
学科分类：08[工学] 081105[工学-导航、制导与控制] 080203[工学-机械设计及理论] 0802[工学-机械工程] 0811[工学-控制科学与工程]
关 键 词：Machine vision Ground test of aircraft Pose estimation Vector cross
摘 要：In small-size aircraft ground tests,pose measurement is significant for the validation of aircraft control and guidance algorithm development and performance evaluation.In order to achieve a precise value of the pose in the aircraft ground test,Firstly,an aircraft ground test localization system based on machine vision has been established and proposed a global localization method for the smaller sizes of aircraft ground test.Multi-view matching of feature points is realized by using the spatiotemporal consistency method.The method of uncertainty analysis is used to realize the high precision 3D reconstruction of feature points from multiple views.The pose parameters of the aircraft are solved by using the pose estimation algorithm of non-collinear vectors.The feasibility and accuracy of the pose measurement algorithm are verified by simulation and actual experiment.The experimental results show that:under the condition of the measurement range is 6000 mm×6000 mm×3000 mm,the size of the aircraft is about 400 mm,and the 3D positioning accuracy of the feature points is 2.9 mm,the theoretical accuracy of the pose measurement algorithm based on vector crossover can reach 0.5°（3σ） and 3 mm（3σ） respectively.The actual relative measurement accuracy up to 1.3 °（3σ） and 4 mm（3σ）.It meets the requirements of stable and reliable performance evaluation,high precision,and strong anti-jamming ability of small-size aircraft guidance and control system.