A Spider Monkey Optimization Based on Beta-Hill Climbing Optimizer for Unmanned Combat Aerial Vehicle (UCAV)

Abstract

Unmanned Combat Aerial Vehicle (UCAV) path planning is a challenging optimization problem that seeks the optimal or near-optimal flight path for military operations. The problem is further complicated by the need to operate in a complex battlefield environment with minimal military risk and fewer constraints. To address these challenges, highly sophisticated control methods are required, and Swarm Intelligence (SI) algorithms have proven to be one of the most effective approaches. In this context, a study has been conducted to improve the existing Spider Monkey Optimization (SMO) algorithm by integrating a new explorative local search algorithm called Beta-Hill Climbing Optimizer (BHC) into the three main phases of SMO. The result is a novel SMO variant called SMOBHC, which offers improved performance in terms of intensification, exploration, avoiding local minima, and convergence speed. Specifically, BHC is integrated into the main SMO algorithmic structure for three purposes: to improve the new Spider Monkey solution generated in the SMO Local Leader Phase (LLP), to enhance the new Spider Monkey solution produced in the SMO Global Leader Phase (GLP), and to update the positions of all Local Leader members of each local group under a specific condition in the SMO Local Leader Decision (LLD) phase. To demonstrate the effectiveness of the proposed algorithm, SMOBHC is applied to UCAV path planning in 2D space on three different complex battlefields with ten, thirty, and twenty randomly distributed threats under various conditions. Experimental results show that SMOBHC outperforms the original SMO algorithm and a large set of twenty-six powerful and recent evolutionary algorithms. The proposed method shows better results in terms of the best, worst, mean, and standard deviation outcomes obtained from twenty independent runs on small-scale (D = 30), medium-scale (D = 60), and large-scale (D = 90) battlefields. Statistically, SMOBHC performs better on the three battlefields, except in the case of SMO, where there is no significant difference between them. Overall, the proposed SMO variant significantly improves the obstacle avoidance capability of the SMO algorithm and enhances the stability of the final results. The study provides an effective approach to UCAV path planning that can be useful in military operations with complex battlefield environments.