Particle Swarm Optimization of Non Uniform Rational B-Splines for Robot Manipulators Path Planning

Abstract

Keywords robot manipulators, path planning, B-splines, particle swarm optimization IntroductionIn recent years as the technology advanced the use of robots become common and human life become easier. Months or even years needed to cover long distances on feet or using animals have been reduced to hours using planes. Planets which appear in nights, and seen only by eyes has been explored by human using planetary robots. Under sea that represent puzzles for human-being have been excavated using under-water robots. Complex hard works that need many workers and long period to be accomplished have become short time tasks through exploiting industrial robots. Chirurgical operations that need the presence of doctors and nurses are simplified by taking advantage of service robots.All kinds of robots whether they are mobile or manipulators operate in unknown environments or in environments that changed continuously. Therefore, they require an important step to accomplish their tasks even though simple or complex. This step consists of generating a path that allows the robots to navigate through, without colliding with any of the surrounding obstacles. Besides, the path should start form an initial position and reach the final one. The main aim behind path planning for mobile robots is to make them capable of reacting to any new situation they face, thus increase their autonomies, whereas for manipulators path planning is necessary for high speed, high precision and consequently for increasing productivity and safeness.Generally, existing approaches for solving this problem of path planning can be classified into two classes: conventional and meta-heuristic. On the one hand, the conventional class includes bug algorithms, potential field [1], road map [2,3], cell decomposition and sampling based algorithms [4]. It depends on complex mathematical model, and suffers from common drawbacks such as the limitation to simple two-dimension space, local minima, incompleteness and high computational time, produces long and rough paths resulting from a compilation of straight line which cannot be executed by the robot. On the other hand, the meta-heuristic class that group neural network [5,6], fuzzy logic [7], evolutionary algorithms [8] (i.e., genetic algorithm, genetic programming, evolutionary programming and evolution strategy), ant colony [9] and particle