NEURAL NAVIGATION APPROACH FOR INTELLIGENT AUTONOMOUS VEHICLES (IAV) IN PARTIALLY STRUCTURED ENVIRONMENTS

The use of Neural Networks (NN) is necessary to bring the behavior of Intelligent Autonomous Vehicles (IAV) near the human one in recognitio n, learning, decision-making, and action. First, current navigation ap proaches based on NN are discussed. Indeed, these current approaches r emedy insufficiencies of classical approaches related to real-time, au tonomy, and intelligence. Second, a neural navigation approach essenti ally based on pattern classification to acquire target localization an d obstacle avoidance behaviors is suggested. This approach must provid e vehicles with capability, after supervised Gradient Backpropagation learning, to recognize both six (06) target location and thirty (30) o bstacle avoidance situations using NN1 and NN2 Classifiers, respective ly. Afterwards, the decision-making and action consist of two associat ion stages, carried out by reinforcement Trial and Error learning, and their coordination using a NN3. Then, NN3 allows to decide among five (05) actions (move towards 30 degrees, move towards 60 degrees, move towards 90 degrees, move towards 120 degrees, and move towards 150 deg rees). Third, simulation results which display the ability of the neur al approach to provide IAV with capability to intelligently navigate i n partially structured environments are presented. Finally, a discussi on dealing with the suggested approach and how it relates to some othe r works is given.