A GWO-based multi-robot cooperation method for target searching in unknown environments

Abstract

To solve static and dynamic target searching problems involving multiple robots in unknown environments, a novel adaptive robotic grey wolf optimizer (GWO) algorithm, named the RGWO, is proposed. First, an optimal learning strategy is introduced to improve the position updating formula of the GWO to make the algorithm suitable for use in actual mobile situations involving robots, allowing the searching robots to move towards the target (prey) in a step-by-step manner. Then, an adaptive inertial weighting scheme is adopted. By increasing the “aggregation degree” or decreasing the “evolution speed”, the influence of the inertial weight can be increased, which is helpful for maintaining the search diversity of the robots and avoiding premature convergence. In addition, due to the ability of the prey to escape, the pursuing robots are likely to fall into local optima. To avoid this issue, an adaptive speed adjustment strategy and an escape mechanism are adopted. The RGWO is verified and compared with other methods. The RGWO has obvious advantages over other methods in terms of the number of required iterations, success rate and efficiency, and it is superior in both static and dynamic target searching. However, the search trajectories generated with the RGWO are not smoother than those generated with the other investigated methods.