Research On Fusion Positioning Method Of Drilling Robot For Rockburst Prevention

Rockburst is one of the most serious dynamic disasters in deep coal mines in my country.At present,the pressure relief operation mainly relies on the manual operation of staff entering the dangerous area.During the pressure relief process,it is easy to induce rockburst disasters,which in turn cause greater personal injury accident.Therefore,it is urgent to realize the robotization of the drilling pressure relief operation,and the key technology for the robotized operation of the drilling pressure relief area is the precise positioning and attitude determination of the drilling robot.Inertial navigation and positioning technology is widely used in the field of coal mine robot positioning because of its strong autonomy,high reliability,and no need to rely on the external environment.However,this method will generate cumulative errors,and long-term operation will lead to unsatisfactory positioning results.Therefore,this thesis proposes a fusion positioning technology of drilling robot for rockburst prevention based on spatial array inertial measurement unit(IMU)and visual images.On the basis of inertial navigation,visual positioning is used to suppress the generation of accumulated errors,and the operation process of the drilling robot for rockburst prevention is realized accurate locating.The main work and research results of the thesis are as follows:(1)The basic structure and working principle of the drilling robot for rockburst prevention are analyzed,the inertial navigation coordinate system and visual coordinate system of the drilling robot for rockburst prevention are established.The inertial navigation positioning and visual positioning of the drilling robot for rockburst prevention are studied,and the fusion positioning scheme of the drilling robot for rockburst prevention based on IMU and visual image is proposed.(2)Aiming at the deterministic drift and non-deterministic drift generated during the inertial navigation and positioning process of the drilling robot for rockburst prevention,the spatial array IMU data fusion model of the drilling robot for rockburst prevention is established,the posture solution method of the drilling robot for rockburst prevention based on the space array IMU is studied,the initial alignment method of inertial navigation based on the Kalman filtering algorithm is designed,the inertial navigation positioning accuracy of the drilling robot for rockburst prevention is improved,and the superiority of the space array IMU solving algorithm is verified through simulation analysis.(3)Common image feature extraction and matching algorithms are analyzed.Aiming at the problems of atomization,glare,uneven illumination and noise interference in coal mine,an image enhancement algorithm based on the improved Retinex algorithm is designed to improve the accuracy of image feature extraction and matching.In order to further eliminate the cumulative error of spatial array IMU,a camera pose estimation model based on continuous frame images is constructed,and a visually corrected drilling robot for rockburst prevention fusion positioning process is designed.(4)Design and build an experimental platform for drilling robot for rockburst prevention positioning,and conduct experimental analysis in the simulated roadway of the coal mine.The experimental results show that the average absolute errors of the robot positioning method proposed in this thesis are 0.036 m,0.028 m,and 0.018 m in the x,y,and z directions,respectively.The average absolute errors of the roll angle,pitch angle,and heading angle are the errors are 0.382°,0.345°,and 0.389° respectively,which meet the positioning requirements of the drilling robot for rockburst prevention.The thesis has 45 figures,16 tables,and 100 references.