Research On Path Planning Of 3D Object Surface Processing

Laser processing path planning is an important issue in the field of laser processing.With the development of laser processing technology,this issue has also received extensive attention and research.Among the object models of laser processing,the most widely used is the three-dimensional model,so the surface path planning of the three-dimensional model is the core of the problem.A good processing trajectory route can greatly reduce the running time for invalid positioning in the route,improve the processing efficiency of a single component,and reduce the cost of the entire production link.In recent years,with the discovery and improvement of many swarm intelligence algorithms,more swarm intelligence algorithms are applied to combinatorial optimization problems.Since swarm intelligence algorithms can solve nonlinear and high-dimensional complex problems faster than traditional algorithms in a reasonable time,this article relies on the national key research and development program projects it participates in,and aims at the relatively weak 3D model surface in the domestic laser processing and manufacturing field.Research on processing path planning issues.In order to solve the problem of laser processing path planning for 3D model surface,this paper proposes an improved chicken swarm algorithm(ICSO).The chicken swarm algorithm is a new and efficient swarm intelligence algorithm that can converge quickly on high-dimensional and complex problems,and has the characteristics of multi-group collaborative search.It has strong global search capabilities and is used to solve various combinations in real life.Optimization.In this paper,the chicken group optimization algorithm is used to solve the traveling salesman problem,the algorithm and fitness function are re-encoded,the algorithm adaptive scheme is added,and it is applied to the 3D model surface laser processing path planning problem.The main work completed and contributions made in this paper are as follows:1.Researched the field of laser processing path planning,summarized the existing problems of existing algorithms,and compared and analyzed the application of existing algorithms in actual laser processing path planning.2.In view of the fact that the chicken flock algorithm cannot solve the combinatorial optimization problem,the algorithm's coding strategy is changed,and the algorithm's fitness function is redefined,and the position movement formula of the rooster,hen and chicken is improved.The 2-opt algorithm is introduced,which expands the range of activities of the flock and enhances the anti-interference ability of the algorithm.3.Aiming at the problem of laser processing path planning for the surface of the 3D model,the simulated annealing algorithm is added to the improved chicken flock algorithm,and the self-exploration process is added to increase the convergence ability of the algorithm.4.In order to reduce the number of initial points used in the chicken flock algorithm,the surface of the 3D model is divided into regions.The positioning points in each area are extracted as the initial points of the algorithm,which greatly reduces the number of vertices that the algorithm needs to calculate.5.All the algorithms used in this article have been tested.Eleven data sets with different dimensions are selected from the TSPLIB database to test the improved chicken flock algorithm,and compared with several other typical algorithms,it is proved that the algorithm can solve the TSP problem faster.For the application of 3D model surface laser processing path planning,several typical algorithms are used to test on different iteration times and running times,and the algorithm stability,convergence speed and actual planning effect are compared,and the results are better than other algorithms The convergence speed is faster and the path planning results are better.It shows that the improved chicken flock algorithm can be effectively applied to the 3D model surface path planning problem in laser processing.