Research On The Real-Time Motion Parameters Measuring System

In the field of astronautical and robot development, motion parameters can not only correct the drifts from Inertial Measurement Unit, but also can act as significant basis for experimental results identification and incident causes judgement. And it need to monitor target's movement for a number of applications, Therefore, develop the real-time motion parameters measuring system is imperative. So this paper makes the research work of the following aspects:First, design the programs of the real-time motion parameters measuring system. it gives the system overall design plan based on the Motion parameters vision measuring principle which is based on feature lines, on this basis, this paper presents the method of integration of the coordinates, and aiming at the problem of cameras synchronization, it gives the corresponding design scheme; hardware equipments and development environment are briefly introduced.Second, research the algorithm of image acquisition, feature dectection and extraction based on the LabVIEW FPGA. Aiming at the system of image acquisition and processing for the NI FlexRIO FPGA modules, using the FIFO transfer, ping-pong operations and DMA transfer realizes real-time acquisition and display for image; in order to slove the feature dectection and extraction, the paper take advantage of the parallel processing capabilities of the FPGA to presents a operator for dectect which can quickly detect the features without duplication, and then use the pipelining technique divide gray weighted centroid algorithm into discrete steps to realizes the feature extraction.Next, research the problems of features tracking and motion parameters real-time estimation, In the act of measurement, features disappear or false features appear which is caused by the visual field of cameras and other reasons, in practice, aiming at there problems, this paper establishes motion modles of feature points based on movement smooth constraints, and uses the hungary algorithm to search the movement trajectory of feature points, it can slove the feature tracking even in the case of occlusion and flase points. And in order to improve the accuracy of motion parameters and deal the problem of motion estimation with the occlusion, so this paper establishes the modle for motion parameters estimation, and the corresponding measurement equation transition matrix are discussed in detail, on this basis, using the extended Kalman filter to achieve the real-time estimation of motion parameters.Finally, the paper design the software and hardware platform for the measuring system based the research results, after this, an dynamic experiment of the motion parameters real-time measuring system was carried on, and the results was analyzed in detail.