Parallelized Collision Detection Algorithm for Non-contact Machining Simulation Based on Space Partitioning and Path Point Equalization

Abstract

High-energy beam CNC machining, such as laser, water jet, and plasma machining, is widely applied in modern manufacturing due to its non-contact processing, precise energy control, and broad adaptability. However, potential collisions between tools and machine tools can cause serious accidents, making pre-processing simulation essential. As tasks become more complex, the increase in triangular facets of models and tool path points leads to a significant rise in computational overhead for CNC machining simulations. Although multi-core processors provide powerful computing capabilities, traditional serial collision detection methods fail to fully utilize these resources. To address this issue, we propose a parallelized collision detection algorithm for high-energy beam CNC machining simulations. By dividing the tool paths into sub-tasks and leveraging multi-core processors for concurrent processing, the algorithm enhances efficiency while maintaining accuracy. Experiments in water jet and laser machining scenarios demonstrate that our approach effectively reduces computation time and improves real-time performance and reliability. This study introduces parallel computing strategies to improve collision detection efficiency, combines the parallelization idea with the collision detection algorithm in CNC machining simulation offering an efficient solution for complex high-energy beam machining tasks.