Dynamic reconfiguration of autonomous underwater vehicles propulsion system using genetic optimization

In this work, a method for the dynamic reconfiguration of autonomous underwater vehicles (AUVs) propulsion system is developed and tested in simulation. Global optimization is used to find suitable propulsive configurations and control parameters in order to achieve successive robotic tasks. This is done using a genetic algorithm used here as a task-based design method. Evaluation of the AUV configuration is made using a dynamic simulation of the robot performing its mission involving a non linear control (computed torque method). The objective function minimizes the trajectory tracking error and the energy consumption. The results of this optimization could be used as a preliminary design step for a specialized AUV, finding the fittest propulsive configuration. Here, we propose the dynamic reconfiguration of a multitasking AUV that adapts its propulsion configuration to the changing task specifications during its mission. The method implementation is based on the dynamic model of an existing AUV (IRDL-RSM 4-fixed thrusters under-actuated robot), but searching for new thrusters directions and controller parameters to perform successive tasks.