The Integrated Navigation And Control Of Hexacopter Basing On Vision

In recent years, with the progress of science and technology, Unmanned Aerial Vehicles(UAVs) came to the mass consumer fields from military applications gradually. Due to the advantages of flight stability, simple structure, easy manipulation and easy maintenance, they are widely used in the fields of aerial photography, military reconnaissance, disaster monitoring. In academic research, it is also perfect platform to conduct researches such as navigation in complex environments [1], to build aircraft mathematical model[2], the algorithm[3-6] attitude control, indoor and outdoor flight obstacle avoidance[7], Localization, navigation[8], Simultaneous Localization and Mapping(SLAM)[9] and formation flying [10]. Integrated navigation based on visual information and the traditional inertial navigation system is the solution of the urgent problem of indoor navigation which will not only improve the accuracy, but also pave the way for the popularity and intelligence of the aircraft. Therefore, this article forcus on the visual integrated navigation and control. The main results of the work are the following:1, proposed visual position, attitude measurement programs, gave a method of measure aircraft position and attitude through specific markers and verified the feasibility and accuracy of the algorithm. The core of visual measurements includes’ external parameter of images solving, fast corners detection and decoding two-dimensional code recognition algorithms;2, as the traditional measurement of the gyro of inertial navigation system(INS) drifts by time, accelerometers and inertial navigation compass are used to correct the data of INS, which alleviate the drift problem to some extent; optimal estimation Kalman filter algorithm is used to fuse the results of visual and INS measurement. To solve the problem of frequency inconsistent for visual and INS measurement and the possible visual gross errors, Kalman filter algorithm is modified to improve performence. Finally, the effect of fusion experiments using Matlab to show testing results;3, proposed the ADRC position and attitude controllers for aircraft navigation control. The designed six ADRC controlers for three positional parameters X 、Y 、Z and three attitude parameters? 、? 、?. ADRC controlers don’t rely on the system model, can eliminate channel coupling, and overcome the external disturbance like wind which are ideal choice for aircraft control.4, basing to the analysis of the mathematical model of hexacopter, we built the hexacopter simulation platform.The platform using Simulink and Virtual Reality(VR) toolbox is feasible to verify the control algorithm, and it can display dynamic aircraft process during flight in 3D. We can arrange practical aricraft application after the simulation which not only has a positive meaning to reduce cost but also improve research efficiency.