Place recognition in 3D data is a challenging task that has been commonly approached by adapting imagebased solutions. Methods based on local features suffer from ambiguity and from robustness to environment changes while methods based on global features are viewpoint dependent. We propose SegMatch, a reliable place recognition algorithm based on the matching of 3D segments. Segments provide a good compromise between local and global descriptions, incorporating their strengths while reducing their individual drawbacks. SegMatch does not rely on assumptions of 'perfect segmentation', or on the existence of 'objects' in the environment, which allows for reliable execution on large scale, unstructured environments. We quantitatively demonstrate that SegMatch can achieve accurate localization at a frequency of 1Hz on the largest sequence of the KITTI odometry dataset. We furthermore show how this algorithm can reliably detect and close loops in real-time, during online operation. In addition, the source code for the SegMatch algorithm is made publicly available 1 .

Precisely estimating a robot’s pose in a prior, global map is a fundamental capability for mobile robotics, e.g., autonomous driving or exploration in disaster zones. This task, however, remains challenging in unstructured, dynamic environments, where local features are not discriminative enough and global scene descriptors only provide coarse information. We therefore present SegMap: a map representation solution for localization and mapping based on the extraction of segments in 3D point clouds. Working at the level of segments offers increased invariance to view-point and local structural changes, and facilitates real-time processing of large-scale 3D data. SegMap exploits a single compact data-driven descriptor for performing multiple tasks: global localization, 3D dense map reconstruction, and semantic information extraction. The performance of SegMap is evaluated in multiple urban driving and search and rescue experiments. We show that the learned SegMap descriptor has superior segment retrieval capabilities, compared with state-of-the-art handcrafted descriptors. As a consequence, we achieve a higher localization accuracy and a 6% increase in recall over state-of-the-art handcrafted descriptors. These segment-based localizations allow us to reduce the open-loop odometry drift by up to 50%. SegMap is open-source available along with easy to run demonstrations.

When performing localization and mapping, working at the level of structure can be advantageous in terms of robustness to environmental changes and differences in illumination. This paper presents SegMap: a map representation solution to the localization and mapping problem based on the extraction of segments in 3D point clouds. In addition to facilitating the computationally intensive task of processing 3D point clouds, working at the level of segments addresses the data compression requirements of real-time single-and multi-robot systems. While current methods extract descriptors for the single task of localization, SegMap leverages a data-driven descriptor in order to extract meaningful features that can also be used for reconstructing a dense 3D map of the environment and for extracting semantic information. This is particularly interesting for navigation tasks and for providing visual feedback to endusers such as robot operators, for example in search and rescue scenarios. These capabilities are demonstrated in multiple urban driving and search and rescue experiments. Our method leads to an increase of area under the ROC curve of 28.3% over current state of the art using eigenvalue based features. We also obtain very similar reconstruction capabilities to a model specifically trained for this task. The SegMap implementation is available open-source along with easy to run demonstrations at www.github.com/ethz-asl/segmap.

Fossil soils and grasses from the well-known Miocene mammal locality of Fort Ternan, southwestern Kenya, are evidence of a mosaic of grassy woodland and wooded grassland some 14 million years ago. This most ancient wooded grassland yet known on the African continent supported more abundant and diverse antelopes than known earlier in Africa. Ape fossils at Fort Ternan, including Kenyapithecus wickeri, were associated with woodland parts of the vegetation mosaic revealed by paleosols. Grassland habitats were available in East Africa long before the evolutionary divergence of apes and humans some 5 to 10 million years ago.

At the well-known fossil mammal locality of Fort Ternan in southwestern Kenya, radiometrically dated at about 14 million years old (middle Miocene), fossil grasses have been preserved by nephelinitic sandstone in place of growth above a brown paleosol (type Onuria clay). Large portions of grass plants as well as fragments of leaves have revealed details of silica bodies, stomates, and other taxonomically important features under the scanning electron microscope. The computer database for grass identification compiled by Leslie Watson and colleagues was used to determine the most similar living grass genera to the five distinct kinds of fossil found. Two of the fossil species are assigned to Cleistochloa kabuyis sp. nov. and C. shipmanae sp. nov. This genus includes one species from low fertility dry woodland soils of New South Wales and Queensland and a second species from “raw clay soils” in western New Guinea. A third fossil species, represented by a large portion of a branching culm, is assigned to Stereochlaena miocenica sp. nov. This genus includes five species of low-fertility woodland soils in southeastern Africa. Both Cleistochloa and Stereochlaena are in the supertribe Panicanae of the subfamily Panicoideae. A fourth species is assigned to Distichlis africana sp. nov. and provides a biogeographic link between the single species of this genus now living in coastal grasslands in southeastern Australia and the 12 species of dunes and deserts found throughout the Americas from Patagonia and the West Indies to the United States and Canada. A fifth species is, like D. africana, in the subfamily Chloridoideae, but its stomata were not seen and it could belong to Cyclostachya, Pogoneura, or Polevansia. This earliest known wooded grassland flora in Africa is taxonomically unlike the modern grass flora of fertile volcanic African landscapes, and may have been recruited from an archaic grass flora of Gondwanan desert and lateritic soils.