面向车间生产过程的并行碰撞检测算法研究

Research on Parallel Collision Detection Algorithm for Workshop Production Process

针对车间大规模生产过程仿真场景中碰撞检测实时性不足的问题,提出一种基于改进线性层次包围盒的GPU并行碰撞检测算法。对不同类型车间设备进行特征分析以简化碰撞模型,利用空间填充曲线为对象模型分配莫顿码并排序,并将有序莫顿码与节点的特定布局相结合实现层次包围盒的并行化构建,最后使用分离轴定理并行测试得到碰撞检测结果。通过OpenCL并行运算架构实现上述算法,并与其他并行碰撞检测算法对比分析,验证了该方法可以有效提升车间仿真过程中碰撞检测的准确性与实时性。

To solve the problem of insufficient real-time collision detection in the simulation of workshop mass production process,a GPU parallel collision detection algorithm based on linear bounding volume hierarchy was proposed.The collision model was simplified based on the analysis of workshop equipment characteristics.The space filling curve was used to assign Morton codes to the object model and sort them.The ordered Morton codes were combined with the specific layout of nodes to realize the complete parallel construction of the bounding volume hierarchies.Finally,the collision detection results were obtained by parallel testing with the separation axis theorem.The proposed collision detection algorithm is implemented based on OpenCL parallel computing architecture.Compared with other parallel collision detection algorithms,it is verified that the algorithm can effectively improve the accuracy and real-time performance of the collision detection process.