Based On Improved Ant Colony Algorithm For Mobile Robot Path Planning And Implementation

Path planning is one of the most fundamental and most important tasks in mobile robots. It shows the robot's exchanging ability with around environment. Ant Colony System (ACS) is a novel bio-inspired optimization algorithm, which simulates the foraging behavior of ants. Path planning has an essential relation with the behavior of real ants searching for food, so it is logical to apply ACS to path planning.In this thesis, ACS is improved as follows:Firstly, the improved ACS adjusts the parameters q0 dynamically to adapt to different requirements, which improved the length of path and the speed of convergence.Secondly, the improved ACS changes the value of pheromone by using opposition-based learning to increase the search scope and decrease the search time.Thirdly, the improved ACS changes the pheromone updation rule, i.e., adjusts the volatile parameters of the pheromone dynamically, limits the value of pheromone betweenτmin andτmax, and sets the best local circulation value increased and the worst local circulation value decreased at the end of the local circulation.In modeling and planning, a mobile robot is considered as a non-full particle to deal with the obstacles, and a new evaluation criterion is proposed so that the planning is not only concerned about the path length and convergence rate, but also concerned about the actual movement of turning number of the robot.Dynamic programming is carried out on the basis of the hybrid architecture for mobile robot. The local path planning is based on the global path planning. During the movement, the robot detects the real-time local information by using the mounted sensors, and predicts the dynamic obstacle position by using straight-line method.Simulation and actual experimental results show that the improved ACS has better results in the path length and planning time.