Research On Positioning Method Of Boom-type Roadheader Based On UWB-IMU

In the context of the fourth industrial revolution,coal is a vital basic energy source for China,which demands higher standards for the intelligence of coal mining equipment.The intelligence of coal mining equipment can effectively lower the risk of on-site workers and enhance the mining efficiency.The boom-type roadheader is a crucial device for coal excavation in mines,and its precise positioning is a key technique for its intelligence.This essay presents a positioning method for the boom-type roadheader based on ultra-wideband-inertial measurement(UWB-IMU),which adapts to the actual working condition of the boom-type roadheader in coal mine tunnels.By constructing a dataset,a multi-input multi-output machine learning model is built to fuse the UWB-IMU data,and the validity of this method is verified by experiments.The main research work and results are summarized as follows:(1)A positioning system for boom-type roadheader was established.Aiming at the positioning demand of boom-type roadheader,the structure of positioning system in roadway was designed,the pose estimation model of roadheader was determined,the positioning system of roadheader was constructed,and the hardware selection and accuracy calibration of positioning system were completed.The positioning software development based on ROS was realized.(2)The two-way ranging method(TWR)and time difference of arrival(TDOA)positioning model of UWB were studied,and the pose estimation algorithm based on Newton iteration method was analyzed and experimentally verified.The ranging experiment was carried out in a narrow and long space,and the effects of antenna orientation,transmission power,base station geometric distribution,and distance between base station and tag on UWB ranging accuracy were investigated.The UWB pose estimation algorithm based on Newton iteration method adopted TWR ranging method to avoid the clock synchronization problem in TDOA,and designed a Z-shaped dynamic experiment of roadheader to verify the accuracy of this algorithm.(3)A UWB-IMU data fusion method based on machine learning was proposed.Based on UWB ranging,the advantages of high accuracy of IMU roll angle and pitch angle were utilized to compensate for the low accuracy of pose and Z axis in UWB pose estimation.The nonlinear transformation relationship between 16 ranging values of UWB,2 poses of IMU and position and yaw angle of roadheader was fitted by machine learning method.The experimental results show that the average absolute error of threeaxis position is less than 0.10 m,and the average absolute error of three-axis pose is less than 1°.This positioning method can meet the requirements of boom-type roadheader positioning in roadway excavation process.