Mobile robot trajectory planning using enhanced artificial bee colony optimization algorithm

Abstract

An Enhancement of Artificial Bee Colony (ABC) algorithm, reflecting the multi-objective features of robot trajectory planning problems, is designed and simulated results are presented. Even though the basic ABC algorithm used in mobile robot can obtain a quick absolute trajectory, it still has its own drawbacks. This bio-inspired meta-heuristic algorithm, which emulates the bee's behavior of foraging around their bee-hives, is used to obtain the optimal trajectory from a source position to a destination position. The trajectories of the robots' motion from predetermined source points to target points in the prescribed scenarios are evaluated with the eventual motive to acquire collision-free trajectories and to reduce the trajectory lengths of all the motion of robots in the scenario. The trajectories that are to be developed locally for ‘n’ robots are adequately small with minimal spacing with the obstacles, if any, in the prescribed scenario. Based on the traditional Artificial Bee Colony algorithm, another objective function for each and every robot's trajectory is inferred for global trajectory planning. The fitness function and initialization strategy are also optimized to enhance the capabilities of the proposed algorithm. Simulation results of the proposed Enhanced ABC algorithm are compared with the basic ABC algorithm. It reveals that the proposed optimization algorithm performs better than the basic ABC algorithm in terms of trajectory distance and path deviations and also shows that the proposed algorithm is effective for use in real-time trajectory planning of mobile robots.