Improved genetic algorithm for mobile robot path planning in static environments

Abstract

The genetic algorithm (GA) is a well-known meta-heuristic technique for addressing the static mobile robot global path planning (MRGPP) issue. Current GA, however, has certain shortcomings, such as inefficient population initialization and low-quality solutions. As an enhanced GA, a Linear Rank-based, or Clearance-based Probabilistic Road Map (CBPRM), technique is proffered to overcome these difficulties. The new model guides the population initialization process by using the fitness score of each cell in the environment, lowering the number of infeasible pathways created. Furthermore, a genetic operator combination is proposed to balance the global and local search and increase the quality of the optimum path created in terms of path length and safety. Two experiments were carried out to assess the suggested GA. The novel population initialization strategy was compared to two current models in the first experiment, and the findings revealed that the suggested approach greatly decreases the number of infeasible pathways created and the time required for the process. The ideal genetic operator combination was determined in the second experiment, and the results revealed that the suggested combination improves the quality of the optimal path created in fewer iterations. In summary, the proposed GA improves on previous models by proposing a novel population initialization method and combining numerous genetic operators. These alterations improve the quality of the optimum path and indicate the suggested model’s potential for solving the MRGPP challenge.