Cui, Bo, Chaomin Luo, Mohan Krinisnan, and Mark Paulik
Presentably, computational intelligence (CI) algorithms have increasingly drawn great attention to their applications to Unmanned Ground Vehicles (UGV) path planning due to their advantages. In this research project, a Glasius-type biologically inspired neural network model is simulated for UGV path planning. The vehicle motion is generated through the dynamic activity landscape and neural activity propagation of the neural network. No templates, no prior knowledge of the dynamic environments, and no learning procedures are needed. An efficient Glasius biologically inspired neural networks approach is proposed for real-time map building and path planning of UGVs in a completely unknown environment. Simulation and comparison studies of the proposed approach with square-cell-map-based UGV path planning demonstrate that the proposed method is capable of planning even more reasonable and shorter collision-free UGVs trajectory in unknown environments on Player/Stage. The proposed strategy has been successfully implemented on actual autonomous vehicle.