Research On Path Planning Of A New Intelligent Cotton Picking Robot Picking Cotton Head

With more and more applications of various types of robots,the research on robots has also reached a level of in-depth.Due to the wide range of topics involved in robotics,according to the needs of different applications,a number of research areas have been developed.The research on path planning is a hot research direction.Based on the new intelligent cotton picker project as research background,in order to improve picking efficiency and optimize the picking path,this paper proposed the research on the path planning technology of picking cotton picker,and validated the proposed planning algorithm through experiments and simulations.This article mainly carried out research from the following aspects:The principles and theories of basic genetic algorithms are studied and analyzed,and the advantages and disadvantages of the basic genetic algorithms(GA)are pointed out.For their shortcomings,the common improved genetic algorithms studied by scholars are briefly analyzed,and integrated other scholars' research on genetic algorithms,summarized the main improvement directions of this algorithm in briefly.The improvement of multi-population genetic algorithm was proposed,and the improvement of the efficiency and capability of the improved algorithm in path planning was verified.this paper presented a new theory of improved innovative path planning algorithm that combines simulated annealing and multi-population genetic algorithms.Aiming at solve the problem of premature convergence of genetic algorithm in robot path planning,an improved algorithm(multi-population genetic algorithm,MPGA)was proposed.While the improvement of multi-population genetic algorithm in the optimization efficiency was not obvious,it's mainly due to the easiness of the elite population to fall into the local optimal solution.In order to improve the global optimal solution search efficiency further,the Metropolis criterion was used to innovatively jump out of the local optimal solution for the elite population,combined elite population return strategy of the multi-population genetic algorithm with the simulated annealing(SA)algorithm theory,solved the prroblem of premature convergence of genetic algorithms relatively perfect.This article elaborated the implementation process of this improved multi-population genetic algorithm,and summarized its comprehensive improvement with respect to the basic genetic algorithm.Last,the Matlab algorithm program verified the improvement of the improved algorithm for its improvement in global optimization efficiency,optimization speed,and reduced premature convergence problem.The optimization efficiency and optimization ability of the hybrid particle swarm optimization algorithm in the path planning are studied under different particle numbers.Based on the basic particle swarm algorithm,the evolutionary approach of particle evolution using information exchange with individual extremum and global extremum was combined,the crossover and mutation operations of genetic algorithm are combined to make the hybrid particle swarm optimization algorithm jump out of the local optimal solution as much as possible to the global maximum,and get global optimal solution.Afterwards,the gap for the optimization rate and the optimization ability between hybrid particle swarm optimization algorithm and the improved multi-population genetic algorithm was verified by simulation experiments.Finally,it is still decided to adopt the improved multi-population genetic algorithm in physical experiments.