Robot Path Planning Based On Improved Artificial Fish Swarm Algorithm

Robot is a closer and huge potential technology. It is an important part of robot and artificial intelligent. Robot has a broad development space in many fields such as aerospace, industrial manufacture, service industries, agriculture and so on. The complexity and uncertainty of the environment when robot moves, determines the path planning is one of its research emphasis and difficulty. The path planning means that find out a collision-free path from the start to the end in the known or unknown environment.Intelligent bionic algorithm is a kind of intelligent compute method for simulating the biological behavior. Intelligent bionic algorithm are concerning to the experts and scholars because it has parallel, fast and robustness advantage.Intelligent bionic algorithm has their own characteristics, by applying the advantages and disadvantages that we can use them to solve many complex problems. The newer intelligent bionic algorithms are artificial bee colony algorithm, cast swarm and Artificial Fish Swarm Algorithm(AFSA) which we will use in this subject. Artificial fish swarm algorithm is getting eyes in recent years. It has strong robustness and efficient convergence rate so that it is widely used in various fields. This paper mainly through improves AFSA to complete robot path planning. First it systematically introduces the principle and ideological in artificial fish swarm algorithm, and deeply analyzes its pros and cons. We design a new fitness function to effectively improve the comprehensive path. And at the same time we add the directional operator to accelerate the speed of path finding and solve the shortcoming that the search algorithm efficiency is reduced in the late part. We increase immune algorithm to enhance the global search ability to overcome the disadvantages of AFSA which is easy to fall into local optima. Finally we experiment the Immune Directional Artificial Fish Swarm Algorithm(IDAFSA).The experiment was built by language C++, through visual studio software we respectively experiment immune directional artificial fish swarm algorithm and basic artificial fish swarm algorithm. The results show that the immune directional artificial fish swarm algorithm can effectively improve the performance and efficiency. And results prove that immune directional artificial fish swarm algorithm is better then artificial fish swarm algorithm in robot path planning. And it can solve the shortcoming well. Finally we make a comprehensive summary of this paper, and we make a prospect about robot path planning in the future.