Research On Monitoring System Of Roadheader Body Position And Posture

The cantilever roadheader is one of the key equipment used in coal mine excavation and is an important part of the complete set of equipment for the tunneling face.With the improvement of the mechanization and intelligence level of my country's fully mechanized mining face,the annual consumption of the length of the mining roadway has increased significantly,but the automation and intelligence level of the driving face has not been significantly improved,and the imbalance of coal mining has become more and more obvious,which has become a constraint on my country's coal the "stuck neck" problem of industry development.The autonomous directional tunneling of the tunnel boring machine plays a vital role in improving the intelligence level of the tunneling face,and the monitoring of the body's posture of the tunnel boring machine is the core issue of the tunneling machine's autonomous directional tunneling.At present,most coal mines in my country only use laser pointers to instruct roadheaders to carry out directional tunneling operations.This is not only inferior in accuracy,but also low in automation and high in risk factors for workers.It deviates from the development direction of unmanned mines and cannot be real-time during tunneling.Obtaining the posture information of the roadheader's fuselage cannot solve the problem of autonomous directional tunneling of the roadheader.Based on the above problems,this article takes the roadheader fuselage pose monitoring as the research goal,designs the roadheader fuselage pose monitoring system,and studies the roadheader fuselage pose measurement method based on monocular vision.The main research contents of the thesis are as follows:(1)Analyzed the structure and working process of the cantilever roadheader,established the roadheader body pose monitoring coordinate system,simplified the roadheader model,determined the pose monitoring parameters,and finally completed the overall plan of the pose monitoring system design.(2)Analyzed the environment and imaging characteristics of the tunneling face,proposed an improved non-local mean denoising image preprocessing algorithm,designed a feature extraction algorithm according to the selected target features,and improved the extraction accuracy according to the characteristics of the feature neighborhood up to the sub-pixel level,the image preprocessing effect and feature extraction accuracy were verified through simulation.(3)Aiming at the problem of the posture monitoring of the roadheader,an improved particle swarm algorithm was designed.By adding two improvement measures of dynamic inertia weight and Lévy flight,the optimization speed and search accuracy of the algorithm were improved,and the application of the particle swarm algorithm would be improved.In the pose calculation of the roadheader fuselage,the accuracy of the pose calculation is improved.(4)The experimental platform of the roadheader's body pose monitoring system was built,the roadheader's virtual model was established,and the monitoring software was developed to verify the image preprocessing algorithm and the accuracy of the roadheader's body pose.Experiments show that the proposed improved non-local mean denoising algorithm can effectively improve the image quality,the designed feature point extraction algorithm can achieve accurate extraction of the reference target feature points,and the accuracy of the roadheader's body pose can meet actual needs.The thesis includes 64 pictures,23 tables,and 82 references.