César Cadena

Samuel Looper, Javier Rodriguez-Puigvert, Roland Siegwart et al.

Numerous applications require robots to operate in environments shared with other agents, such as humans or other robots. However, such shared scenes are typically subject to different kinds of long-term semantic scene changes. The ability to model and predict such changes is thus crucial for robot autonomy. In this work, we formalize the task of semantic scene variability estimation and identify three main varieties of semantic scene change: changes in the position of an object, its semantic state, or the composition of a scene as a whole. To represent this variability, we propose the Variable Scene Graph (VSG), which augments existing 3D Scene Graph (SG) representations with the variability attribute, representing the likelihood of discrete long-term change events. We present a novel method, DeltaVSG, to estimate the variability of VSGs in a supervised fashion. We evaluate our method on the 3RScan long-term dataset, showing notable improvements in this novel task over existing approaches. Our method DeltaVSG achieves an accuracy of 77.1% and a recall of 72.3%, often mimicking human intuition about how indoor scenes change over time. We further show the utility of VSG prediction in the task of active robotic change detection, speeding up task completion by 66.0% compared to a scene-change-unaware planner. We make our code available as open-source.

Chunwei Xing, Xinyu Sun, Andrei Cramariuc et al. · eth-zurich

Current descriptors for global localization often struggle under vast viewpoint or appearance changes. One possible improvement is the addition of topological information on semantic objects. However, handcrafted topological descriptors are hard to tune and not robust to environmental noise, drastic perspective changes, object occlusion or misdetections. To solve this problem, we formulate a learning-based approach by modelling semantically meaningful object constellations as graphs and using Deep Graph Convolution Networks to map a constellation to a descriptor. We demonstrate the effectiveness of our Deep Learned Constellation Descriptor (Descriptellation) on two real-world datasets. Although Descriptellation is trained on randomly generated simulation datasets, it shows good generalization abilities on real-world datasets. Descriptellation also outperforms state-of-the-art and handcrafted constellation descriptors for global localization, and is robust to different types of noise. The code is publicly available at https://github.com/ethz-asl/Descriptellation.

Lukas Schmid, Mansoor Nasir Cheema, Victor Reijgwart et al.

Exploration of unknown environments is a fundamental problem in robotics and an essential component in numerous applications of autonomous systems. A major challenge in exploring unknown environments is that the robot has to plan with the limited information available at each time step. While most current approaches rely on heuristics and assumption to plan paths based on these partial observations, we instead propose a novel way to integrate deep learning into exploration by leveraging 3D scene completion for informed, safe, and interpretable exploration mapping and planning. Our approach, SC-Explorer, combines scene completion using a novel incremental fusion mechanism and a newly proposed hierarchical multi-layer mapping approach, to guarantee safety and efficiency of the robot. We further present an informative path planning method, leveraging the capabilities of our mapping approach and a novel scene-completion-aware information gain. While our method is generally applicable, we evaluate it in the use case of a Micro Aerial Vehicle (MAV). We thoroughly study each component in high-fidelity simulation experiments using only mobile hardware, and show that our method can speed up coverage of an environment by 73% compared to the baselines with only minimal reduction in map accuracy. Even if scene completions are not included in the final map, we show that they can be used to guide the robot to choose more informative paths, speeding up the measurement of the scene with the robot's sensors by 35%. We validate our system on a fully autonomous MAV, showing rapid and reliable scene coverage even in a complex cluttered environment. We make our methods available as open-source.

Hermann Blum, Marcus G. Müller, Abel Gawel et al.

In order to operate in human environments, a robot's semantic perception has to overcome open-world challenges such as novel objects and domain gaps. Autonomous deployment to such environments therefore requires robots to update their knowledge and learn without supervision. We investigate how a robot can autonomously discover novel semantic classes and improve accuracy on known classes when exploring an unknown environment. To this end, we develop a general framework for mapping and clustering that we then use to generate a self-supervised learning signal to update a semantic segmentation model. In particular, we show how clustering parameters can be optimized during deployment and that fusion of multiple observation modalities improves novel object discovery compared to prior work. Models, data, and implementations can be found at https://github.com/hermannsblum/scim

Landson Guo, Andres M. Diaz Aguilar, William Talbot et al.

Accurate point-wise velocity estimation in 3D is crucial for robot interaction with non-rigid dynamic agents, enabling robust performance in path planning, collision avoidance, and object manipulation in dynamic environments. To this end, this paper proposes a novel RADAR, LiDAR, and camera fusion pipeline for point-wise 3D velocity estimation named CaRLi-V. This pipeline leverages raw RADAR measurements to create a novel RADAR representation, the velocity cube, which densely encodes RADAR radial velocities. By combining the velocity cube for radial velocity extraction, optical flow for tangential velocity estimation, and LiDAR for point-wise range measurements through a closed-form solution, our approach can produce 3D velocity estimates for a dense array of points. Developed as an open-source ROS2 package, CaRLi-V has been field-tested on a custom dataset and achieves low velocity error metrics relative to ground truth while outperforming state-of-the-art scene flow methods.

Zhizheng Liu, Francesco Milano, Jonas Frey et al.

An increasing amount of applications rely on data-driven models that are deployed for perception tasks across a sequence of scenes. Due to the mismatch between training and deployment data, adapting the model on the new scenes is often crucial to obtain good performance. In this work, we study continual multi-scene adaptation for the task of semantic segmentation, assuming that no ground-truth labels are available during deployment and that performance on the previous scenes should be maintained. We propose training a Semantic-NeRF network for each scene by fusing the predictions of a segmentation model and then using the view-consistent rendered semantic labels as pseudo-labels to adapt the model. Through joint training with the segmentation model, the Semantic-NeRF model effectively enables 2D-3D knowledge transfer. Furthermore, due to its compact size, it can be stored in a long-term memory and subsequently used to render data from arbitrary viewpoints to reduce forgetting. We evaluate our approach on ScanNet, where we outperform both a voxel-based baseline and a state-of-the-art unsupervised domain adaptation method.

Michael Pantic, Cesar Cadena, Roland Siegwart et al.

This work investigates the use of Neural implicit representations, specifically Neural Radiance Fields (NeRF), for geometrical queries and motion planning. We show that by adding the capacity to infer occupancy in a radius to a pre-trained NeRF, we are effectively learning an approximation to a Euclidean Signed Distance Field (ESDF). Using backward differentiation of the augmented network, we obtain an obstacle gradient that is integrated into an obstacle avoidance policy based on the Riemannian Motion Policies (RMP) framework. Thus, our findings allow for very fast sampling-free obstacle avoidance planning in the implicit representation.

Matthias Brucker, Andrei Cramariuc, Cornelius von Einem et al.

Reliable obstacle detection on railways could help prevent collisions that result in injuries and potentially damage or derail the train. Unfortunately, generic object detectors do not have enough classes to account for all possible scenarios, and datasets featuring objects on railways are challenging to obtain. We propose utilizing a shallow network to learn railway segmentation from normal railway images. The limited receptive field of the network prevents overconfident predictions and allows the network to focus on the locally very distinct and repetitive patterns of the railway environment. Additionally, we explore the controlled inclusion of global information by learning to hallucinate obstacle-free images. We evaluate our method on a custom dataset featuring railway images with artificially augmented obstacles. Our proposed method outperforms other learning-based baseline methods.

Jin Jin, Chong Zhang, Jonas Frey et al.

Autonomous robots must navigate reliably in unknown environments even under compromised exteroceptive perception, or perception failures. Such failures often occur when harsh environments lead to degraded sensing, or when the perception algorithm misinterprets the scene due to limited generalization. In this paper, we model perception failures as invisible obstacles and pits, and train a reinforcement learning (RL) based local navigation policy to guide our legged robot. Unlike previous works relying on heuristics and anomaly detection to update navigational information, we train our navigation policy to reconstruct the environment information in the latent space from corrupted perception and react to perception failures end-to-end. To this end, we incorporate both proprioception and exteroception into our policy inputs, thereby enabling the policy to sense collisions on different body parts and pits, prompting corresponding reactions. We validate our approach in simulation and on the real quadruped robot ANYmal running in real-time (<10 ms CPU inference). In a quantitative comparison with existing heuristic-based locally reactive planners, our policy increases the success rate over 30% when facing perception failures. Project Page: https://bit.ly/45NBTuh.

Guido Caccianiga, Julian Nubert, Cesar Cadena et al.

Reconstructing the 3D shape of a deformable environment from the information captured by a moving depth camera is highly relevant to surgery. The underlying challenge is the fact that simultaneously estimating camera motion and tissue deformation in a fully deformable scene is an ill-posed problem, especially from a single arbitrarily moving viewpoint. Current solutions are often organ-specific and lack the robustness required to handle large deformations. Here we propose a multi-viewpoint global optimization framework that can flexibly integrate the output of low-level perception modules (data association, depth, and relative scene flow) with kinematic and scene-modeling priors to jointly estimate multiple camera motions and absolute scene flow. We use simulated noisy data to show three practical examples that successfully constrain the convergence to a unique solution. Overall, our method shows robustness to combined noisy input measures and can process hundreds of points in a few milliseconds. MultiViPerFrOG builds a generalized learning-free scaffolding for spatio-temporal encoding that can unlock advanced surgical scene representations and will facilitate the development of the computer-assisted-surgery technologies of the future.

Mike Zhang, Kaixian Qu, Vaishakh Patil et al.

Large Language Models (LLM) have emerged as a tool for robots to generate task plans using common sense reasoning. For the LLM to generate actionable plans, scene context must be provided, often through a map. Recent works have shifted from explicit maps with fixed semantic classes to implicit open vocabulary maps based on queryable embeddings capable of representing any semantic class. However, embeddings cannot directly report the scene context as they are implicit, requiring further processing for LLM integration. To address this, we propose an explicit text-based map that can represent thousands of semantic classes while easily integrating with LLMs due to their text-based nature by building upon large-scale image recognition models. We study how entities in our map can be localized and show through evaluations that our text-based map localizations perform comparably to those from open vocabulary maps while using two to four orders of magnitude less memory. Real-robot experiments demonstrate the grounding of an LLM with the text-based map to solve user tasks.

Julian Nubert, Turcan Tuna, Jonas Frey et al.

Seamless operation of mobile robots in challenging environments requires low-latency local motion estimation (e.g., dynamic maneuvers) and accurate global localization (e.g., wayfinding). While most existing sensor-fusion approaches are designed for specific scenarios, this work introduces a flexible open-source solution for task- and setup-agnostic multimodal sensor fusion that is distinguished by its generality and usability. Holistic Fusion formulates sensor fusion as a combined estimation problem of i) the local and global robot state and ii) a (theoretically unlimited) number of dynamic context variables, including automatic alignment of reference frames; this formulation fits countless real-world applications without any conceptual modifications. The proposed factor-graph solution enables the direct fusion of an arbitrary number of absolute, local, and landmark measurements expressed with respect to different reference frames by explicitly including them as states in the optimization and modeling their evolution as random walks. Moreover, local smoothness and consistency receive particular attention to prevent jumps in the robot state belief. HF enables low-latency and smooth online state estimation on typical robot hardware while simultaneously providing low-drift global localization at the IMU measurement rate. The efficacy of this released framework is demonstrated in five real-world scenarios on three robotic platforms, each with distinct task requirements.

Christopher E. Mower, Yuhui Wan, Hongzhan Yu et al.

We present a framework for intuitive robot programming by non-experts, leveraging natural language prompts and contextual information from the Robot Operating System (ROS). Our system integrates large language models (LLMs), enabling non-experts to articulate task requirements to the system through a chat interface. Key features of the framework include: integration of ROS with an AI agent connected to a plethora of open-source and commercial LLMs, automatic extraction of a behavior from the LLM output and execution of ROS actions/services, support for three behavior modes (sequence, behavior tree, state machine), imitation learning for adding new robot actions to the library of possible actions, and LLM reflection via human and environment feedback. Extensive experiments validate the framework, showcasing robustness, scalability, and versatility in diverse scenarios, including long-horizon tasks, tabletop rearrangements, and remote supervisory control. To facilitate the adoption of our framework and support the reproduction of our results, we have made our code open-source. You can access it at: https://github.com/huawei-noah/HEBO/tree/master/ROSLLM.

Nicolaj Schmid, Cornelius von Einem, Cesar Cadena et al.

Autonomous mobile robots are an increasingly integral part of modern factory and warehouse operations. Obstacle detection, avoidance and path planning are critical safety-relevant tasks, which are often solved using expensive LiDAR sensors and depth cameras. We propose to use cost-effective low-resolution ranging sensors, such as ultrasonic and infrared time-of-flight sensors by developing VIRUS-NeRF - Vision, InfraRed, and UltraSonic based Neural Radiance Fields. Building upon Instant Neural Graphics Primitives with a Multiresolution Hash Encoding (Instant-NGP), VIRUS-NeRF incorporates depth measurements from ultrasonic and infrared sensors and utilizes them to update the occupancy grid used for ray marching. Experimental evaluation in 2D demonstrates that VIRUS-NeRF achieves comparable mapping performance to LiDAR point clouds regarding coverage. Notably, in small environments, its accuracy aligns with that of LiDAR measurements, while in larger ones, it is bounded by the utilized ultrasonic sensors. An in-depth ablation study reveals that adding ultrasonic and infrared sensors is highly effective when dealing with sparse data and low view variation. Further, the proposed occupancy grid of VIRUS-NeRF improves the mapping capabilities and increases the training speed by 46% compared to Instant-NGP. Overall, VIRUS-NeRF presents a promising approach for cost-effective local mapping in mobile robotics, with potential applications in safety and navigation tasks. The code can be found at https://github.com/ethz-asl/virus nerf.

Lukas Schmid, Victor Reijgwart, Lionel Ott et al.

Exploration is a fundamental problem in robot autonomy. A major limitation, however, is that during exploration robots oftentimes have to rely on on-board systems alone for state estimation, accumulating significant drift over time in large environments. Drift can be detrimental to robot safety and exploration performance. In this work, a submap-based, multi-layer approach for both mapping and planning is proposed to enable safe and efficient volumetric exploration of large scale environments despite odometry drift. The central idea of our approach combines local (temporally and spatially) and global mapping to guarantee safety and efficiency. Similarly, our planning approach leverages the presented map to compute global volumetric frontiers in a changing global map and utilizes the nature of exploration dealing with partial information for efficient local and global planning. The presented system is thoroughly evaluated and shown to outperform state of the art methods even under drift-free conditions. Our system, termed GLoca}, will be made available open source.

Berta Bescos, Cesar Cadena, Jose Neira

In this paper we present a data-driven approach to obtain the static image of a scene, eliminating dynamic objects that might have been present at the time of traversing the scene with a camera. The general objective is to improve vision-based localization and mapping tasks in dynamic environments, where the presence (or absence) of different dynamic objects in different moments makes these tasks less robust. We introduce an end-to-end deep learning framework to turn images of an urban environment that include dynamic content, such as vehicles or pedestrians, into realistic static frames suitable for localization and mapping. This objective faces two main challenges: detecting the dynamic objects, and inpainting the static occluded back-ground. The first challenge is addressed by the use of a convolutional network that learns a multi-class semantic segmentation of the image. The second challenge is approached with a generative adversarial model that, taking as input the original dynamic image and the computed dynamic/static binary mask, is capable of generating the final static image. This framework makes use of two new losses, one based on image steganalysis techniques, useful to improve the inpainting quality, and another one based on ORB features, designed to enhance feature matching between real and hallucinated image regions. To validate our approach, we perform an extensive evaluation on different tasks that are affected by dynamic entities, i.e., visual odometry, place recognition and multi-view stereo, with the hallucinated images. Code has been made available on https://github.com/bertabescos/EmptyCities_SLAM.

Andrei Cramariuc, Aleksandar Petrov, Rohit Suri et al.

Self-diagnosis and self-repair are some of the key challenges in deploying robotic platforms for long-term real-world applications. One of the issues that can occur to a robot is miscalibration of its sensors due to aging, environmental transients, or external disturbances. Precise calibration lies at the core of a variety of applications, due to the need to accurately perceive the world. However, while a lot of work has focused on calibrating the sensors, not much has been done towards identifying when a sensor needs to be recalibrated. This paper focuses on a data-driven approach to learn the detection of miscalibration in vision sensors, specifically RGB cameras. Our contributions include a proposed miscalibration metric for RGB cameras and a novel semi-synthetic dataset generation pipeline based on this metric. Additionally, by training a deep convolutional neural network, we demonstrate the effectiveness of our pipeline to identify whether a recalibration of the camera's intrinsic parameters is required or not. The code is available at http://github.com/ethz-asl/camera_miscalib_detection.

Yu Liu, Jie Li, Xia Yuan et al.

Semantic Scene Completion (SSC) refers to the task of inferring the 3D semantic segmentation of a scene while simultaneously completing the 3D shapes. We propose PALNet, a novel hybrid network for SSC based on single depth. PALNet utilizes a two-stream network to extract both 2D and 3D features from multi-stages using fine-grained depth information to efficiently captures the context, as well as the geometric cues of the scene. Current methods for SSC treat all parts of the scene equally causing unnecessary attention to the interior of objects. To address this problem, we propose Position Aware Loss(PA-Loss) which is position importance aware while training the network. Specifically, PA-Loss considers Local Geometric Anisotropy to determine the importance of different positions within the scene. It is beneficial for recovering key details like the boundaries of objects and the corners of the scene. Comprehensive experiments on two benchmark datasets demonstrate the effectiveness of the proposed method and its superior performance. Models and Video demo can be found at: https://github.com/UniLauX/PALNet.

Renaud Dubé, Andrei Cramariuc, Daniel Dugas et al.

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 end-users 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 will be made available open-source along with easy to run demonstrations at www.github.com/ethz-asl/segmap. A video demonstration is available at https://youtu.be/CMk4w4eRobg.

Matías Mattamala, Jonas Frey, Piotr Libera et al.

Natural environments such as forests and grasslands are challenging for robotic navigation because of the false perception of rigid obstacles from high grass, twigs, or bushes. In this work, we present Wild Visual Navigation (WVN), an online self-supervised learning system for visual traversability estimation. The system is able to continuously adapt from a short human demonstration in the field, only using onboard sensing and computing. One of the key ideas to achieve this is the use of high-dimensional features from pre-trained self-supervised models, which implicitly encode semantic information that massively simplifies the learning task. Further, the development of an online scheme for supervision generator enables concurrent training and inference of the learned model in the wild. We demonstrate our approach through diverse real-world deployments in forests, parks, and grasslands. Our system is able to bootstrap the traversable terrain segmentation in less than 5 min of in-field training time, enabling the robot to navigate in complex, previously unseen outdoor terrains. Code: https://bit.ly/498b0CV - Project page:https://bit.ly/3M6nMHH

Kaixian Qu, Jie Tan, Tingnan Zhang et al.

Navigating efficiently to an object in an unexplored environment is a critical skill for general-purpose intelligent robots. Recent approaches to this object goal navigation problem have embraced a modular strategy, integrating classical exploration algorithms-notably frontier exploration-with a learned semantic mapping/exploration module. This paper introduces a novel informative path planning and 3D object probability mapping approach. The mapping module computes the probability of the object of interest through semantic segmentation and a Bayes filter. Additionally, it stores probabilities for common objects, which semantically guides the exploration based on common sense priors from a large language model. The planner terminates when the current viewpoint captures enough voxels identified with high confidence as the object of interest. Although our planner follows a zero-shot approach, it achieves state-of-the-art performance as measured by the Success weighted by Path Length (SPL) and Soft SPL in the Habitat ObjectNav Challenge 2023, outperforming other works by more than 20%. Furthermore, we validate its effectiveness on real robots. Project webpage: https://ippon-paper.github.io/

Manthan Patel, Fan Yang, Yuheng Qiu et al.

We present TartanGround, a large-scale, multi-modal dataset to advance the perception and autonomy of ground robots operating in diverse environments. This dataset, collected in various photorealistic simulation environments includes multiple RGB stereo cameras for 360-degree coverage, along with depth, optical flow, stereo disparity, LiDAR point clouds, ground truth poses, semantic segmented images, and occupancy maps with semantic labels. Data is collected using an integrated automatic pipeline, which generates trajectories mimicking the motion patterns of various ground robot platforms, including wheeled and legged robots. We collect 910 trajectories across 70 environments, resulting in 1.5 million samples. Evaluations on occupancy prediction and SLAM tasks reveal that state-of-the-art methods trained on existing datasets struggle to generalize across diverse scenes. TartanGround can serve as a testbed for training and evaluation of a broad range of learning-based tasks, including occupancy prediction, SLAM, neural scene representation, perception-based navigation, and more, enabling advancements in robotic perception and autonomy towards achieving robust models generalizable to more diverse scenarios. The dataset and codebase are available on the webpage: https://tartanair.org/tartanground

Rajitha de Silva, Jonathan Cox, Marija Popovic et al.

Robust robot navigation in outdoor environments requires accurate perception systems capable of handling visual challenges such as repetitive structures and changing appearances. Visual feature matching is crucial to vision-based pipelines but remains particularly challenging in natural outdoor settings due to perceptual aliasing. We address this issue in vineyards, where repetitive vine trunks and other natural elements generate ambiguous descriptors that hinder reliable feature matching. We hypothesise that semantic information tied to keypoint positions can alleviate perceptual aliasing by enhancing keypoint descriptor distinctiveness. To this end, we introduce a keypoint semantic integration technique that improves the descriptors in semantically meaningful regions within the image, enabling more accurate differentiation even among visually similar local features. We validate this approach in two vineyard perception tasks: (i) relative pose estimation and (ii) visual localisation. Across all tested keypoint types and descriptors, our method improves matching accuracy by 12.6%, demonstrating its effectiveness over multiple months in challenging vineyard conditions.

Yanqing Shen, Turcan Tuna, Marco Hutter et al.

Place recognition is essential to maintain global consistency in large-scale localization systems. While research in urban environments has progressed significantly using LiDARs or cameras, applications in natural forest-like environments remain largely under-explored. Furthermore, forests present particular challenges due to high self-similarity and substantial variations in vegetation growth over time. In this work, we propose a robust LiDAR-based place recognition method for natural forests, ForestLPR. We hypothesize that a set of cross-sectional images of the forest's geometry at different heights contains the information needed to recognize revisiting a place. The cross-sectional images are represented by \ac{bev} density images of horizontal slices of the point cloud at different heights. Our approach utilizes a visual transformer as the shared backbone to produce sets of local descriptors and introduces a multi-BEV interaction module to attend to information at different heights adaptively. It is followed by an aggregation layer that produces a rotation-invariant place descriptor. We evaluated the efficacy of our method extensively on real-world data from public benchmarks as well as robotic datasets and compared it against the state-of-the-art (SOTA) methods. The results indicate that ForestLPR has consistently good performance on all evaluations and achieves an average increase of 7.38\% and 9.11\% on Recall@1 over the closest competitor on intra-sequence loop closure detection and inter-sequence re-localization, respectively, validating our hypothesis

Rajitha de Silva, Jonathan Cox, James R. Heselden et al.

Accurate localisation is critical for mobile robots in structured outdoor environments, yet LiDAR-based methods often fail in vineyards due to repetitive row geometry and perceptual aliasing. We propose a semantic particle filter that incorporates stable object-level detections, specifically vine trunks and support poles into the likelihood estimation process. Detected landmarks are projected into a birds eye view and fused with LiDAR scans to generate semantic observations. A key innovation is the use of semantic walls, which connect adjacent landmarks into pseudo-structural constraints that mitigate row aliasing. To maintain global consistency in headland regions where semantics are sparse, we introduce a noisy GPS prior that adaptively supports the filter. Experiments in a real vineyard demonstrate that our approach maintains localisation within the correct row, recovers from deviations where AMCL fails, and outperforms vision-based SLAM methods such as RTAB-Map.

Richard Kuhlmann, Jakob Wolfram, Boyang Sun et al.

Autonomous inspection is a central problem in robotics, with applications ranging from industrial monitoring to search-and-rescue. Traditionally, inspection has often been reduced to navigation tasks, where the objective is to reach a predefined location while avoiding obstacles. However, this formulation captures only part of the real inspection problem. In real-world environments, the inspection targets may become visible well before their exact coordinates are reached, making further movement both redundant and inefficient. What matters more for inspection is not simply arriving at the target's position, but positioning the robot at a viewpoint from which the target becomes observable. In this work, we revisit inspection from a perception-aware perspective. We propose an end-to-end reinforcement learning framework that explicitly incorporates target visibility as the primary objective, enabling the robot to find the shortest trajectory that guarantees visual contact with the target without relying on a map. The learned policy leverages both perceptual and proprioceptive sensing and is trained entirely in simulation, before being deployed to a real-world robot. We further develop an algorithm to compute ground-truth shortest inspection paths, which provides a reference for evaluation. Through extensive experiments, we show that our method outperforms existing classical and learning-based navigation approaches, yielding more efficient inspection trajectories in both simulated and real-world settings. The project is avialable at https://sight-over-site.github.io/

Udayanga G. W. K. N. Gamage, Xuanni Huo, Luca Zanatta et al.

Small Unmanned Aerial Vehicle (UAV) based visual inspections are a more efficient alternative to manual methods for examining civil structural defects, offering safe access to hazardous areas and significant cost savings by reducing labor requirements. However, traditional frame-based cameras, widely used in UAV-based inspections, often struggle to capture defects under low or dynamic lighting conditions. In contrast, Dynamic Vision Sensors (DVS), or event-based cameras, excel in such scenarios by minimizing motion blur, enhancing power efficiency, and maintaining high-quality imaging across diverse lighting conditions without saturation or information loss. Despite these advantages, existing research lacks studies exploring the feasibility of using DVS for detecting civil structural defects.Moreover, there is no dedicated event-based dataset tailored for this purpose. Addressing this gap, this study introduces the first event-based civil infrastructure defect detection dataset, capturing defective surfaces as a spatio-temporal event stream using DVS.In addition to event-based data, the dataset includes grayscale intensity image frames captured simultaneously using an Active Pixel Sensor (APS). Both data types were collected using the DAVIS346 camera, which integrates DVS and APS sensors.The dataset focuses on two types of defects: cracks and spalling, and includes data from both field and laboratory environments. The field dataset comprises 318 recording sequences,documenting 458 distinct cracks and 121 distinct spalling instances.The laboratory dataset includes 362 recording sequences, covering 220 distinct cracks and 308 spalling instances.Four realtime object detection models were evaluated on it to validate the dataset effectiveness.The results demonstrate the dataset robustness in enabling accurate defect detection and classification,even under challenging lighting conditions.

Jonas Frey, Matias Mattamala, Nived Chebrolu et al.

Natural environments such as forests and grasslands are challenging for robotic navigation because of the false perception of rigid obstacles from high grass, twigs, or bushes. In this work, we propose Wild Visual Navigation (WVN), an online self-supervised learning system for traversability estimation which uses only vision. The system is able to continuously adapt from a short human demonstration in the field. It leverages high-dimensional features from self-supervised visual transformer models, with an online scheme for supervision generation that runs in real-time on the robot. We demonstrate the advantages of our approach with experiments and ablation studies in challenging environments in forests, parks, and grasslands. Our system is able to bootstrap the traversable terrain segmentation in less than 5 min of in-field training time, enabling the robot to navigate in complex outdoor terrains - negotiating obstacles in high grass as well as a 1.4 km footpath following. While our experiments were executed with a quadruped robot, ANYmal, the approach presented can generalize to any ground robot.

Jonas Frey, Hermann Blum, Francesco Milano et al.

Semantic segmentation networks are usually pre-trained once and not updated during deployment. As a consequence, misclassifications commonly occur if the distribution of the training data deviates from the one encountered during the robot's operation. We propose to mitigate this problem by adapting the neural network to the robot's environment during deployment, without any need for external supervision. Leveraging complementary data representations, we generate a supervision signal, by probabilistically accumulating consecutive 2D semantic predictions in a volumetric 3D map. We then train the network on renderings of the accumulated semantic map, effectively resolving ambiguities and enforcing multi-view consistency through the 3D representation. In contrast to scene adaptation methods, we aim to retain the previously-learned knowledge, and therefore employ a continual learning experience replay strategy to adapt the network. Through extensive experimental evaluation, we show successful adaptation to real-world indoor scenes both on the ScanNet dataset and on in-house data recorded with an RGB-D sensor. Our method increases the segmentation accuracy on average by 9.9% compared to the fixed pre-trained neural network, while retaining knowledge from the pre-training dataset.

Stefan Lionar, Lukas Schmid, Cesar Cadena et al.

We present a novel 3D mapping method leveraging the recent progress in neural implicit representation for 3D reconstruction. Most existing state-of-the-art neural implicit representation methods are limited to object-level reconstructions and can not incrementally perform updates given new data. In this work, we propose a fusion strategy and training pipeline to incrementally build and update neural implicit representations that enable the reconstruction of large scenes from sequential partial observations. By representing an arbitrarily sized scene as a grid of latent codes and performing updates directly in latent space, we show that incrementally built occupancy maps can be obtained in real-time even on a CPU. Compared to traditional approaches such as Truncated Signed Distance Fields (TSDFs), our map representation is significantly more robust in yielding a better scene completeness given noisy inputs. We demonstrate the performance of our approach in thorough experimental validation on real-world datasets with varying degrees of added pose noise.

Boyang Sun, Jiaxu Xing, Hermann Blum et al.

Autonomous robots deal with unexpected scenarios in real environments. Given input images, various visual perception tasks can be performed, e.g., semantic segmentation, depth estimation and normal estimation. These different tasks provide rich information for the whole robotic perception system. All tasks have their own characteristics while sharing some latent correlations. However, some of the task predictions may suffer from the unreliability dealing with complex scenes and anomalies. We propose an attention-based failure detection approach by exploiting the correlations among multiple tasks. The proposed framework infers task failures by evaluating the individual prediction, across multiple visual perception tasks for different regions in an image. The formulation of the evaluations is based on an attention network supervised by multi-task uncertainty estimation and their corresponding prediction errors. Our proposed framework generates more accurate estimations of the prediction error for the different task's predictions.

Lukas Schmid, Jeffrey Delmerico, Johannes Schönberger et al.

For robotic interaction in environments shared with other agents, access to volumetric and semantic maps of the scene is crucial. However, such environments are inevitably subject to long-term changes, which the map needs to account for. We thus propose panoptic multi-TSDFs as a novel representation for multi-resolution volumetric mapping in changing environments. By leveraging high-level information for 3D reconstruction, our proposed system allocates high resolution only where needed. Through reasoning on the object level, semantic consistency over time is achieved. This enables our method to maintain up-to-date reconstructions with high accuracy while improving coverage by incorporating previous data. We show in thorough experimental evaluation that our map can be efficiently constructed, maintained, and queried during online operation, and that the presented approach can operate robustly on real depth sensors using non-optimized panoptic segmentation as input.

Florian Tschopp, Juan Nieto, Roland Siegwart et al.

Introducing semantically meaningful objects to visual Simultaneous Localization And Mapping (SLAM) has the potential to improve both the accuracy and reliability of pose estimates, especially in challenging scenarios with significant view-point and appearance changes. However, how semantic objects should be represented for an efficient inclusion in optimization-based SLAM frameworks is still an open question. Superquadrics(SQs) are an efficient and compact object representation, able to represent most common object types to a high degree, and typically retrieved from 3D point-cloud data. However, accurate 3D point-cloud data might not be available in all applications. Recent advancements in machine learning enabled robust object recognition and semantic mask measurements from camera images under many different appearance conditions. We propose a pipeline to leverage such semantic mask measurements to fit SQ parameters to multi-view camera observations using a multi-stage initialization and optimization procedure. We demonstrate the system's ability to retrieve randomly generated SQ parameters from multi-view mask observations in preliminary simulation experiments and evaluate different initialization stages and cost functions.

Hermann Blum, Francesco Milano, René Zurbrügg et al.

We propose a novel robotic system that can improve its perception during deployment. Contrary to the established approach of learning semantics from large datasets and deploying fixed models, we propose a framework in which semantic models are continuously updated on the robot to adapt to the deployment environments. By combining continual learning with self-supervision, our robotic system learns online during deployment without external supervision. We conduct real-world experiments with robots localising in 3D floorplans. Our experiments show how the robot's semantic perception improves during deployment and how this translates into improved localisation, even across drastically different environments. We further study the risk of catastrophic forgetting that such a continuous learning setting poses. We find memory replay an effective measure to reduce forgetting and show how the robotic system can improve even when switching between different environments. On average, our system improves by 60% in segmentation and 10% in localisation accuracy compared to deployment of a fixed model, and it maintains this improvement while adapting to further environments.

Lukas Bernreiter, Lionel Ott, Juan Nieto et al.

In this paper, we propose a robust end-to-end multi-modal pipeline for place recognition where the sensor systems can differ from the map building to the query. Our approach operates directly on images and LiDAR scans without requiring any local feature extraction modules. By projecting the sensor data onto the unit sphere, we learn a multi-modal descriptor of partially overlapping scenes using a spherical convolutional neural network. The employed spherical projection model enables the support of arbitrary LiDAR and camera systems readily without losing information. Loop closure candidates are found using a nearest-neighbor lookup in the embedding space. We tackle the problem of correctly identifying the closest place by correlating the candidates' power spectra, obtaining a confidence value per prospect. Our estimate for the correct place corresponds then to the candidate with the highest confidence. We evaluate our proposal w.r.t. state-of-the-art approaches in place recognition using real-world data acquired using different sensors. Our approach can achieve a recall that is up to 10% and 5% higher than for a LiDAR- and vision-based system, respectively, when the sensor setup differs between model training and deployment. Additionally, our place selection can correctly identify up to 95% matches from the candidate set.

Jiapeng Zhong, Zheyu Ye, Andrei Cramariuc et al.

Many mobile robotic platforms rely on an accurate knowledge of the extrinsic calibration parameters, especially systems performing visual stereo matching. Although a number of accurate stereo camera calibration methods have been developed, which provide good initial "factory" calibrations, the determined parameters can lose their validity over time as the sensors are exposed to environmental conditions and external effects. Thus, on autonomous platforms on-board diagnostic methods for an early detection of the need to repeat calibration procedures have the potential to prevent critical failures of crucial systems, such as state estimation or obstacle detection. In this work, we present a novel data-driven method to estimate the calibration quality and detect discrepancies between the original calibration and the current system state for stereo camera systems. The framework consists of a novel dataset generation pipeline to train CalQNet, a deep convolutional neural network. CalQNet can estimate the calibration quality using a new metric that approximates the degree of miscalibration in stereo setups. We show the framework's ability to predict from a single stereo frame if a state-of-the-art stereo-visual odometry system will diverge due to a degraded calibration in two real-world experiments.

Giancarlo Di Biase, Hermann Blum, Roland Siegwart et al.

The inability of state-of-the-art semantic segmentation methods to detect anomaly instances hinders them from being deployed in safety-critical and complex applications, such as autonomous driving. Recent approaches have focused on either leveraging segmentation uncertainty to identify anomalous areas or re-synthesizing the image from the semantic label map to find dissimilarities with the input image. In this work, we demonstrate that these two methodologies contain complementary information and can be combined to produce robust predictions for anomaly segmentation. We present a pixel-wise anomaly detection framework that uses uncertainty maps to improve over existing re-synthesis methods in finding dissimilarities between the input and generated images. Our approach works as a general framework around already trained segmentation networks, which ensures anomaly detection without compromising segmentation accuracy, while significantly outperforming all similar methods. Top-2 performance across a range of different anomaly datasets shows the robustness of our approach to handling different anomaly instances.

Michael Pantic, Lionel Ott, Cesar Cadena et al.

This paper presents a novel on-line path planning method that enables aerial robots to interact with surfaces. We present a solution to the problem of finding trajectories that drive a robot towards a surface and move along it. Triangular meshes are used as a surface map representation that is free of fixed discretization and allows for very large workspaces. We propose to leverage planar parametrization methods to obtain a lower-dimensional topologically equivalent representation of the original surface. Furthermore, we interpret the original surface and its lower-dimensional representation as manifold approximations that allow the use of Riemannian Motion Policies (RMPs), resulting in an efficient, versatile, and elegant motion generation framework. We compare against several Rapidly-exploring Random Tree (RRT) planners, a customized CHOMP variant, and the discrete geodesic algorithm. Using extensive simulations on real-world data we show that the proposed planner can reliably plan high-quality near-optimal trajectories at minimal computational cost. The accompanying multimedia attachment demonstrates feasibility on a real OMAV. The obtained paths show less than 10% deviation from the theoretical optimum while facilitating reactive re-planning at kHz refresh rates, enabling flying robots to perform motion planning for interaction with complex surfaces.

Lukas Bernreiter, Lionel Ott, Juan Nieto et al.

We propose PHASER, a correspondence-free global registration of sensor-centric pointclouds that is robust to noise, sparsity, and partial overlaps. Our method can seamlessly handle multimodal information and does not rely on keypoint nor descriptor preprocessing modules. By exploiting properties of Fourier analysis, PHASER operates directly on the sensor's signal, fusing the spectra of multiple channels and computing the 6-DoF transformation based on correlation. Our registration pipeline starts by finding the most likely rotation followed by computing the most likely translation. Both estimates are distributed according to a probability distribution that takes the underlying manifold into account, i.e., a Bingham and Gaussian distribution, respectively. This further allows our approach to consider the periodic-nature of rotations and naturally represent its uncertainty. We extensively compare PHASER against several well-known registration algorithms on both simulated datasets, and real-world data acquired using different sensor configurations. Our results show that PHASER can globally align pointclouds in less than 100ms with an average accuracy of 2cm and 0.5deg, is resilient against noise, and can handle partial overlap.

Janis Postels, Hermann Blum, Yannick Strümpler et al.

The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution moreover correlates with a model's epistemic uncertainty, thus indicates its ability to generalise to novel inputs. We first empirically verify that epistemic uncertainty can be identified with the surprise, thus the negative log-likelihood, of observing a particular latent representation. Moreover, we demonstrate that the output-conditional distribution of hidden representations also allows quantifying aleatoric uncertainty via the entropy of the predictive distribution. We analyse epistemic and aleatoric uncertainty inferred from the representations of different layers and conclude that deeper layers lead to uncertainty with similar behaviour as established - but computationally more expensive - methods (e.g. deep ensembles). While our approach does not require modifying the training process, we follow prior work and experiment with an additional regularising loss that increases the information in the latent representations. We find that this leads to improved OOD detection of epistemic uncertainty at the cost of ambiguous calibration close to the data distribution. We verify our findings on both classification and regression models.

Le Chen, Yunke Ao, Florian Tschopp et al.

Visual-inertial systems rely on precise calibrations of both camera intrinsics and inter-sensor extrinsics, which typically require manually performing complex motions in front of a calibration target. In this work we present a novel approach to obtain favorable trajectories for visual-inertial system calibration, using model-based deep reinforcement learning. Our key contribution is to model the calibration process as a Markov decision process and then use model-based deep reinforcement learning with particle swarm optimization to establish a sequence of calibration trajectories to be performed by a robot arm. Our experiments show that while maintaining similar or shorter path lengths, the trajectories generated by our learned policy result in lower calibration errors compared to random or handcrafted trajectories.

Julia Nitsch, Masha Itkina, Ransalu Senanayake et al.

Neural networks (NNs) are widely used for object classification in autonomous driving. However, NNs can fail on input data not well represented by the training dataset, known as out-of-distribution (OOD) data. A mechanism to detect OOD samples is important for safety-critical applications, such as automotive perception, to trigger a safe fallback mode. NNs often rely on softmax normalization for confidence estimation, which can lead to high confidences being assigned to OOD samples, thus hindering the detection of failures. This paper presents a method for determining whether inputs are OOD, which does not require OOD data during training and does not increase the computational cost of inference. The latter property is especially important in automotive applications with limited computational resources and real-time constraints. Our proposed approach outperforms state-of-the-art methods on real-world automotive datasets.

Alexander Millane, Helen Oleynikova, Christian Lanegger et al.

Localization of a robotic system within a previously mapped environment is important for reducing estimation drift and for reusing previously built maps. Existing techniques for geometry-based localization have focused on the description of local surface geometry, usually using pointclouds as the underlying representation. We propose a system for geometry-based localization that extracts features directly from an implicit surface representation: the Signed Distance Function (SDF). The SDF varies continuously through space, which allows the proposed system to extract and utilize features describing both surfaces and free-space. Through evaluations on public datasets, we demonstrate the flexibility of this approach, and show an increase in localization performance over state-of-the-art handcrafted surfaces-only descriptors. We achieve an average improvement of ~12% on an RGB-D dataset and ~18% on a LiDAR-based dataset. Finally, we demonstrate our system for localizing a LiDAR-equipped MAV within a previously built map of a search and rescue training ground.

Timo Hinzmann, Tobias Stegemann, Cesar Cadena et al.

In this paper, we present our deep learning-based human detection system that uses optical (RGB) and long-wave infrared (LWIR) cameras to detect, track, localize, and re-identify humans from UAVs flying at high altitude. In each spectrum, a customized RetinaNet network with ResNet backbone provides human detections which are subsequently fused to minimize the overall false detection rate. We show that by optimizing the bounding box anchors and augmenting the image resolution the number of missed detections from high altitudes can be decreased by over 20 percent. Our proposed network is compared to different RetinaNet and YOLO variants, and to a classical optical-infrared human detection framework that uses hand-crafted features. Furthermore, along with the publication of this paper, we release a collection of annotated optical-infrared datasets recorded with different UAVs during search-and-rescue field tests and the source code of the implemented annotation tool.

Hermann Blum, Julian Stiefel, Cesar Cadena et al.

This paper presents a localization system for mobile robots enabling precise localization in inaccurate building models. The approach leverages local referencing to counteract inherent deviations between as-planned and as-built data for locally accurate registration. We further fuse a novel image-based robust outlier detector with LiDAR data to reject a wide range of outlier measurements from clutter, dynamic objects, and sensor failures. We evaluate the proposed approach on a mobile robot in a challenging real world building construction site. It consistently outperforms the traditional ICP-based alingment, reducing localization error by at least 30%.

Lukas Schaupp, Patrick Pfreundschuh, Mathias Buerki et al.

Visually poor scenarios are one of the main sources of failure in visual localization systems in outdoor environments. To address this challenge, we present MOZARD, a multi-modal localization system for urban outdoor environments using vision and LiDAR. By extending our preexisting key-point based visual multi-session local localization approach with the use of semantic data, an improved localization recall can be achieved across vastly different appearance conditions. In particular we focus on the use of curbstone information because of their broad distribution and reliability within urban environments. We present thorough experimental evaluations on several driving kilometers in challenging urban outdoor environments, analyze the recall and accuracy of our localization system and demonstrate in a case study possible failure cases of each subsystem. We demonstrate that MOZARD is able to bridge scenarios where our previous work VIZARD fails, hence yielding an increased recall performance, while a similar localization accuracy of 0.2m is achieved

Yu Liu, Jie Li, Qingsen Yan et al.

This paper tackles the problem of data fusion in the semantic scene completion (SSC) task, which can simultaneously deal with semantic labeling and scene completion. RGB images contain texture details of the object(s) which are vital for semantic scene understanding. Meanwhile, depth images capture geometric clues of high relevance for shape completion. Using both RGB and depth images can further boost the accuracy of SSC over employing one modality in isolation. We propose a 3D gated recurrent fusion network (GRFNet), which learns to adaptively select and fuse the relevant information from depth and RGB by making use of the gate and memory modules. Based on the single-stage fusion, we further propose a multi-stage fusion strategy, which could model the correlations among different stages within the network. Extensive experiments on two benchmark datasets demonstrate the superior performance and the effectiveness of the proposed GRFNet for data fusion in SSC. Code will be made available.

Marius Fehr, Tim Taubner, Yang Liu et al.

Safe and efficient path planning is crucial for autonomous mobile robots. A prerequisite for path planning is to have a comprehensive understanding of the 3D structure of the robot's environment. On MAVs this is commonly achieved using low-cost sensors, such as stereo or RGB-D cameras. These sensors may fail to provide depth measurements in textureless or IR-absorbing areas and have limited effective range. In path planning, this results in inefficient trajectories or failure to recognize a feasible path to the goal, hence significantly impairing the robot's mobility. Recent advances in deep learning enables us to exploit prior experience about the shape of the world and hence to infer complete depth maps from color images and additional sparse depth measurements. In this work, we present an augmented planning system and investigate the effects of employing state-of-the-art depth completion techniques, specifically trained to augment sparse depth maps originating from RGB-D sensors, semi-dense methods and stereo matchers. We extensively evaluate our approach in online path planning experiments based on simulated data, as well as global path planning experiments based on real world MAV data. We show that our augmented system, provided with only sparse depth perception, can reach on-par performance to ground truth depth input in simulated online planning experiments. On real world MAV data the augmented system demonstrates superior performance compared to a planner based on very dense RGB-D depth maps.

David Haldimann, Hermann Blum, Roland Siegwart et al.

There has been a remarkable progress in the accuracy of semantic segmentation due to the capabilities of deep learning. Unfortunately, these methods are not able to generalize much further than the distribution of their training data and fail to handle out-of-distribution classes appropriately. This limits the applicability to autonomous or safety critical systems. We propose a novel method leveraging generative models to detect wrongly segmented or out-of-distribution instances. Conditioned on the predicted semantic segmentation, an RGB image is generated. We then learn a dissimilarity metric that compares the generated image with the original input and detects inconsistencies introduced by the semantic segmentation. We present test cases for outlier and misclassification detection and evaluate our method qualitatively and quantitatively on multiple datasets.

Timo Hinzmann, Cesar Cadena, Juan Nieto et al.

In this paper, we propose a visual-inertial framework able to efficiently estimate the camera poses of a non-rigid trinocular baseline for long-range depth estimation on-board a fast moving aerial platform. The estimation of the time-varying baseline is based on relative inertial measurements, a photometric relative pose optimizer, and a probabilistic wing model fused in an efficient Extended Kalman Filter (EKF) formulation. The estimated depth measurements can be integrated into a geo-referenced global map to render a reconstruction of the environment useful for local replanning algorithms. Based on extensive real-world experiments we describe the challenges and solutions for obtaining the probabilistic wing model, reliable relative inertial measurements, and vision-based relative pose updates and demonstrate the computational efficiency and robustness of the overall system under challenging conditions.