A Novel 3-D Bio-Inspired Neural Network Model for the Path Planning of An Auv in Underwater Environments
Mingzhong Yan, Daqi Zhu, Simon X. Yang · Intelligent Automation & Soft Computing · 2013
A three-dimensional (3-D) neural network model based on bio-inspired neurodynamics is proposed for the path planning of an autonomous underwater vehicle (AUV) in underwater environments. The model is topologically organized according to the 3-D underwater workspace of the AUV. Each neuron in the neural network uniquely represents a discretized subspace in the workspace. The excitatory and inhibitory inputs to the neural network come from the mission of the AUV and obstacles in the workspace respectively. The AUV is globally attracted through excitatory neural activity and the propagation among the network for its mission. Meanwhile, it is locally pushed away by the inhibitory neural activities to avoid collisions. Simulation results show that the proposed 3-D bio-inspired neural network model is suitable for an AUV to plan various paths in a 3-D underwater workspace.