基于CPU-GPU的B样条曲面并行刀具路径规划方法

摘要

Aiming at the inefficiency of legacy serial tool path algorithms and incompatibility issues on heterogeneous hardware platforms, a tool path planning method based on CPU-GPU heterogeneous parallel computing is proposed.The method contraposes bi-cubic B-spline surface which is abstracted as a matrix multiplication on the principle of isoparametric line tool path planning method, and then designs parallel algorithm in accordance with OpenCL specification.Adopting data parallel programming model, it executes multiple work-items of the GPU on the core under control of the CPU logic, and reconstructs the isoparametric method as parallel execution instead of traditional serial execution.Obviously, simulation results show that this algorithm takes less time to generate tool paths on the CPU-GPU heterogeneous platforms, reduced by 1.5 to 11.9 times compared with traditional serial algorithm and it is of great significance to the tool path planning's real-time or near real-time generation.%针对传统串行刀具路径规划算法效率低下和在异构硬件平台上的不兼容问题,提出一种基于CPU-GPU异构并行计算的刀具路径规划方法.方法针对双三次均匀B样条曲面,依据等参数线刀具路径规划方法的原理和OpenCL规范设计并行算法,在CPU的逻辑控制下,采用数据并行的编程模型在GPU的多个工作项上并行执行内核,将传统串行执行的等参数线法进行了并行化重构.仿真实验结果表明,该算法在CPU-GPU异构平台上生成刀具路径的时间较传统串行算法缩短1.5~11.9倍,对实现刀具路径的实时或准实时生成具有重大意义.