Chen De, Yan Qingdong, Zhou Junxiong, and Du Yixian
Point clouds, clustering, LiDAR, localisation, mobile robot (MR)
Mobile robot (MR) is favoured because of the broad application prospects. This work proposes a localisation method based on the density-based spatial clustering of applications with noise (DBSCAN) algorithm. The main purpose is to direct localisation based on the coordinates of the connected wharves legs point clouds, as well as solve the problem of low localisation accuracy or localisation failure caused by the influence of miscellaneous point clouds such as obstacles in the operating environment and so on. In this method, four filters are designed to preprocess the initial point clouds data obtained by LiDAR, to delete a large amount of useless or noise point clouds to obtain the point clouds data with large contribution to the algorithm, and improve the computational efficiency at the same time. The practical application experiments are employed and the results show that the proposed methodology can effectively realise the MR localisation function and offer the advantages of high localisation accuracy and low dependence on environmental stability in complex work environment.
Important Links:
Go Back
