| Electro-optical stable platform is one form of inertia systems, which are applied in dynamic circumstances. As the limitations for equipment size and containment space get stricter, the use of executive part for electro-optical platform shifts from direct torque motor to micro gear drive motor gradually. Micro motor has high speed, tiny size and excellent robustness, which to some extent satisfy the demand. After long term usage, however, gear drive system brings in gap problem unavoidably. The control issue, such as turn-around jump and mechanical noisy, stands out evidently when gear turn round, slow down or speed up. As a consequence, system with high precision demand is not able to bear such uncontrollable question. So, in the essay blow, I aim to study the gap drive issue in order to improve the stability and rapidity.Firstly, the essay gives a brief description about the background and present status of the nonlinear gap issue. Basing on the context, it introduce the origin, development, present study and future direction about backstepping algorithm. Secondly, I send the main content. Researchers have found out that nonlinear gap issue exists multiple models. At first, the essay describes several universal models, such as gap model and dead-zone model. In order to illustrate the target choice of essay model, dead-zone model, I make comparisons between these different models, and list the advantages and disadvantages of all of them. Then, I make a thorough description of backstepping control algorithm, which is the core algorithm of whole essay. Introduction starts from stability criterion, which in this case Lyapunov function. Though the method of Lyapunov function, I have steps of construction for the controller. After the construction of controller, along with outer gear drive system, I scratch the whole picture of state equation scheme. According to the scheme, I have the key for simulation model. Simulation model doesn’t adapts internal "simulink block", on the other hand, I design function file, f.y.i m-file, instead. During the simulation, non-optimization control model produce divergence output response curve, so the system is not stable. To find out the un-stable factor, I make a fully analysis for amplitude and phase of the law of controller. By comparison, I find out that there exists maximum in one of the laws, which is caused by inherent parameters and control parameters. Given a weight to the maximum, system begins to act normally, and output curve shows stability. After analyzing several factors that have an impact on system stability, I make a conclusion for simulation pattern. To test the pattern, I design the backstepping control algorithm for spectrum photo system. After repeated experiments for different situation, the motor show expected performance. Comparing simulink and experiment, I find out that most of results in simulink is correct, and the others couldn’t be tested because of limited resource. At last, I reflect on the merits and drawbacks of the whole essay, and give a final summary.After weighing models, designing simulation and performing experiments, I prove the validity of backstepping control algorithm. Besides, during the whole project, I discover various features to be reformed, which guides the future deep experiment. |
