The Research Of Large Space Vision Shape Measurement Technology Based On UAV

With the rapid development of science and technology,large equipment manufacturing industry for the measurement of workpiece shape measurement accuracy,efficiency,scope and flexibility of work put forward higher requirements.Large space vision shape measurement technology can meet the above requirements by its own non-contact,high precision and high efficiency,which becomes one of the most effective methods to solve the large-space shape measurement.The existing large-space vison shape measurement technology has special requirements for measuring equipment,field and personnel,and it is necessary to measure the position of some key parts of large-scale workpieces by artificial form,which limits the flexibility of measurement methods,but also has some security risks.Therefore,from the point of view of improving the flexibility of vision shape measurement and expanding its application range,this thesis presents a large space vision shape measurement method based on unmanned aerial vehicle(UAV),which uses UAV platform to carry the measurement equipment to realize vision appearance measurement.Which can solve the problem of positioning information acquisition and target appearance measurement of the measurement platform itself,and realize the synchronization of the position and profile measurement process.The main contents of this thesis are as follows:1.Aiming at the shortcomings of the existing technology,a vision shape measurement method based on multi-direction intersection is proposed.The measurement platform can be carried by the UAV,While solving the measurement of the pose of the measurement platform and the target of the shape measurement,to achieve large-scale objects of the vision shape measurement.2.The design of the optical feature points in the system and the center extraction are completed.ARM + CMOS platform is used to collect and store the target feature images.The software of the shape solution software is built by OpenCV + VS2010 environment.At the same time,a lot of influence factors of pose solving process are analyzed by MATLAB.3.The relationship between the internal reference and the coordinate system of the vision sensor in the large-space vision shape measurement system is calibrated.The calibration method of concentric circle 2D target is used to calibrate the vision sensor.At the same time,the pose relationship between non-public vision field sensors is calibrated according to the relationship between them and vision common coordinate system.4.A static verification experiment platform is set up to verify the feasibility of the measurement method.The experimental results show that the accuracy of the proposed method is within 5mm within the working range of 3m x 3m,which can meet the needs of large space vision shape measurement.