Ameliorated equilibrium optimizer with application in smooth path planning oriented unmanned ground vehicle

To enhance the performance of equilibrium optimizer (EO) and expand its application for smooth path planning of the unmanned ground vehicle (UGV), an ameliorated equilibrium optimizer (AEO) is developed and applied to the UGV smooth path planning problem. The main characteristics of AEO are the population initialization by an opposition-based learning (OBL) strategy, the concentration updating by a centroid opposition-based learning (COBL) strategy, and the concentration updating by a proposed self-learning strategy. The three improvement strategies in the AEO enhance the optimization performance of EO through utilizing the information of the opposite space, the neighborhood space, and the whole population. The performance of AEO is examined by comparing it with several well-known algorithms on 29 commonly used benchmark functions. The comparison results show that the AEO is superior to the compared algorithms and ranks first in performance evaluation. Furthermore, a smooth path planning method AEO-HB is proposed by optimizing the control points from high-order Bezier curve based on the AEO. Simulation experiment results manifest that the AEO-HB solves the smooth path planning problem and ranks first among the compared algorithms for the performance evaluation in three different cases. The above results of numerical experiments and smooth path planning experiments indicate that the proposed improvement strategies in the AEO enhance the performance of solving global optimization problems, which makes the AEO have the potential to deal with global optimization problems in more types of application scenarios.