Automotive is one of the most important industries in the world, which lead to a need for continuous improvement of vehicles and their internal systems. Suspension systems have been improved for better vehicle performance and passenger comfort, keeping the tire in contact with the road surface. Active suspensions require optimal control to modulate the flow of energy and generate the control force by implementing active actuators able to provide negative damping and a wider range of forces and velocities. This article aims the design of an active suspension system based on LQG and LQR controller evaluating its performance in a distributed parameters simulation using COMSOL Multiphysics® and MATLAB®. The quarter car model is proposed and is linearized to design the optimal control (LQG and LQR) respectively. An early mathematical simulation is developed in MATLAB (R) software to verify and compare the open and closed loop results. Finally, the full system model is implemented in COMSOL Multiphysics (R) software considering rigid materials and the controller to analyze the distributed parameters simulation results.

Remotely Operated Vehicles, ROVs, are useful devices that can greatly assist humans in the development deep-sea exploration and underwater tasks. These unmanned vehicles require human intervention for the realization of its underwater jobs and have the ability to carry multiple instruments, sensors and actuators according to the application. There are numerous commercial ROV platforms, ranging from inexpensive to very advanced and costly systems. Being an underactuated system, one of the most important factors in the development of a ROV is the design of its control system. Depending on the quality of the implemented control strategy, the functioning of the vehicle may or may not fail, or the accuracy with which its assignments are done may be seriously compromised. Different control strategies can be utilized for the stabilization and maneuvering of a ROV. The design of these strategies often require system parameter identification. Appropriate modeling and knowledge about the dynamic behavior of the system is essential for a successful parameter identification. One of the main parameters that must be identified is the drag coefficient of the ROV as it moves in the fluid. This parameter can be either experimentally measured, or estimated using the finite element method, to quantify the forces due to fluid-structure interaction. This work seeks the design and comparison of different advanced control techniques as applied to a small ROV. A commercial small ROV system has been chosen as the object of study and finite element simulations were carried out to estimate some of its mechanic parameters, using the commercial software COMSOL Multiphysics®. The nonlinear model of the system is developed and linearized to obtain its state space representation. The state space representation of the system is then used in the design of a LQR control system. The comparisons of the responses of the compensated systems allows assessing the suitability of the optimal control strategy for stabilization of ROVs.

No abstract