| With the continuous innovation and development of science and technology,artificial intelligence,computer control and other frontier fields of breakthrough and progress,people for mobile robot technology research has also made more fruitful results.It is an important research direction of mobile robot technology to improve the efficiency of robot autonomous movement and effectively avoid complex obstacles in the working environment.Among them,map optimization and path planning algorithm optimization in simulation environment are very important to solve the above problems.Therefore,this topic focuses on the two hot spots and difficulties of autonomous mobile robots,and the specific research content is as follows:1).Map optimization.Aiming at the problems that the traditional raster map model builds the robot working environment information is relatively simple,the robot is not accurate enough to detect the obstacle information,the planned path is long,and the quality is poor,this paper proposes a map optimization algorithm based on triangular mesh to improve the raster map model.Improved model is the core idea of grid map is a grid map of obstacles as reference point of constructing triangular mesh,the improved model maps in the square on the grid and triangular grid points can be used as a node traverse robots,the autonomous mobile robot can traverse the nodes are added at work,so as to realize accurate obstacle avoidance.The experimental results show that the path length generated by path planning on triangular mesh improved raster map is reduced by about 12.22%-22.27%,and the path planning time is reduced by about 2.01%-3.68% compared with that on traditional raster map.2).Path planning.Aiming at the problems of long path length and more traversal time in the path planning process of mobile robot using A* algorithm,this paper makes improvements.The specific research contents are as follows:(1)In order to improve the problem of sharp inflection point and long path in the path planned by A* algorithm,the path planning algorithm based on Bessel modified A* algorithm is studied.Firstly,the traditional algorithm is used to obtain the initial path,and then the sharp inflection point in the initial path is taken as the reference point of the improved algorithm.Considering the turning radius and minimum lateral distance of the robot,the second order Bessel curve is used to optimize the local path,and finally the overall planning path is obtained.The experimental results show that the path length planned by the improved algorithm is reduced by about 13.45%-15.58%on the raster map and about 7.18%-10.48% on the improved raster map,which eliminates the sharp inflection point in the path and makes the path smoother.(2)In order to improve the problem that the path planning of A* algorithm takes too much time to traverse,a path planning algorithm based on the improved average cumulative deviation of A* algorithm is proposed.Firstly,the mean and variance of the components were normalized according to the different distributions of the components in each dimension of the dataset.Then,on the basis of standardization,the average cumulative deviation coefficient of the original algorithm is adjusted,so as to eliminate the influence of data sets on the degree of variance of the calculated data,and finally realize the reduction of the number of target nodes and search time.Through the debugging of the algorithm,the optimal solution is finally selected when the value is 3,and the path planning time of the improved algorithm is reduced by about 5.07%-6.24%compared with the original algorithm.Figure 34 Table 5 Reference 83... |
PDF Full Text Request