Grid-Based UGV Navigation in a Dynamic Environment Using Neural Network

A grid environment based algorithm using neural network for an Unmanned Ground Vehicle (UGV) navigation is presented in this paper. An UGV is helpful in applications including traffic planning for smart cities, military operations, handling of goods in warehouses. The basic challenge in a cluttered environment is to find a collision free path while navigating from a starting position to a target position. Depending on neural activity of neighbouring grid cells the UGV starts moving towards its target grid cell based on the information getting from sensor. The proposed algorithm is verified on a topologically grid-based map for diverse environmental situations.