Research On The Path Planning Of Agricultural Information Collecting Robot Based On Ant Colony Algorithm

With the rapid development of robotics,robots has played a significant role in manufacturing industry,medical treatment service and military areas.And our government is also speeding up the mechanization of our agriculture,and pursuit of the production mode that of low labor cost and high quality.This requires a mobile robot can find the optimal path itself to get to the work place and finish the task,so that to meet the requirements of agricultural production.The route planning involves several important factors such as maps building,obstacle avoidance and how to rapidly get optimum solution and so on.So it demands a broad resource management system for calculating walking route calling the algorithm,saving date and performing the task.This paper explains what is Cloud Computing and,determine the elements can make the robot cooperate with it,so that to complete robot scheduling in the cloud computing.In the context of cloud computing to bulid the cloud operation platform system about the agricultural information collection robot,to achieve the front-robot and end-cloud system.It is turn to analysis how to plan the optimal path for the cloud robot shuttling it's way through a field.This paper presented a route planning system based on ant colony system for optimization(ACO).By analyzing the mathematical model and basic concepts of existing ant colony algorithm(ACA),it is clearly that this kind of algorithm has several defects in path planning such as the usual phenomenon of premature convergence and local solution.So it need to be optimized.Firstly,introduce a new strategy into existing ACA by combining random selection strategy and perturbation strategy,and to effect the probability of the path selection a disturbance factor ? is cited;then improve the mode of updating the pheromone by using at the optimal solution.After all these work,the optimization has been finished.But the improved ACA is not good enough.So it should be combined with particle swarm optimization(PSO).By analysis of the characteristics of PSO,a inertia weight coefficient w is needed to bring in improve global and local searching ability.And the values of w should be decreased to the lower limit linearly.Then finding the junction where the improved ACA meets the PSO to solve blindness searching and pulling through the oscillation slowly that questions only come out when just using ACA.Meanwhile,it also increases the diversity of path selection.After the algorithm setup,the MATLAB is used to simulate and compare the two algorithms which are before coalescing and after coalescing,through the simulation of the grid model.And the significance and value of the proposed algorithm are verified.Through the connection building of the cloud platform,the robot and the PC,According to the data transmission of the background cloud and the way of the algorithm call,the front-end robot can provide the medium for the field path planning task.