A SIGMOID FUNCTION OPTIMIZATION FOR OSTACLE AVOIDANCE MECHANISM ON PATH FOLLOWING CONTROL OF AN AUTONOMOUS TRUCK-TRAILER

• Main Author: Asrofudin, Bani
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/57446
• Institution: Institut Teknologi Bandung

The increasing intensity of loading and unloading in the world of port logistics causes the accumulation of containers because they aren’t accommodated in the container yard so that the traffic lane becomes blocked. An obstacle avoidance mechanism is needed to accommodate these conditions so that a collision does not occur. In the application of a track-following autonomous truck-trailer, the avoidance path reference must be dynamically adjusted to bring the system to change lane. Sigmoid function optimization is proposed to obtain optimal obstacle avoidance path reference for steering comfort and avoiding tractor trailer effects. The Fmincon Nonlinear Programming function is used to get the optimal solution of the optimization problem. The optimization results obtained parameter of curve concavity K_?s= 0,55 and parameter distance of inhibiting objects K_rs= 2,0247. The optimal turning path is integrated with the Lyapunov stability-based path follower control to obtain a global asymptotic stable state. Global asymptotic stability is achieved based on the origin of error of the continuous trailer which converges to zero with a total value of RMSE of 0,26 which is better than the system with unoptimization path. The simulation results show the average speed achieved is 0,51 m/s. The overall performance value of IAE, ISE, ITAE, and ITSE shows a system with optimized path is efective.