Research On Path Planning Method For The Equipment Camouflage Painting Robot

Pattern painting camouflage is one of the most simple and effective measures of equipment protection.In recent years,with the rapid development of spray painting robots,camouflage patterns of equipment have gradually changed from manual painting to automatic painting by robots.The path planning of spray painting robots is difficult and crucial to the realization of automatic painting of equipment camouflage.It has the most direct impact on the efficiency and quality of equipment camouflage operation,and then affects the timeliness and camouflage effects of military equipment combat support.At present,there are few researches on the path planning of spray painting robots for camouflage pattern,painting paths are generated by human experience or rule strategies mostly,which result in problems such as redundant paths,coating thickness inconsistent and low planning efficiency,and affect the actual efficiency and quality of robot painting ultimately.Therefore,we research on the path planning method for the equipment camouflage robot in this thesis.The research and results are as follows:Firstly,the camouflage pattern robot spraying model is established in this paper.The model is mainly used to provide spraying process parameters such as spraying speed,spraying distance and spraying width for the research of path planning method of camouflage pattern.This paper is aimed at the special spraying equipment for camouflage robot spraying,combined with a variety of common spraying models,through experimental data and the characteristics of the spray gun's own shape,the digital camouflage elliptical spray gun uses the elliptical double beta distribution model to modeling the coating thickness,speckle camouflage round spray gun uses the beta distribution model to modeling the coating thickness.According to the coating deposition rate,the coating thickness distribution function and spray gun spraying flow function with gamma function are derived,which solved the problem that the formula was too complicated to be solved.Spraying experiments were designed to measure the coating thickness data,and then the parameters of the spraying model were fitted based on the least square method,and the fitted parameters were verified in practice.The results show that the curve average error of the fitted digital camouflage spraying model is about 5%,and the curve average error of the fitted speckle camouflage spraying model is about 4%.Secondly,the path planning method for digital camouflage pattern painting is researched.According to grid shape characteristics of digital camouflage pattern,grid method is used to model the environment of digital camouflage pattern,and then the camouflage pattern is divided into several local zones by the domain segmentation with color block aggregation.The path planning problem is divided into local zone path planning and path planning among the local zones.The local zone path is planned by Zig Zag path.And the path planning among the local zones is equivalent to the special traveling salesman problem,the we propose an improved genetic algorithm for the local zones' path planing.Compared with the traditional path by sequential regular rules,the length of our optimized path is shortened by 37.0%,as a result the spray painting efficiency will be improved obviously.Thirdly,the path planning method for spot camouflage patterns painting is researched.The spot camouflage pattern path consists of the local path of an individual spot unit(including the internal fill path and the boundary path)and the path among spot units.We propose an internal fill path planning method that adaptively adjusts the width of the internal filling path of the spot unit for spot camouflage pattern based on spraying model parameters,which solves the problem that the traditional fixed internal filling path width cannot be flexibly adjusted,which leads to the problem that the last spraying area is too narrow,the coating thickness will be not uniform if an additional gun is sprayed,if not the bottom will be exposed.The improved genetic algorithm previously proposed by the digital camouflage pattern is also used to optimize the path among spot units.Compared with the traditional path,the length of our optimized path is shortened by 40.3%.Finally,the actual effect of the path generated by our proposed algorithm is verified by using the self-developed off-line programming software and ABB Robot Studio software.The results show from simulation,compared with the traditional path,the actual running time of the optimized path of digital camouflage pattern can be reduced by 16.3%.While the actual running time of the optimized path of the spot camouflage pattern can be reduced by 17.0%.