Reliable and accurate localization and mapping are key components of most autonomous systems. Besides geometric information about the mapped environment, the semantics plays an important role to enable intelligent navigation behaviors. In most realistic environments, this task is particularly complicated due to dynamics caused by moving objects, which can corrupt the mapping step or derail localization. In this paper, we propose an extension of a recently published surfelbased mapping approach exploiting three-dimensional laser range scans by integrating semantic information to facilitate the mapping process. The semantic information is efficiently extracted by a fully convolutional neural network and rendered on a spherical projection of the laser range data. This computed semantic segmentation results in point-wise labels for the whole scan, allowing us to build a semantically-enriched map with labeled surfels. This semantic map enables us to reliably filter moving objects, but also improve the projective scan matching via semantic constraints. Our experimental evaluation on challenging highways sequences from KITTI dataset with very few static structures and a large amount of moving cars shows the advantage of our semantic SLAM approach in comparison to a purely geometric, state-of-the-art approach.
Mapping and localization are essential capabilities of robotic systems. Although the majority of mapping systems focus on static environments, the deployment in real-world situations requires them to handle dynamic objects. In this paper, we propose an approach for an RGB-D sensor that is able to consistently map scenes containing multiple dynamic elements. For localization and mapping, we employ an efficient direct tracking on the truncated signed distance function (TSDF) and leverage color information encoded in the TSDF to estimate the pose of the sensor. The TSDF is efficiently represented using voxel hashing, with most computations parallelized on a GPU. For detecting dynamics, we exploit the residuals obtained after an initial registration, together with the explicit modeling of free space in the model. We evaluate our approach on existing datasets, and provide a new dataset showing highly dynamic scenes. These experiments show that our approach often surpass other state-of-the-art dense SLAM methods. We make available our dataset with the ground truth for both the trajectory of the RGB-D sensor obtained by a motion capture system and the model of the static environment using a high-precision terrestrial laser scanner. Finally, we release our approach as open source code.
Micro aerial vehicles (MAVs) are an excellent platform for autonomous exploration. Most MAVs rely mainly on cameras for buliding a map of the 3D environment. Therefore, vision-based MAVs require an efficient exploration algorithm to select viewpoints that provide informative measurements. In this paper, we propose an exploration approach that selects in real time the next-best-view that maximizes the expected information gain of new measurements. In addition, we take into account the cost of reaching a new viewpoint in terms of distance and predictability of the flight path for a human observer. Finally, our approach selects a path that reduces the risk of crashes when the expected battery life comes to an end, while still maximizing the information gain in the process. We implemented and thoroughly tested our approach and the experiments show that it offers an improved performance compared to other state-of-the-art algorithms in terms of precision of the reconstruction, execution time, and smoothness of the path.
ABSTRACT:Most micro aerial vehicles (MAV) are flown manually by a pilot. When it comes to autonomous exploration for MAVs equipped with cameras, we need a good exploration strategy for covering an unknown 3D environment in order to build an accurate map of the scene. In particular, the robot must select appropriate viewpoints to acquire informative measurements. In this paper, we present an approach that computes in real-time a smooth flight path with the exploration of a 3D environment using a vision-based MAV. We assume to know a bounding box of the object or building to explore and our approach iteratively computes the next best viewpoints using a utility function that considers the expected information gain of new measurements, the distance between viewpoints, and the smoothness of the flight trajectories. In addition, the algorithm takes into account the elapsed time of the exploration run to safely land the MAV at its starting point after a user specified time. We implemented our algorithm and our experiments suggest that it allows for a precise reconstruction of the 3D environment while guiding the robot smoothly through the scene.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with đź’™ for researchers
Part of the Research Solutions Family.