Title :
Depth control of an autonomous underwater vehicle, STARFISH
Author :
Hong, Eng You ; Soon, Hong Geok ; Chitre, Mandar
Author_Institution :
Tropical Marine Sci. Inst., Nat. Univ. of Singapore, Singapore, Singapore
Abstract :
We present a depth controller design for a torpedo-shaped autonomous underwater vehicle (AUV) known as STARFISH. It is common to design an AUV to be positively buoyant, so that it will float to the surface in case of power failure. However, most depth controllers are designed with a neutral buoyancy assumption by regarding the extra buoyancy as a disturbance. In this paper, we study the effect of buoyancy on both pitch and heave dynamics of an AUV, and propose a controller scheme that specifically compensates for the effect. We propose a simplified model for pitch dynamics that takes into account the buoyancy of the AUV. We identify the parameters of the model from field data from a closed loop depth maneuver. We adopt dual loop control methodology with inner pitch control loop and outer depth control loop. The inner pitch controller is designed using sliding mode control (SMC) with integrator effect to overcome a constant offset term due to positive buoyancy of the AUV. Then, a simple proportional controller is designed in the outer loop for depth control. Positive buoyancy of the vehicle will induce heave motion of the AUV. Thus, in order to maintain depth, the AUV need to be pitch down at certain angle. An adaptive feedforward controller is designed to compensate for this angle. The dual loop design with inner SMC and outer proportional control with feedforward loop was shown to be effective through experiments in both lake and sea.
Keywords :
adaptive control; closed loop systems; control system synthesis; feedforward; parameter estimation; remotely operated vehicles; underwater vehicles; variable structure systems; STARFISH; adaptive feedforward controller; autonomous underwater vehicle; closed loop depth maneuver; depth control design; dual loop control methodology; feedforward loop; inner pitch control loop; integrator effect; neutral buoyancy assumption; parameter identification; pitch dynamics; power failure; sliding mode control; Data models; Elevators; Feedforward neural networks; Payloads; Steady-state; Tuning; Vehicle dynamics; Autonomous Underwater Vehicle (AUV); Positive Buoyancy; Sliding Mode Control;
Conference_Titel :
OCEANS 2010 IEEE - Sydney
Conference_Location :
Sydney, NSW
Print_ISBN :
978-1-4244-5221-7
Electronic_ISBN :
978-1-4244-5222-4
DOI :
10.1109/OCEANSSYD.2010.5603566