LWSGA-MO

Evolutionary algorithms (EAs) have been used to solve multi-objective optimisation problems in various applicationsincluding industrial and simulated environments. EAs can solve several objective functions in the simulated environment.Although numerous studies utilise evolutionary algorithms to evaluate simulated environments, there has been littleresearch on evaluating a multi-objective optimisation algorithm in the industrial and space-simulated environment. Inthis paper, we introduce a new Evolutionary Multi-Objective (EMO) optimisation algorithm that is applied in one ofthe most significant simulated-environment research areas such as NASA. More specifically, this paper presents a newalgorithm called Limited Weighted Sum Genetic Algorithm for Multi-Objectives optimisation (LWSGA-MO) which is amodified WBGA that allows weights to be limited or prioritised when an upper limit is unknown. The new algorithmLWSGA is used to evaluate the effectiveness of multi-objective optimisation when designing Martian rovers by usinggenetic algorithms within a simulated environment. The proposed evolutionary algorithm is used to find a solution set ofMartian rovers in the simulated environment that are lightweight, cost effective and have the ability to move successfullyin the simulated terrain. The results of the LWSGA-MO show that changing the rate of the mutation, implementing anew genetic operator bias, and increasing the selection range; all can lead to finding an optimal solution in the simulatedenvironments. In particular, in the new algorithm increasing the rate of the mutation can be beneficial to achieve apopulation of fit and diverse optimal solutions. It is also found that, by amending the crossover pairing with the best pairfittest solutions from each generation, it will achieve an optimal solution in the final population. Moreover, adding manynew solutions in each generation is likely to produce diverse solutions for fitness functions in the final population.