Method on pose estimation of excavators based on onboard depth camera

Abstract Pose estimation of construction machines describes their motions and orientations in 3D space, which provides basic information for remote control, safety monitoring, productivity analysis. Among the commonly used machines, excavator is an important one whose pose has significant value development automatic driving systems operation monitoring. Regarding vision-based excavators, previous studies mainly rely on image data from digital cameras installed sites. However, robustness may drop due to occlusions by surrounding clutters. To address these problems fundamentally, this study proposes a method partial with onboard depth camera cabin targeted excavator. This solution inspired visual effect produced operator’s eyes, according Bionics principles. First, processing data, spatial coordinates several pre-defined keypoints are obtained. Afterward, combining keypoint prior knowledge geometric relationship triangle computed Finally, feasibility proposed algorithm preliminarily verified experimental real contributes not only solve problem occlusion when using sensors sites, but also provide theoretical basis design control excavators.