Design of four axis aircraft based on Android control

A four axle aircraft is designed with the open source firmware MWC is used as the flight control board .The MPU6050 gyro sensor and BMP085 pressure sensor are used to acquire the attitude data of aircraft in real time. Combining the Calman filtering and mean filtering, the attitude data is filtered and fused to obtain more accurate attitude information. In the control algorithm, the PID control algorithm is adopted to control the speed of the motor, which ensures that the aircraft can quickly and stably change the attitude of the flight. The Android handset is connected to an aircraft by using the WiFi signal which is launched on the four axle aircraft. The user can control the aircraft by using the appropriate App. Experimental results show that: The four axle aircraft can realize the functions of attitude, heading and hover under the control of App , and the system has friendly interface, flexible operation, stable and reliable.


Introduction
The four axis aircraft is becoming more and more popular.Its common control is to use radio remote control [1].There are three main types of remote control transmitters for four axis aircraft at present: The first one is a portable lever type remote control transmitter, the second one is a hand-held gun remote control transmitter, and another one is a small hand-held remote control transmitter for cassette keys.
Gun type transmitter is the most popular remote control system at present, The gun type transmitter is a runner and a similar trigger lever to control the direction and speed respectively, remote control system of switch type analog circuit, the general model of ordinary toy remote control will use the remote control system, operation instruction is relatively simple, only the "on" and "off" of the two states, Performance and remote control distance is low.The four axis aircraft is controlled by Android phones.Using Android mobile phones to control four axis aircraft removes the traditional handheld wireless remote controller and the operation is simple.

Introduction of control algorithms
If a gyroscope or accelerometer is used alone to detect the attitude of the flight, it is impossible to provide reliable data for the balance of the aircraft.According to the characteristics of both, if the method used for filtering, fusion of gyroscope and accelerometer output signal to compensate for the cumulative error, gyro drift error and dynamic error of the acceleration sensor, can make the two mutual calibration, which can play the advantages, can also avoid their disadvantages, so the measurement can get a dip a better approximation, thus more accurate attitude measurement.
The design of the MPU6050 sensor output data using Calman filtering and mean filtering to filter and fusion, in order to obtain more accurate attitude information.Using the PID control algorithm to control the speed of the motor, to maintain the four axis aircraft in changing the attitude of the flight can be fast and stable.

Calman filtering algorithm
Calman filter [3] can measure a series of signals that do not contain noise completely, and can estimate the state of dynamic system optimally.It is an efficient recursive filter [4].
It is assumed that a linear stochastic differential equation is used to describe a discrete control process system: The measured values of the system are: is the system state of the k , and u is the amount of control that is added to the system at the k moment.In formulas, and are intrinsic parameters of the system, and when the system is multiple modules, they are matrices.k time measurement value is , when it is in the measurement system, parameter measurement system is H matrix. W and V respectively represent system processes and system measurements of noise.
Assuming that the Gauss white noise does not vary with the system state, then the covariance of is Q and R. The Calman filter is the best processor for the process satisfying the above conditions.
First use system process model to predict the next state: is the value predicted by the previous state.
is the optimal result of the previous state, and ) (k U is the control of the current state.
The covariance equation corresponding to The current state measurement value of the system is collected, and the optimal estimation value of the current state can be obtained by fusing the predicted value and the measured value The g K is Calman gain in the formula: In order to ensure that the Calman filter can continue running, the covariance of the

PID control algorithm
In the control of four axis aircraft, because there are many unknown disturbances and disturbances that affect the flight of an aircraft, the PID control [5] is adopted in this design.
The controller based on the ratio of deviation, integral and differential is called the PID controller.This automatic control law is widely used in the control system [6,7].
As shown in Fig. 1, The control formula of analog PID controller is: In the formula,

App send and manipulate instructions
After a successful connection, the user through the mobile button on the left and right sides to control four axis aircraft, when the user presses the button, the program will capture the current screen touch points in the X and Y axis coordinate point, when the finger moves in the circle area, the program will change X and Y axis control with respect to the amount of acquisition and, to control set new coordinates, so that the control can be moved with the movement of the finger.In the process of moving, the program will take the measured coordinate value and coordinate value changes as parameters, calculate the control data, and then send the control data to the four axis aircraft through the network.Then, the data is received by the WiFi module on the four axis aircraft and transmitted to the master chip through the serial port.

Summary
The four axis aircraft designed in this paper is a portable aircraft with a small volume, which is

Four
axis aircraft using MWC open source firmware for flight control board, and has a digital pressure sensor BMP085, MPU6050 digital three axis accelerometer and three axis gyro sensors and HMC5883L three axis digital magnetoresistive sensor, can help determine the four axis aircraft flying state with the sensors, and can adjust the flight attitude according to the parameters of each sensor.Therefore, the open source firmware is suitable for the development of four axis aircraft.In addition, the firmware has reserved other interface pins, such as I2C interface and serial port interface, which makes the ports for other sensors to be added later.The four axis flight control module consists of the following components: and T A is the transpose of the A , and the value of Q is the covariance of the system process.MATEC Web of Conferences 139, 00160 (2017) DOI: 10.1051/matecconf/201713900160 ICMITE 2017 unit matrix in the formula and I=1 is in the single model single measurement system.When the system is in the this point.In this way, we can do the regression calculation and realize the optimal estimation.The average filter takes the average value of N values as the measured value, so that the measured value is close to the actual value, and can reduce the random error and weaken the influence of the external disturbance on the control process.

powered by 4
brushless motors, with high flexibility and flexibility.The whole machine is fully electronic stabilized and easy to operate.It has a wide range of applications in many fields.Four axis aircraft flight control system based on Android design in this paper, the Android mobile phone is connected with the four axis aircraft by launching four aircraft wifi, off user with the appropriate App can control the four axis aircraft, landing and other attitude change.Not only the operation is simple, but also eliminates the traditional handheld wireless remote control, increasing the user's operating experience.