Shearer positioning method based on anti-disturbance cubature Kalman Filter

MA Rui; SONG Jiancheng; GENG Pulong; TIAN Muqin; SONG Danyang; XU Chunyu; CHANG Mengxiao; ZHANG Jie; LI Lei

doi:10.11799/ce202510019

Abstract：Traditional positioning techniques integrated on coal shearers primarily depend on inertial navigation units and odometers.However,in the intricate environment of fully mechanized mining faces,inertial navigation is prone to accumulated deviations,while the coal shearer may encounter abnormal phenomena such as slippage during operation,leading to inaccurate odometer data and further diminishing positioning precision.To address this issue,we introduce an integrated positioning technique based on robust cubature Kalman filtering.By loosely coupling the strapdown inertial navigation unit with the odometer,the speed and movement distance data provided by the odometer are merged with the computed results of the inertial navigation unit,serving as deviation observations input into the filtering algorithm for calibration.Additionally,the filtering algorithm incorporates robust processing based on innovation chi-square testing,constructing a variance inflation factor to adaptively adjust the observation covariance matrix,thereby enhancing the robust performance and deviation suppression capability of the cubature Kalman filter.The filtered results are fed back into the system as deviation corrections,and finally,the validity of the integrated state is evaluated after the entire computation is completed.Simulation and experimental results demonstrate that this algorithm can effectively mitigate the impact of outliers,achieving a positioning precision of 0.1 meter when employing this integrated navigation algorithm.