Research On RKSF-RUKF Aided Error Suppression For Coal Mine Personnel Ranging And Positioning

Coal occupies a major position in Chinese primary energy production and consumption structure.As one of the "six safety avoidance systems",the coal mine personnel positioning system has played an important role in coal mine production management and accident emergency rescue.However,the current mainstream coal mine personnel positioning methods in my country have relatively large errors.Therefore,it is of great significance that in-depth research on the causes and suppression methods of personnel positioning errors in coal mines for improving the accuracy of mine personnel positioning and ensuring coal mine safety production.The method of personnel positioning in coal mines is treated as the research object in this paper.After analyzing and comparing the current mainstream wireless positioning methods,based on the time-of-arrival(TOA)ranging method,firstly conducting an in-depth analysis of the ranging error under this method,it is concluded that the non-Gaussian distributed non-line-of-sight error(NLOS)is the main source of the personnel ranging error under the symmetric double-sided two way ranging communication mode.In order to suppress the influence of the non-line-of-sight error on the ranging result,it is proposed to base on the Gaussian sum filtering theory robust Kalman sum filtering(RKSF)ranging optimization method.Secondly,the influence of the special geometric dimensions of the roadway is analyzed on the traditional personnel positioning method.In the roadway environment where the aspect ratio tends to be infinite,the positioning error is mainly borne by the radial direction of the roadway.In order to adapt to this characteristic of roadway positioning error and enhance the robustness of the positioning algorithm to abnormal observations,a robust unscented Kalman filter(RUKF)positioning optimization method based on hypothesis testing is proposed.Finally,a method for locating people underground in coal mines based on the RKSF-RUKF model is constructed.For the purpose of verifying the effectiveness of the proposed method,a simulation experiment based on Robot Operating System(ROS)and Gazebo platform was designed and implemented,and the three most common positioning scenarios in underground coal mine positioning were compared and analyzed.Through simulation experiments,in the line-of-sight environment,the method in this paper is not far from the traditional method;in the non-line-of-sight environment,the method in this paper has a good error suppression ability,which verifies the effectiveness of the method in this paper and improves the positioning accuracy of mine personnel,guaranteeing the safe and efficient production of coal mines,and assisting mine emergency rescue,provide a certain reference significance.