Path planning is one of the most important tasks in mobile robots. This task is computationally intensive, and it is intensified by the complexity of the environment. This work presents the membrane pseudo-bacterial potential field algorithm with GPU (graphics processing unit) acceleration for mobile robot path planning. This proposal includes the combination of membrane computing, the pseudo-bacterial genetic algorithm, and the artificial potential field method. In this work, we will focus on testing the proposed algorithm in four test environments with distinctive challenges for each one. These experiments are carried out to test the membrane pseudo-bacterial potential field algorithm in terms of path length and computation time. We show how data-intensive tasks in mobile robots can be processed efficiently using the GPU through a parallel computing implementation, in which different paths are concurrently calculated to select the best one. Experiments and simulation results are provided to show the effectiveness of the proposal in computational performance achieving a factor above \(8\times \) on CPU (central processing unit) and above \(133\times \) on GPU.