Design And Research Of Attitude Measurement System For Small UAV

Small unmanned aerial vehicle(UAV)has the advantages of small size,light weight and low cost,and is widely used in civil and military fields.The attitude measurement system is an important part of the small UAV system.It senses the position and attitude information of the UAV through sensor measurement to make the UAV easier to control.The small UAV system is a nonlinear and underactuated system.It is very difficult to measure and control its attitude.Accurate attitude measurement is the prerequisite for its efficient control.Based on the in-depth study of the small UAV attitude measurement system,this paper designs a small UAV attitude measurement system based on inertial and visual sensor data fusion,and describes its working principle,software and hardware design and attitude algorithm.The hardware circuit of small UAV attitude measurement system is mainly built with the control chip STM32F427VIT6,inertial sensor MPU9250(internally integrated 3-axis gyroscope,3-axis accelerometer and 3-axis magnetometer)and visual sensor ATK-MT9V034.The attitude algorithm is constructed based on hardware circuit.Firstly,the attitude representation method based on Euler angle method and quaternion method is introduced;Secondly,in the small UAV attitude measurement based on inertial sensors,due to the poor dynamic and stable performance of the accelerometer,and the good dynamic and stable performance of the gyroscope,the complementary filtering algorithm is used to learn from each other’s strengths and complement each other’s weaknesses to coordinate the performance of the two sensors,which solves the shortage of single sensor measurement;Thirdly,in the small UAV attitude measurement based on vision sensor,the geometric perspective projection method is used to measure the small UAV attitude.The main work includes image preprocessing,image region segmentation,image target feature point information extraction based on wavelet transform technology and attitude estimation;Finally,in order to solve the problem of error accumulation when the inertial sensor takes too long to measure,and the problem of target error detection and missing detection caused by image degradation due to the high-speed movement and rotation of UAV when the visual sensor is measuring,an adaptive volume Kalman filter algorithm based on the fusion of inertial and visual sensor data is introduced.Small UAV with inertial sensors and visual sensors is used to build an experimental platform with an upper PC mechanism.Static experiments and dynamic experiments are designed using matlab software.The experimental results show that the proposed multi-sensor data fusion algorithm has higher accuracy than the algorithm using only one sensor,and realizes the complementarity of attitude measurement based on inertial sensors and visual sensors.