Route Planning Base On Ant Colony Optimization And Panicle Swarm Optimization

Path planning technology is a hot spot of many areas at present. It has broadapplication prospects and scientific value.And the research of the path planning algorithm is the core of it. The main methods which solving the path planning probleminclude A*algorithm, particle swarm optimization algorithm, genetic algorithm, antcolony optimization algorithm and so on.Particle swarm optimization algorithm and ant colony optimization algorithm arebionic algorithms, which mocks the process of flock foraging and the process whenants seeking food. In recent years, Both of algorithms are more and more applied,path planning has proved its good application performance. Particle swarmoptimization algorithm is a new kind of random search algorithm, has a strong globalsearch ability and high efficiency and concision, But easy to fall into local optimum.Ant colony optimization algorithm has features of parallelizability， positivefeedback，high accuracy and Low efficiency of the convergence.According to theabove problem, specific improvement strategy is as follows:(1) An enhanced ant colony optimization algorithm is proposed to plan pathglobally for mobile robots under complicicated environments. The search deadlock isruled out through modifying both initial environment pheromone and state transitionprobability. The roundabout of trajectory is improved by combining determinatedsearch with stochastic search. Trajectory smooth operation is utilized toeliminate redundant waypoint for maintaining better consistency of path. Localpheromone diffusion mechanism is introduced to advance global optimizationperformance. Experimental results show that the new algorithm can provide desirablesafety path under the complex situation, which contains dense obstacles or manyconcave blocks, as well as computational time can meet the requirement of particleapplications.(2) The new algorithm generates the distribution of the initial information forACO by using the merits of high efficiency and concision of PSO，and then uses theadvantages of parallelizability，positive feedback and solution with high accuracy ofACO to get global optimum solution． Because of the characteristics of the multi-threaded. Path planning simulationplatform is developed using visual C++6.0environment. Three algorithm such asPSO、EACO、PSO-EACO presented in this article are implemented and tested,theperformance are compared each other sequentially.Experimental results show that thePSO-EACO algorithm performance is better than EACO in time. Better than ACO inthe quality of the solution. Show the effectiveness of the strategy and practical.