Skripsi
IMPLEMENTASI FUZZY LOGIC CONTROL TYPE-2 SEBAGAI SISTEM KENDALI POSISI PADA AUTONOMOUS BUOY
This research analyzes the application of the Fuzzy Logic Controller (FLC) type-2 for Autonomous vehicle steering control, using input values in the form of error and delta error from the difference between the output produced by the primary controller and the steering angle value obtained from the calculation of the pulse encoder installed on the steering wheel. This data is then processed through ROS (robot perating system). This study compares the performance of type-2 FLCs with 7 members and 5 members, as well as PID controllers in various scenarios. The results show that FLC type-2 with 7 members achieves an average error of 4.97%, better than the 5 member configuration which has an error of 7.71%. In Obstacle avoidance tests, FLC type-2 demonstrated superior accuracy with average errors of 1.54% for human avoidance, 4.28% for one parked car, 1.2% for two cars parked on the left, and 2.13% for two cars parking on the left and one on the right. This compares to PID controllers registering errors of 2.19%, 3.49%, 1.12%, and 3.49% respectively. Full route testing from the front of the Electrical Engineering Department to the Faculty of Engineering and back recorded an average error of 8.87% for FLC type-2 and 12.35% for PID, while the return route error was 4.52% for FLC type-2 and 7.57% for PID. Type-2 FLC with 7 members has proven to be more effective in maintaining accuracy and performance in dynamic driving conditions than PID, although PID has a smoother response to small error values. These findings demonstrate the potential of type-2 FLC in improving steering accuracy and overall performance of Autonomous cars.