The Hilti SLAM Challenge Dataset

Research in Simultaneous Localization and Mapping (SLAM) has made outstanding progress over the past years. SLAM systems are nowadays transitioning from academic to real world applications. However, this transition posed new demanding challenges terms of accuracy robustness. To develop that can address these challenges, datasets containing cutting-edge hardware realistic scenarios required. We propose Hilti Challenge Dataset . Our dataset contains indoor sequences offices, labs, construction environments outdoor sites parking areas. All characterized by featureless areas varying illumination conditions typical real-world pose great algorithms have been developed confined lab environments. Accurate sparse ground truth, at millimeter level, is provided for each sequence. The sensor platform used record data includes a number visual, lidar, inertial sensors, which spatially temporally calibrated. purpose foster research fusion be deployed tasks where high robustness required, xmlns:xlink="http://www.w3.org/1999/xlink">e.g. , Many industrial groups tested their on proposed Challenge results challenge, summarized paper, show an important asset development ready real-world.