| Airborne head-up display system is one of the important display devices in modern aircraft,which plays an important role in improving the combat effectiveness of aircraft and ensuring the safety of flight.Video and graphics generation system is used to achieve the graphics rendering and video real-time processing,which is one of the HUD’s core components.In this paper,the deficiencies of the existing video and graphics generation system are studied in depth,aiming at optimizing the system architecture,reducing resource consumption,improving the quality of graphics display and video processing speed.The main work is as follows:(1)The function test and performance test of the existing video and graphics generation system are completed.According to the integrity of the function and the actual image display quality,pointed out the direction of system improvement and optimization.(2)The system logic structure is optimized.By adjusting the execution order and operation steps between the modules,the system function is guaranteed to be correct,and the complexity of some functions is reduced.According to the actual demand,the BRAM resource allocation is optimized.After optimization,a total of 58.3% of the BRAM resources are used,and reduced by 25.3%,which improves the system stability and scalability.(3)The improved multi-line-width triangulation is proposed,which makes the vertices of multi-line-width triangles sharp and distortion-free,and it is also suitable for polygon vertex-free rendering.An improved arc rapid drawing algorithm is proposed,which overcomes the problems of the calculation of a large number of trigonometric functions and the judgment of redundant points in the angle discrimination method,and effectively improves the speed and efficiency of arc drawing.(4)An improved bidirectional mapping rotation algorithm is proposed,which greatly reduces the complexity of the algorithm by using the lookup table technique.When the resolution is 1024 × 1024,the multiplication calculation is reduced by 3 orders of magnitude compared with the original algorithm,which effectively improves the video processing speed.The algorithm of anti-aliasing characters and video superposition is studied,and the hardware implementation is optimized.The edge of the overlaid character is smooth and has no sawtooth,which effectively improves the display effect.At the same time,the complexity of DDR3 interrupt processing and the system processing delay are reduced.(5)The design,optimization and realization of high-speed communication interface are completed.The communication between CPU and GPU through PCI-E bus is achieved,which support PIO and DMA two kinds of data transmission modes.Optimize and realize DDR3 interrupt nesting function,improve the system speed and reliability.The system self-checking function is designed and implemented,and the detection information is fed back to the CPU in real time through PCI-E bus,which effectively improves the system fault fast positioning capability.According to the test results,the chip in the non-working state is set to the power saving mode,which reduces the system power consumption.The test results show that the quality of the system graphics is improved significantly,the delay of the video processing is reduced,the maximum delay of PAL is 39.2ms,reduced by 64.8ms.The overall power consumption is 11.13 W,which is 70% lower than that of the previous generation.The system has passed the joint testing and acceptance,has excellent performance,and the indicators are in line with technical agreement requirements. |
PDF Full Text Request