high performance computing on graphics processing units: hgpu.org

High-velocity and high-accuracy are the development direction of the numerical control technology. During the machining of complicated curves and surfaces, by CAD/CAM software, the massive micro-segments are generated. Then the micro-segments are inputted into numerical control system (CNC) to process the velocity planning and high-velocity interpolation. This whole procedure is the core algorithm of CNC. Because the data quantity is very large, and during the procedure of velocity planning, the look-ahead iterative calculation is used for several specified future micro-segments, the calculation quantity is very large. Additionally, the CNC is a typical hard real-time system. Every interpolation step must be finished in specified interpolation period, so the high calculating ability of CPU is needed. If the performance of CPU is not excellent enough, the interpolation accuracy and velocity will be lower, and sometimes even the system real-time property is broken. The concept of GPU is quite different with the traditional CPU. It has a very strong parallel computing ability. In this project, the GPU technology is introduced into CNC algorithm. The parallel structure model is construted by CUDA, and a discrete planning unit look-ahead velocity control strategy is proposed. The GPU parallel computing mechanism of massive micro-segments is discussed and base on the research, a high-performance velocity planning algorithm is presented. This research result can highly raise the velocity planning and interpolation efficiency. This brand new solution can help to improve the performance of CNC system dramatically.