Study On Environmental Perception System Of Coal Mine Ground Rail Transportation

As an important part of the coal mine transportation system,the coal mine ground rail transportation plays a key role of the coal mine.Its safe and efficient operation is directly related to the efficiency of the coal mine production and transportation.At present,the operation of the tramcar mainly depends on manual control.The operation efficiency is low,and the transportation safety factor is not high.In addition,the human and material resources are wasted seriously.In view of this,in this thesis,the machine vision and millimeter wave radar information were combined,and a set of coal mine ground rail transportation environment perception system was designed,to realize the automatic real-time detection and tracking of obstacles in front of the running tramcar.The main work of this thesis includes the following parts:According to the functional requirements of the environmental awareness system of coal mine ground rail transportation,the hardware selection and installation of sensor equipment were completed after the analysis and comparison of the advantages and disadvantages of a variety of radar and machine vision sensors.The hardware structure design idea was provided,and the overall research scheme of the system was formulated.Obstacle detection of coal mine ground track based on machine vision: Firstly,the image graying,image denoising and image enhancement methods were used to complete the image preprocessing.Secondly,Canny operator was used to complete the edge detection of coal mine ground track,and the segmented curve model and cubic spline fitting were combined to extract the track limit.Finally,the recognition model based on YOLOv3 algorithm was constructed to detect obstacles in front of the running tramcar under different categories and light brightness.Obstacle detection and tracking in coal mine ground track based on millimeter wave radar: Firstly,according to the set of distance threshold and shortest distance principle,the effective target was selected initially.Secondly,the effective target was selected accurately by the method of effective life cycle.Finally,the Kalman filter algorithm was used to achieves effective target tracking and improves the accuracy of radar detection and tracking of obstacles.The thesis introduced the idea of information fusion,and analyzed the detection results of machine vision and millimeter wave radar technology.The final detectionresults of the coal mine ground rail transportation environment perception system were obtained through the fusion of two kinds of data space and time.Based on the above research,using MATLAB and Microsoft Visual Studio,a set of environment awareness system software for coal mine ground rail transportation was designed,and the system interface design and construction were completed.This thesis has 64 figures,14 tables and 83 references.