GLASIUS BIO-INSPIRED NEURAL NETWORK ALGORITHM-BASED SUBSTATION INSPECTION ROBOT DYNAMIC PATH PLANNING

This study presents a glasius bio-inspired neural network (GBNN) algorithm for intelligent substation inspection robot autonomous path planning. First, a GBNN Neural map is established to represent the working environment of the inspection robot. In this model, each neuron corresponds to a grid map position unit. The GBNN model was trained to map the environment, including obstacles and potential paths, into a discrete neural network representation. Second, the motion path of the inspection robot was planned autonomously based on the activation output values of the neurons in the neural network. The robot selected the path with the highest activation output value for the next movement direction. The simulation results under dynamic obstacle scenarios or in uncertain environments demonstrated the effectiveness of the GBNN algorithm in path planning.