Path Planning Of Intelligent Mobile Robot Based On Ant Colony Algorithm

Mobile robot is a integrate system that research began in the late1960s, as one of themost magnificent inventions of mankind in the history of the20th century, has undergone atremendous change within the extremely short-term50years. In the resent years, the robot hasbecome the most representative of the strategic objectives in the high-tech field goal, mobilerobots will be everywhere. Swarm intelligence, a recent rise of a new type of intelligentcomputing method for solving NP problems, due to the application of the method is very wide,and therefore subject to the attention of scholars from various countries."Swarm Intelligence"argues that human intelligence derives from the interactions of individuals in a social worldand further, that this model of intelligence can be effectively applied to artificially intelligentsystems. Swarm intelligence has so many advantages, such as distributivity, collaboration,robustness, rapidity, and so, these advantages of the engineering problems solve a series of theproblem that difficult to solve in complex engineering optimization problems. New swarmintelligence, including particle swarm, fish stocks, and ant colony, the ant colony optimization(ACO) is an emerging intelligent optimization algorithm which is widespread concern andresearch by a large number of experts and scholars, it has good positive feedback and strongrobustness, so it is widely used in machine learning, dynamic environment optimization,function optimization, neural network and so on. This paper mainly based on theimprovement of ant colony algorithm to complete path planning of the robot.This subject systematic introduced the formation of the basic ant colony algorithm theoryand mathematical model, in-depth analysis of based on the ant colony optimization algorithmshortcomings and the formation of the causes, and improves the ant colony optimizationalgorithm using rough set theory that improve the speed and accuracy of the convergence ofthe algorithm in robot path planning. This improved fusion algorithm to overcome effectivelythe shortcomings of the algorithm is as follows: easy to fall into local optimal, the slowersolution search time, the worse search results.The following is a major work of this paper:In the first, this article describes a brief history of robot development, definition andmobile robotics, and then systematic summarizes the research status, research methods, keytechnologies of the robot path planning at home and abroad, analyses their respectiveadvantages and disadvantages, laid an important theoretical foundation for further study，andhighlight the problem of global path planning and environmental modeling problems as belowthe theoretical basis. Secondly, the principle, the model, the characteristics and the management about thebasic algorithm of ant colony optimization are also presented, and then this paper analyses thedisadvantages of the algorithm, and proposes improving measures. Ant colony algorithmcombined with rough set algorithm to enhance the global optimization capability androbustness of the algorithm to improve robot path planning. Using rough set optimizationalgorithm to find a collection of libraries of initial feasible path, and then search using ACOand rough set. This will be handled both, not only overcome the ACO algorithm vulnerable toinadequate local optimal solutions, but also can help solve the problem of rough set algorithmcan only find the approximate optimal solution. The algorithm improves the search speed，andalso to avoid the ACO select the repeatability of the initial value.After that, in the model established by this article, the feasibility of the grid eightdirections to establish the initial decision-making table, and then using the rough set theoryalgorithm to simplify the decision-making table, which rough sets can be trained a series ofpaths and the path for a feasible path, this method reduces the size of the hunting zone of theant colony algorithm. Next to the use of Ant Colony algorithm for solving optimal path,inspiring information on probability of selection, used as heuristic information for importanceof rough set feature, makes the ants select the path more quickly, in addition, the pheromonestrategy also joined rough set attribute dependence and attribute importance updated strategy,compared with the traditional method of path planning, this method searches the efficiencyfaster and performance better. Through Matlab software tests the method, compare the twomethods, initial of related data parameter for simulation, the simulation results show that thismethod improves the robot path planning speed and planning capacity, basic ant colonyalgorithm based on rough set ant colony algorithm application effect is better than the effectof robot path planning, also shows that the rough set and the combination of the ant colonyalgorithm has some theoretical significance and certain feasibility.Finally, we get a comprehensive summary of the paper work done, and give the outlookfor the future need to study the problem of robot path planning.