Hawks Swarm Optimization And Applications Of Path Planning On The Surfaces Of Cubes

With the development of robotics and artificial intelligence,single robots are unable to complete increasingly complex and scaled engineering tasks,and the scenarios in which robots move are becoming more and more complex.The path planning problem on the surfaces of cubes for multiple intelligences has received increasing attention from researchers and practitioners in related industries and has become one of the important applied research topics.Swarm intelligence algorithms are one of the common methods for solving global path-planning problems.The Harris Hawk Swarm Optimisation algorithm is a powerful new population intelligence algorithm,so it has some research value to apply the Harris Hawk Swarm Optimisation algorithm to solve the three-dimensional path planning problem of an intelligent body.In this paper,the Harris Hawk Swarm Optimisation algorithm and the application to path planning problems on the surfaces of cubes are studied in depth,and the following three main aspects are carried out:(1)A hybrid hawk swarm optimization algorithm with a dynamic backward learning strategy is proposed.The original Harris Hawk Swarm Optimisation algorithm tends to fall into a local optimum prematurely and is sensitive to the initial value of the search population.The algorithm proposed in this chapter first uses a dynamic opposite learning mechanism to enhance population diversity and population quality,then incorporates the vortex effect mechanism of the marine predator algorithm in order to enhance the global search capability and search accuracy,and finally uses a non-linear escape energy parameter to increase the global search in the early stage and the local search in the later stage to prevent the search of the hawk swarm algorithm from ’premature ’.With 20 test functions,the improved algorithm’s search capability was verified,effectively improving the convergence speed of the algorithm,reducing premature convergence,and enhancing the optimization performance of the algorithm.(2)On the basis of the proposed hybrid hawk swarm optimization algorithm,a modified hawk swarm algorithm based on greedy neighborhood search is proposed.The original Harris hawk swarm optimization algorithm is not suitable for solving discrete or combinatorial optimization problems.In order to enhance the initial solution quality of the path planning problem on the surfaces of cubes,the k-means clustering method is used to initialize the population in conjunction with the characteristics of the three-dimensional space.Finally,a neighborhood search operator with a greedy strategy is added to the local search strategy in order to improve the late convergence speed of the algorithm.The ability of the algorithm is verified on a variety of 2D test maps and random 3D maps,and the comparison with other classical swarm intelligence algorithms shows the superior performance of the algorithm for path planning.(3)The path planning on the surfaces of the cubes simulation platform designed and built provides visualization capabilities,including visualization of experimental results of path planning simulations and visualization of results related to testing of basic test functions.The simulation platform provides a convenient tool for researchers to study the use of population intelligence algorithms to solve path-planning problems on the surfaces of cubes.