TRAJECTORY AND ATTITUDE CONTROL OF UNDERWATER ROBOT FOR MINIMUM ENERGY CONSUMPTION

Abstract

This work presents the trajectory and attitude control of an underwater robot for minimum energy consumption. It involves the integration of bondgraph modeling technique and genetic algorithm as optimization technique. The robot is considered to be performing a repetitive motion task in underwater environment. In bondgraph modeling, most significant hydrodynamic forces pertaining to added mass effects, buoyancy and gravity, linear and quadratic damping effects are considered. The non-significant hydrodynamic forces like wave currents and dynamic fluid forces due to fluid flow are not considered in modeling. The modeling is done with bondgraph modeling software Symbol Shakti. The program generated by Symbol Shakti provides the mathematical expressions for model using which a program for numerical integration of these expressions is created. This program is then used as an objective function for the genetic program which compares power required for different trajectories. As a result, the genetic program yields the coefficients of desired energy optimal trajectory. As another method for energy conservation in case of underwater robot, optimization with minimum number of thrusters in attitude controller is presented. The attitude controller having two pairs of thrusters about two principle axes and a thruster to counter gravity in vertical direction is used