Attitude Estimation Of Crane Hook Based On Strap-down Inertial Navigation Technology

With the development of the Chinese economy, truck crane has been widely used in industrial fields such as cargo handling, housing construction, station construction, etc. How to satisfy the demand of increasing carrying weight and working speed is becoming an urgent issue. The key point is to provide high accurate attitude of crane hook to operator by detecting the swing angle, orientation angle and position of hook. This can guarantee the efficiency and safety of crane operation and have great engineering significance. The proposed solution integrates strap-down inertial navigation with attitude estimation of crane hook. On the basis of establishing 3D motion model, data fusion is implemented to achieve real-time detection.To adapt to the specific structure of crane hook, different algorithms such as direction cosine matrix, attitude quaternion, Euler angles etc. solving attitude angles are discussed. We select the quaternion algorithm as the best method by combining advantages and disadvantages of proposed algorithms. A three-dimensional hook motion mathematical model is built to analyze the hook motion. Meanwhile, the overall block diagram of proposed system is given. To eliminate the effects of hook torsion, the spherical triangle theorem is applied to the motion model which acquires precise attitude angles without be affected by torsion.Multi-sensor data fusion algorithm is proposed to tackle the drift problem of gyroscopes, accelerometers and magnetometers. An improved complementary filter named adaptive complementary filter is developed to fuse data acquired from different sensors along with an extended Kalman filter which has the highest precision. The following is attitude error range: yaw angle, pitch angle, roll angle, swing angle and orientation angle are less than 2°,1°,1.5°, 0.5° and 2.5% full range respectively.This research has designed the attitude detection platform of crane hook based on iNEMO module which integrated inertial sensors and magnetometers. Hook attitude is transmitted via wireless communication to PC. A reference platform has also been developed based on visual inspection to verify the estimation. The results indicate the proposed system can effectively fuse the multi-sensor data, monitor hook status by wireless transmission and achieve highly precise attitude which has a promising perspective of industrial application.