Internal Model Control for AUVs with Output Time Delays and Input Disturbances

Abstract

Autonomous Underwater Vehicles (AUVs) are increasingly being used for offshore inspection tasks. This paper investigates how navigation using Simple Internal Model Control for Proportional-Integral-Derivative Control (SIMC-PID) operates in a realistic offshore environment with waves and ocean current acting as input disturbances while the available underwater sensors introduce time delays on the output signals. First, the time delays are determined by investigating available absolute positioning sensor systems. Then, a model of the AUV and the external disturbances is established. The model-based SIMC-PID controller is tuned and examined based on acceptable disturbance rejection while tolerating the dominant time delays. Two simulation case studies show that the heave controller, in both cases, struggles to stabilize in 0 to 8 meters depths, while the surge and sway controllers tolerate the Doppler Velocity Log case (DVL) acceptably. Short Baseline (SBL) shows unacceptable performance in 0 to 15 meters depths. It is concluded that the simplicity of the SIMC-PID controller is an advantage and, therefore, useful when time delays are relatively small, but more advanced techniques must be applied for larger delays such as those introduced by SBL systems.