Research On Path Planning Of Multi-task Heterogeneous Robot Group

In the field task of robot,the robot path planning directly determines the efficiency of task implementation.With the continuous improvement of field operation demand,the robot operation mode has gradually changed from single robot operation to multi-robot operation,and the task type has gradually changed from single task to multi-task.In multi-task job for multi-robot mode,the traditional algorithm has been difficult to meet the requirements of optimal path planning.In recent years,more and more researchers use swarm intelligence optimization algorithm to plan multi-robot multi-task path,and have made outstanding achievements.However,the performance of swarm intelligence optimization algorithm is affected by parameter design and environmental factors.In this paper,operator design,parameter improvement and hybrid algorithm are studied to solve the above problems.Firstly,the path planning problem of swarm intelligence algorithm is studied.Aiming at the shortcomings of traditional algorithm in path planning,such as easy to fall into local turbulence and solving non-global optimal solution,the traditional algorithm was improved from two aspects of parameter optimization design and multi-algorithm fusion.To improve particle swarm optimization algorithm for search performance,by adjusting the weighting coefficient of inertia of improvement strategies.For the problem that the ant colony algorithm has too fast convergence speed in the early stage of searching,an improved strategy based on segmented algorithm is adopted.Secondly,taking the shortest path value as the optimization goal,the multi-task problem solving' model of single robot is established.In order to solve the multi-task problem in obstacle environment,genetic algorithm has the disadvantages of slow evolutionary convergence and easy to fall into local optimum in the later stage of the algorithm,an improved genetic algorithm is proposed.In order to reduce the complexity of the algorithm,the grid method is used to model the environment map,and the grid ordinal coding method is selected to reduce the complexity of the algorithm.After the selection,crossover and mutation operators are finished,the acceptance criterion of simulated annealing algorithm is integrated to judge whether to accept the new solution and improve the local optimization ability of the algorithm.The simulation results show that the proposed improved algorithm shortens the planned path value by 9%compared with the traditional genetic algorithm,and reduces the number of convergence iterations by 67%,which proves the effectiveness and feasibility of the improved algorithm.Finally,an adaptive improved genetic algorithm based on the fusion of two algorithms is proposed to solve the problem of multi-robot multi-task.The robot operating environment was modeled by grid method,and the grid number was coded as real number.The crossover and mutation operations of the genetic algorithm are adjusted dynamically.After the completion of the genetic operator operation,the acceptance criterion is incorporated to judge the acceptance probability.Then smooth the planned path.The simulation results show that compared with the traditional genetic algorithm,the proposed algorithm has faster speed and stronger global optimization ability for solving the multi-task point multi-robot problem.The average path value of the robot is reduced by 5%,and the number of converging iterations is reduced by 86%.