Dynamic Behavior And Control Strategy Of A Landing AUV

Autonomous underwater vehicle (AUV) is widely used as an effective platformin the oceanic research and exploration. AUV has high flexibility and can work inareas that are hazardous to humans or where humans cannot go, such as natural orman-made disastrous regions, deep ocean, and under ice. AUV needs a significantamount of onboard power to complete missions. However, energy storage is limitedin AUV. A Landing AUV was developed to prolong the monitoring time, also itsdynamic behavior and control strategy is studied. To obtain this behavior, the vehicleis required to have the capability of self ballast. When sitting on the seafloor, theAUV can enter into a sleep mode to conserve power, while the monitoring sensors areawake and working. In this dissertation, a dynamic model is developed to analyze theeffects of kinematical parameters on stability and maneuverability of the landing AUV.A landing strategy for the vehicle has been proposed and a fuzzy neural networkalgorithm has been applied in motion controller design. A series of sea trials have alsoconducted to verify the theoretical analysis. Comparison shows that the simulationresults are in a good agreement with experimental results.The prototype of the landing AUV has been designed based on concept ofmodular mechanical structure and distributed control architecture. The prototype canachieve long-range navigation, underwater landing, bottom-sitting and rising.A nonlinear dynamic model of the landing AUV has been formulated usingNewton-Euler Equations and hydrodynamic parameters have been obtained usingcomputational fluid dynamics method and physical experiments. Simulation of thebasic motions, such as rectilinear motion, rotary motion, and helix motion, has alsobeen studied. The simulation results provide fundamental support for motion control.The stability and maneuverability of the landing AUV in the horizontal planeand the vertical plane have been analyzed based on the dynamic model derived above.The stability curved surface which is a function of the fin coefficient, the coordinatesof the center of gravity, and the velocity of AUV. It is established to evaluate the stability of Landing AUV when its overall shape is determined. The important role offins in AUV motion stability is analyzed and the result is used to guide the design offins. The stability analysis also indicates that stability of the landing AUV increaseswith its velocity. Raising the center of gravity can also enhance the stability.The motion control system of the landing AUV has been studied using a fuzzyneural network algorithm. Controllers, including speed controller, orientationcontroller, and depth controller, have been respectively designed. Considering thelimitations of PID controllers, the fuzzy neural network PID controller is proposed. Alanding strategy has been studied regarding the topography of the seafloor. Landingof the AUV is divided into four stages according to the presented strategy. Thelanding trajectory has been designed by lines of sight and cross track error trackingmethod to ensure the navigation stability and safety.