Research On The Method Of Multi-parameter System Error Calibration Of Airborne LiDAR

As a new measurement technology, airborne LiDAR can rapidly abtain threedimensional(3-D) geo-spatial information with high temporal-spatial resolution, which owns its unique advantages and widely application prospect. Compared with the hardware technology of airborne LiDAR, the development of its data post-processing is slower. Specifically, eliminating the systematic errors has been becoming a key problem which limits its application. In order to acquire high-precision airborne LiDAR data, it is of great theoretic and practical value to research the methods of the elimination of systematic errors. In terms of the problem, this paper has done some researches and experiments, mainly including:(1)Through the geometric positioning equation and simulation experiment, qualitively and quantitively analyze the influences caused by the systematic errors on the airborne LiDAR data. And summarize the methods of evaluating relative and absolute accuracy of LiDAR data, including elevation and level.(2)Based on the number of calibration parameters, propose three multi-parameter system calibration models(three, six and eight parameters, respectively). According to the criterion that coordinate values of conjugate points in overlapping strips are equal, we can use the geometric positioning equation to establish the error equations, calculate the calibration parameters, and then Li DAR strips can be adjusted. Meanwhile, we put forward an integrated workflow of the proposed calibration methods.(3)Using two sets of data, conduct relative experiments to research these three multi-parameter system calibration models(three and six parameters, respectively). And then evaluate and analyze the calibration accuracy.The result indicated that the above methods are capable of eliminating the systematic errors, which can get the same results as manual calibration and cater to the requirement of engineering production. Among them, the accuracy of six parameters calibration model is slightly better than three parameters model.