Variable buoyancy control for a bottom skimming autonomous underwater vehicle

Abstract

A variable buoyancy system (VBS) is a critical component that is widely used to achieve controlled operation of an autonomous underwater vehicle (AUV). In this thesis two on/off feedback controllers are presented that utilize data averaging and model-based estimation to offset the effects of sensor noise and achieve precise control of the VBS developed for a prototype AUV. Operation of the prototype bottom skimming AUV requires a constant reaction force between the seabed and the vehicle. While performing a mission, variable seafloor topography and a changing payload weight requires the use of a VBS to regulate the reaction force. One trait that made development of the VBS system a challenging problem is the presence of sensor noise that could not be mitigated with hardware solutions. The software solutions implemented here needed to minimize any time delays in their response that could cause destabilization when coupled with the system's fast on/off actuation. It was discovered that both of the presented controllers function under these conditions, but the model-based controller provides more precise control of the system. The data averaging controller was able to regulate the buoyancy of the system to within 5.0 pounds of the commanded setpoint, while the model based controller was able to regulate the buoyancy to within 1.25 pounds of the desired setpoint, an improvement in performance by a factor of 4. Presented here is the development and comparison of these two control algorithms based on both simulation results and field experiments conducted in a coastal environment.