Robot Path Planning Based On Artificial Colony Bee Algorithms

In order to improve the speed of the robot path planning in the global static environment, a new robot path planning algorithm based on artificial bee colony is proposed. The algorithm consists of three kind of bees:scout bee、follow bee and leader bee. Each kind of bee corresponding to different search strategies. In order to improve environmental adaptation and searching diversity of the algorithm, the scout bee uses strategy of large step search to find the optimal goal (flower source) quickly. In order to obtain a higher search speed and converge to the optimal path, the algorithm uses the following bee for the local path search and adopts the closest distance strategy to select next node. A lot of simulation experiments show that the algorithm can plan a globally optimal or approximate optimal path in different environment and is faster, more efficient than the similar planning algorithm.In order to solve the problem of robot path planning under more complex environments, such as environment containing continuous obstacle, an improved algorithm based on the above algorithm is proposed. The improved algorithm uses strategy of the short step scout bee to make full use of the information of the environment. To reduce the possibility of robot trap into obstacles, this algorithm evaluates candidate nodes to select node which has the lowest risk as the next node, according to the distribution information of the obstacles detected in each feasible direction. Simulation results show that the algorithm has higher convergence speed and stronger environmental adaptability, and the algorithm can plan a collision avoidance of optimized path rapidly in the static environment with complex obstacles.In order to solve the problem of robot path planning in unknown dynamic environment, a rolling collision avoidance algorithm based on artificial bee colony algorithm for robot path is proposed. The algorithm is mainly divided into two parts: in the first part, the robot detect the surrounding information in real time and take the rolling optimization strategy to plan path in an unknown environment; in the second part the robot avoid dynamic obstacles and plan the path at the same time. First the robot determines if there is dynamic obstacle appear in the current vision domain, and forecast the collision. To avoid the risk of collision, a strategy of dynamic obstacle avoidance and rules for collision avoidance of special circumstances are proposed. Simulation experiments show that the algorithm can avoid the dynamic、static obstacles in the environment and go along an approximate optimal path to reach the destination.