Neuro-fuzzy expert system E-S-CO-V for the obstacle avoidance of intelligent autonomous vehicles (IAV)

The use of hybrid intelligent systems (HIS) is becoming necessary to bring the behavior of intelligent autonomous vehicles (IAV) near the human one in recognition, learning, decision-making and action. First combination of expert systems (ES), neural networks (NN), and fuzzy logic (FL) to provide IAV with more autonomy and intelligence is discussed. Second, a neuro-fuzzy expert system E-S-CO-V (Expert System Controlling Vehicles) for the IAV obstacle avoidance is suggested. Indeed, E-S-CO-V improves Maeda´s obstacle avoidance approach (1990, 1991) by handling size and number of obstacles. On the other hand, the fuzzy reasoning and inference to decide static and dynamic obstacle danger degrees are replaced by fuzzy neural networks (FNN1) and (FNN2) respectively, while the decision table of avoidance direction is replaced by NN3. Finally, simulation results illustrate the efficiency of E-S-CO-V in handling several obstacles with different sizes and display its ability to achieve an intelligent obstacle avoidance