Research And Implementation Of Attitude Estimation Algorithm For Quadtator

The quadrotor aircraft as a symmetrical traditional structure,directly controlled by the speed of DC motor,is a non-traditional UAV.Because of its simple structure,flexible control and being able to adapt to all kinds of complex geographical environment and meteorology,it has been widely applied in various fields.Secondly,as an underactuated and strongly coupled nonlinear system with a typical four input and six output,it is also very meaningful as the object of scientific research.With the continuous progress and development of sensor manufacturing technology,low-cost sensors are emerging,and at the same time,they show low accuracy.Because of their low cost,they are widely used in flight control systems.Therefore,how to make up the deficiency of the sensor by software algorithm and get the accurate information is a very necessary problem.Two attitude estimation algorithms are proposed in this paper,which is aimed at the low precision of the attitude measurement unit of the flight control system.In this paper,the quadrotor aircraft is used as research object,of which the mathematical model is discussed in detail,and the complex model is simplified.Then through the analysis of structure and principle of the flight control system,two kinds of attitude estimation algorithm is discussed,namely,gradient descent algorithm based on quaternion of error(GD)and the extended kalman filter(EKF)algorithm.Algorithms are simulated by using Matlab software tool for the validity of the algorithm and accuracy analysis.The adjustment of the parameters is further prepared for the application of the algorithm.In order to verify the reliability of the attitude estimation algorithm,quadrotor verification platform is set up,which is controlled by STM32F407 processor and uses MPU6050 as the inertial measurement unit.There is a detailed description about the design of platform and write the software architecture.Finally,debugged by PC software,the quadrotor aircraft is tested for design of software and hardware and attitude estimation algorithm.The conclusion is that the proposed attitude estimation algorithm can meet the requirements of aircraft flight.Meanwhile,the attitude estimation of extended kalman filter algorithm is more reliable than gradient descent algorithm based on quaternion of error.