UR dataset

In order for robots to navigate successfully, they need to correctly estimate the traversability of their surroundings. To do so, an orthogonal projection of the spatial obstacles perceived by the robot's sensors is usually utilized. This approach however is challenging, as the sensors can only see the surfaces closest to them, which makes it difficult to estimate the size and shape of the obstacle, effectively prohibiting immediate estimation of occupied space.In this work, we introduce a novel approach that can estimate the obstacle's 2D footprint straight from incomplete 3D data in the form of a point cloud. Unlike existing point cloud completion methods, our approach does not require the reconstruction of the entire 3D object and its projection. Instead, it focuses on rendering a navigation-relevant 2D representation directly from segmented sensor scans with sparse points. At its core, we propose a lightweight, multi-modal autoencoder that takes an input of a voxelized incomplete point cloud and outputs an estimated footprint that is directly applicable to the occupancy grid.We evaluate our method on an open-source synthetic dataset to compare it with the results of point cloud completion algorithms projected on a 2D plane. We also validate the method on a real dataset coined UR, which was collected specifically for this publication, and prove the method's applicability in real-life scenarios. Compared to other pipelines, our method proves superior in terms of accuracy in most cases and achieves the lowest computational resource consumption in complete robotic pipeline tests. Subsequently, UR dataset tests demonstrate favorable results in a real-life scenario.