Indoor Positioning And Control Of Quadrotor Unmanned Aircraft Vehicle Based On SLAM

Quadrotor unmanned aircraft vehicle(QUAV) has become a research focus in the field of UAV in recent years because of its simple structure, low cost, excellent performance and unique flight control methods. In the implementation of the indoor environment monitoring and reconnaissance missions, QUAV has obvious advantages and its application prospect is quite broad. In this dissertation, the nonlinear mathematical model of QUAV is given and the indoor positioning problem is studied by using simultaneous localization and mapping(SLAM) technology, and the control methods of QUAV are investigated with modeling uncertainties, external disturbances, input saturations and attitude constraints by using the position and attitude information estimated by SLAM technology. The main research contents are as follows:Firstly, the nonlinear mathematical model of QUAV is established according to the motion characteristics of QUAV. In order to facilitate the follow-up study, QUAV mathematical model model is divided into fast-loop equation and slow-loop equation formed by attitude loop and position loop respectively, and the forms of affine nonlinear equations are obtained after transformation.Then, aiming at the problem of indoor positioning of QUAV, the SLAM algorithm based on iterative closest points(ICP) is studied by using the laser range finder, the attitude sensor and the altitude sensor. The problem of laser data and height data skew is corrected by using the attitude data measured by attitude sensor. Besides, the indoor localication of QUAV is calculated by the ICP match algorithms. The creating and updating steps of grid map and geometric map are given respectively to solve the problems of map building. The SLAM experimental results show that the ICP positioning algorithms have good stability and high accuracy, and the comparision of two kinds of map representations is given by the map created by SLAM.Thirdly, a backstepping control scheme based on disturbance observer is developed to fast-loop system and slow-loop system of QUAV to deal with the unknown external disturbances. Using disturbance observer to handle the compound disturbances combined by the external disturbances and the modeling errors, and the backstepping control strategy is developed based on disturbance estimations. The rigorous stability of the closed-loop system is proved with Lyapunov method. Simulation results demonstrate the effectiveness of the developed control scheme for the QUAV with the external disturbances and modeling errors.Following, a backstepping-based control strategy is derived for fast-loop system of QUAV with model uncertainties, unknown external disturbances, input saturations and attitude constraints based on disturbance observer and neural network. A neural network method is utilized to estimate the model uncertainties of the system. Besides, a disturbance observer is introduced to compensate for the complex disturbances combined by the external disturbances and the modeling errors. And, the input saturations are smoothed by using hyperbolic tangent function. For the attitude constraints, the barrier Lyapunov function(BLF) is used to guarantee the constraints of the attitude. A backstepping-based controller is given to control the attitude of the QUAV. The closed-loop control system is proved to be uniformly bounded by using Lyapunov stability theory. Simulation results demonstrate the effectiveness of the developed control scheme, when model uncertainties, unknown external disturbances, input saturations and attitude constraints appear simultaneously.Finally, the indoor SLAM experiment platform of QUAV is given and two indoor experiments are accomplished by using small area scene and big area scene respectively. The experiment results indicate the feasibility and validity of the indoor SLAM algorithms.