Research Of Scene Matching Guidance Hardware-in-the-loop (hwil) Simulation System

Hardware-in-the-loop (HWIL) simulation fills the neutral of digital/mixed simulation and the actual outfield experimentation. HWIL has authenticity, real-time, controllability, repeatability, covered, economical. HWIL can test the capability of system components directly and efficaciously, reduce the times of actual trial, debase the cost, economize manpower and material resources and shorten the experimentation period. According to the actual project requirement,this paper researches the biological vision scene matching guidance HWIL simulation system, expounds the composition of the system and its function, and conducts relevant HWIL simulation experimentation.The paper conducts vehicle autonomous velocity measurement HWIL simulation experimentation on condition of linear fly. Using image mosaic technology, referential image was generated. The matching of real-time images and reference image is accomplished using method image matching algorithm based on biological vision. Use based on Kalman filtering and optimal smoothing theory of velocity measurement algorithm and straight-line fitting algorithm to process the experimental date, and compare the results with the standard velocity of the 6-freedom movement equipment. It shows that the relative error of the algorithms is around 3%, and it also shows that the speed measurement algorithm have high accuracy and reliability. The HWIL simulation system provides an economic, efficient and reliable velocity measurement.The scene matching guidance based on muti-reference images HWIL simulation experimentation is conducted in this paper. The change of image resolving power due to the height of the camera fall makes the image matching difficulty, so the algorithm of real image magnified multiple based on changing magnitude of height and based on tracking two points. The principle of the algorithm based on tracking two points is simple, and it does not need added equipment. The algorithm with high operation velocity is prone to be realized in actual project, and brings predicted effect and high target acquisition success rate in the HWIL simulation experimentation.The algorithm of calculating camera place and attitude via image matching is researched in this paper. With the message of reference image and intrinsic parameter, the real image and the reference image match 5 points at lest, then the camera extrinsic parameter can be gained via solving non-linear equations. If the camera angles are known, the number of image match becomes 2 points at lest, and the equations are linear. The HWIL simulation experimentation validates the algorithm is correct and feasible.