Artificial Techniques Based on Neural Network and Fuzzy Logic Combination Approach for Avoiding Dynamic Obstacles

The autonomous mobile robot must be capable of avoiding static and dynamic obstacles in the environment navigating towards target without any human effort. A valid low-cost path from start to goal is obtained by A* algorithm. Neural network used for Zone classification. relative values between obstacle are classification which distance, velocity, angle. Zone1 very dangerous while zone 5 not dangerous. If neural classifies as a activates controller. fuzzy logic makes decision reaction prevent collision. There three inputs (relative angle) obstacle. outputs velocity steering angle robot. Static have been added addition make more complex. Three tested. While one obstacle, another enters critical becomes avoids second it first Then velocities increased prove that proposed system can handle cases with faster velocities. simulation results tested shows capability method fully known environment.