An Attitude Monitoring Method For Broad Width Agricultural Machinery

With the accelerated development of agricultural modernization in China, agricultural machinery towards large-scale development, the complexity, the working width and other parameters of agricultural machinery gradually increased as well. However the agricultural labor force continued to decline, this had put forward higher requirements for the operators and intelligent agricultural machinery and equipment. The attitude of agricultural machinery mainly includes the pitch angle and the roll angle, and the roll angle is an important factor affecting the quality of work, the positioning accuracy and control performance of the equipment. Because of the poor working environment of agricultural machinery, the attitude of the broad width agricultural machinery which working on the surface of slope farmland would constantly change. Therefore, it is vital to monitor the roll angle of the machine effectively to improve the performance of the equipment, which can prevents the waste of energy and preserves soil ecology. So the roll angle has been chosen as the main parameter to monitor the attitude in this paper, and an attitude monitoring method for broad width agricultural machinery working in the field has been studied.This paper analyzed the research status of attitude monitoring method for broad width agricultural machinery in the domestic and foreign respectively. And the problems existing in current research methods have also been analyzed. The attitude monitoring methods based on sensor information, image feature matching and multi-source information fusion have been researched respectively in order to achieve the purpose of accurately monitoring the attitude of broad width agricultural machinery.As for the attitude monitoring method based on sensors information, the date of MTi-300 AHRS inertial sensor and Beidou satellite navigation system have been fused based on Kalman filtering method. The design of software scheme for the monitoring terminal has been completed, and the man-machine conversation interface of the monitoring terminal has been designed by using MATLAB GUI. Tests have been conducted to evaluate the system’s performance, which show good accuracy of the system. After the date fusion of MTi-300 AHRS and Beidou satellite navigation system, the maximum absolute error of the roll angle measurement value for the system was 1.21° and the minimum absolute error was 0.23° and the average value of absolute error was 0.72°, which demonstrate that the integrated navigation algorithm can better monitoring of roll angle, it can meet the demand of the system precision.As for the attitude monitoring method based on image feature matching, according to the principle of image matching, a method to determine the roll angle changes of agricultural machinery based on the dynamic change of farmland environment is presented. The feature extraction algorithm is used to extract the feature of the images in two consecutive frames, and the local feature points are matched to get the rotation angle between two consecutive frames. The man-machine conversation interface has been designed by using MATLAB GUI, and the design of dynamic stability acquisition system based on image feature points has been completed. Tests have been conducted to evaluate the system’s performance, which show that the maximum absolute error and the average absolute error of the static contrast test were 0.13° and 0.07° respectively, the maximum absolute error and the average absolute error of the dynamic contrast test were 0.97°and 0.61°respectively. The test shows that the algorithm can better monitor the attitude change of agricultural machinery in field.An adaptive weighted fusion method is adapted for the attitude detection based on the sensor information and image feature matching. Tests have been conducted based on the attitude monitoring method of multi-source information fusion. Before the fusion, the average absolute error of the attitude monitoring method based on sensor information and image feature matching were 0.65° and 0.72° respectively. After the fusion, the average absolute error of the system was 0.35°, indicating that the fusion algorithm was effective to reduce the measurement error.An attitude monitoring method for broad width agricultural machinery has been studied. And an innovative method has been proposed which based on image feature matching to obtain the roll angle of broad width agricultural machinery. This method provides a new idea for broad width agricultural machinery attitude monitoring, which has high precision and a certain application prospect.