Font Size: a A A

Research On Pose And Position Measurement Method For High Speed Motion Model

Posted on:2013-03-30Degree:MasterType:Thesis
Country:ChinaCandidate:S J LiuFull Text:PDF
GTID:2248330371997624Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
Pose and position measurement techniques have a wide range of applications in the field of aviation, aerospace, industrial automation and robotics research. Especially in the storage separation experiment in wind tunnel, the scaled model is small and separate with a high speed plus the wind acceleration. Meanwhile, the observation field of the wind tunnel is limited and it will adversely affect the flow field if the equipment is installed in the test section, therefore it’s rather difficult to detect the pose and position of the motion model which is quite important in the aerodynamic analysis and model design. Machine vision has been used as a non-contact, high-speed and accurate measurement method and has the potential to be applied in the wind tunnel experiment. In order to improve the level of the wind tunnel experiment conducted in our country, the machine vision system for pose and position measurement of small high speed motion object is studied. This paper mainly focuses on some key techniques as follows:(1) To capture the clear image of the high-speed motion model, two fast cameras are employed as the image acquisition device. According to the environment of the experiment, with the application of constant current LED light source and high reflective material, a target image acquisition program is proposed providing an important hardware basis for the binocular system application.(2) In order to capture the motion of the model in wind tunnel experiment, with the consideration of the vision field has a large range but the target is small, the existing major calibration methods are studied and the plane target is employed to calibrate the cameras in the system. In order to improve the reconstruction accuracy, the feature points reconstruction residuals are introduced as the constraints and the parameters of the system are optimized.(3) Due to the small size of the scaled model, the suitable cooperative target is selected and designed. Meanwhile the recognization of the target and the high precision center extraction algorithm are designed. The target layout and matching method are developed separately for rolling object and non-rolling object. Especially for the rolling object, the target layout based on the spiral acceleration curve is presented to detect the rolling motion. This method can match the targets with the designed model in a single.frame without target tracing which provides a good foundation for the rolling model pose and position measurement. (4) According to the environment of the wind tunnel experiment and the theoretical analysis, the system hardware and software are designed. With the system designed, some experiments are carried out. The feasibility and effectiveness of target pattern and layout design, calibration method and feature extracting method are verified. The method proposed in this paper provides a foundation for the pose and position measurement method in the wind tunnel experiment and also has reference value for the pose and position measurement technique of the high speed motion target.
Keywords/Search Tags:Wind tunnel experiment, High-speed motion, Pose and positionmeasurement, Machine Vision, Image Acquisition, Target matching
PDF Full Text Request
Related items