Research On Multitask Assignment And Path Planning Of Multi Robots

In recent years,the robot is one of the most promissing fields of human development,it has been more and more widely used in industry,agriculture,services and other fields.However,with the continuous development of robot technology,the human demand for robot has also shifted from single robot to multi-robot system,with the deeping research of multi-robot,the multi-robot path planning has become a topics in the fields of robotics.The key of multi-robot path planning problem is to assign the task points in the environment that with multi-robot and multi-task to each robot according to a reasonable allocation strategy and each robot needs to find a shortest path without repetition nor missing of its assigned tasks.Based on this,in a system with multi-robot and multi-task,the tasks in the environment need to be assigned to the robot firstly,and then each robot performs path planning according to its assigned tasks.In the problem of multi-robot system path planning,it is necessary to assign the tasks in the environment to the individual robots in the environment firstly.This paper assigns tasks according to the specific capabilities of each robot,taking into account the power and speed of the robot.When task allocation,the task is assigned to the robot whose cost is the minium,and the formula of caculating cost takes full consideration to the speed and charge of robot,this paper assigns task points in the environment to robots with the least amount of charge and the shortest time.The robot path planning problem is to traversal all tasks in the environment based on some optimization criteria,which is difficult to solve the optimal path.In this paper,the immune genetic algorithm is used to solve the problem because it has both the global searching ability of genetic algorithm and the concentration factor of the immune algorithm.In the later evolution,individuals with high fitness will be extracted to form elite antibody groups.However,the traditional algorithm in the convergence rate and antibody diversity is insufficient,for these problems,the traditional algorithm has been improved in this paper.In the initial antibody group,the nearest neighbor algorithm is used to generate the initial solution,which greatly improves the convergence speed of the algorithm and avoids the algorithm falling into the local optimum.In addition,this paper combines the traditional antibody similarity with the path structure,which improves the population diversity,the convergence rate of the algorithm and finds a better solution.This paper establishes a simulation environment for this problem,which ensures that the location,speed,charge and others of robots or tasks are simulated real environment.The tasks assignment method based on the robot specific capabilities and the robot path planning based immune genetic algorithm experimented and analyzed in the simulation environment.The experimental results show that the above method can solve the problem of multi robot multi task path planning.