Lane graph estimation is an essential and highly challenging task in automated driving and HD map learning. Existing methods using either onboard or aerial imagery struggle with complex lane topologies, out-of-distribution scenarios, or significant occlusions in the image space. Moreover, merging overlapping lane graphs to obtain consistent large-scale graphs remains difficult. To overcome these challenges, we propose a novel bottom-up approach to lane graph estimation from aerial imagery that aggregates multiple overlapping graphs into a single consistent graph. Due to its modular design, our method allows us to address two complementary tasks: predicting ego-respective successor lane graphs from arbitrary vehicle positions using a graph neural network and aggregating these predictions into a consistent global lane graph. Extensive experiments on a large-scale lane graph dataset demonstrate that our approach yields highly accurate lane graphs, even in regions with severe occlusions. The presented approach to graph aggregation proves to eliminate inconsistent predictions while increasing the overall graph quality. We make our large-scale urban lane graph dataset and code publicly available at http://urbanlanegraph.cs.uni-freiburg.de.

Robots operating in the open world encounter various different environments that can substantially differ from each other. This domain gap also poses a challenge for Simultaneous Localization and Mapping (SLAM) being one of the fundamental tasks for navigation. In particular, learning-based SLAM methods are known to generalize poorly to unseen environments hindering their general adoption. In this work, we introduce the novel task of continual SLAM extending the concept of lifelong SLAM from a single dynamically changing environment to sequential deployments in several drastically differing environments. To address this task, we propose CL-SLAM leveraging a dual-network architecture to both adapt to new environments and retain knowledge with respect to previously visited environments. We compare CL-SLAM to learning-based as well as classical SLAM methods and show the advantages of leveraging online data. We extensively evaluate CL-SLAM on three different datasets and demonstrate that it outperforms several baselines inspired by existing continual learning-based visual odometry methods. We make the code of our work publicly available at http://continual-slam.cs.uni-freiburg.de.

Bird's-Eye-View (BEV) semantic maps have become an essential component of automated driving pipelines due to the rich representation they provide for decision-making tasks. However, existing approaches for generating these maps still follow a fully supervised training paradigm and hence rely on large amounts of annotated BEV data. In this work, we address this limitation by proposing the first self-supervised approach for generating a BEV semantic map using a single monocular image from the frontal view (FV). During training, we overcome the need for BEV ground truth annotations by leveraging the more easily available FV semantic annotations of video sequences. Thus, we propose the SkyEye architecture that learns based on two modes of self-supervision, namely, implicit supervision and explicit supervision. Implicit supervision trains the model by enforcing spatial consistency of the scene over time based on FV semantic sequences, while explicit supervision exploits BEV pseudolabels generated from FV semantic annotations and self-supervised depth estimates. Extensive evaluations on the KITTI-360 dataset demonstrate that our self-supervised approach performs on par with the state-of-the-art fully supervised methods and achieves competitive results using only 1% of direct supervision in the BEV compared to fully supervised approaches. Finally, we publicly release both our code and the BEV datasets generated from the KITTI-360 and Waymo datasets.

Audio-visual navigation combines sight and hearing to navigate to a sound-emitting source in an unmapped environment. While recent approaches have demonstrated the benefits of audio input to detect and find the goal, they focus on clean and static sound sources and struggle to generalize to unheard sounds. In this work, we propose the novel dynamic audio-visual navigation benchmark which requires catching a moving sound source in an environment with noisy and distracting sounds, posing a range of new challenges. We introduce a reinforcement learning approach that learns a robust navigation policy for these complex settings. To achieve this, we propose an architecture that fuses audio-visual information in the spatial feature space to learn correlations of geometric information inherent in both local maps and audio signals. We demonstrate that our approach consistently outperforms the current state-of-the-art by a large margin across all tasks of moving sounds, unheard sounds, and noisy environments, on two challenging 3D scanned real-world environments, namely Matterport3D and Replica. The benchmark is available at http://dav-nav.cs.uni-freiburg.de.

Abstract--As robotic systems become more and more capable of assisting humans in their everyday lives, we must consider the opportunities for these artificial agents to make their human collaborators feel unsafe or to treat them unfairly. Robots can exhibit antisocial behavior causing physical harm to people or reproduce unfair behavior replicating and even amplifying historical and societal biases which are detrimental to humans they interact with. In this paper, we discuss these issues considering sociable robotic manipulation and fair robotic decision making. We propose a novel approach to learning fair and sociable behavior, not by reproducing positive behavior, but rather by avoiding negative behavior. In this study, we highlight the importance of incorporating sociability in robot manipulation, as well as the need to consider fairness in human-robot interactions.

Panoptic scene understanding and tracking of dynamic agents are essential for robots and automated vehicles to navigate in urban environments. As LiDARs provide accurate illumination-independent geometric depictions of the scene, performing these tasks using LiDAR point clouds provides reliable predictions. However, existing datasets lack diversity in the type of urban scenes and have a limited number of dynamic object instances which hinders both learning of these tasks as well as credible benchmarking of the developed methods. In this paper, we introduce the large-scale Panoptic nuScenes benchmark dataset that extends our popular nuScenes dataset with point-wise groundtruth annotations for semantic segmentation, panoptic segmentation, and panoptic tracking tasks. To facilitate comparison, we provide several strong baselines for each of these tasks on our proposed dataset. Moreover, we analyze the drawbacks of the existing metrics for panoptic tracking and propose the novel instance-centric PAT metric that addresses the concerns. We present exhaustive experiments that demonstrate the utility of Panoptic nuScenes compared to existing datasets and make the online evaluation server available at nuScenes.org. We believe that this extension will accelerate the research of novel methods for scene understanding of dynamic urban environments.

Multi-view classification is inspired by the behavior of humans, especially when fine-grained features or in our case rarely occurring anomalies are to be detected. Current contributions point to the problem of how high-dimensional data can be fused. In this work, we build upon the deep support vector data description algorithm and address multi-perspective anomaly detection using three different fusion techniques i.e. early fusion, late fusion, and late fusion with multiple decoders. We employ different augmentation techniques with a denoising process to deal with scarce one-class data, which further improves the performance (ROC AUC = 80\%). Furthermore, we introduce the dices dataset that consists of over 2000 grayscale images of falling dices from multiple perspectives, with 5\% of the images containing rare anomalies (e.g. drill holes, sawing, or scratches). We evaluate our approach on the new dices dataset using images from two different perspectives and also benchmark on the standard MNIST dataset. Extensive experiments demonstrate that our proposed approach exceeds the state-of-the-art on both the MNIST and dices datasets. To the best of our knowledge, this is the first work that focuses on addressing multi-perspective anomaly detection in images by jointly using different perspectives together with one single objective function for anomaly detection.

Unmanned Aerial Vehicles (UAVs) equipped with bioradars are a life-saving technology that can enable identification of survivors under collapsed buildings in the aftermath of natural disasters such as earthquakes or gas explosions. However, these UAVs have to be able to autonomously navigate in disaster struck environments and land on debris piles in order to accurately locate the survivors. This problem is extremely challenging as pre-existing maps cannot be leveraged for navigation due to structural changes that may have occurred. Furthermore, existing landing site detection algorithms are not suitable to identify safe landing regions on debris piles. In this work, we present a computationally efficient system for autonomous UAV navigation and landing that does not require any prior knowledge about the environment. We propose a novel landing site detection algorithm that computes costmaps based on several hazard factors including terrain flatness, steepness, depth accuracy, and energy consumption information. We also introduce a first-of-a-kind synthetic dataset of over 1.2 million images of collapsed buildings with groundtruth depth, surface normals, semantics and camera pose information. We demonstrate the efficacy of our system using experiments from a city scale hyperrealistic simulation environment and in real-world scenarios with collapsed buildings.