https://danieledema.github.io/tags/workshop/Recent content in Workshop on Daniele De MartiniHugoen-usFri, 05 Jun 2026 00:00:00 +0000https://danieledema.github.io/publications/20260605-distance_travelled_error/Fri, 05 Jun 2026 00:00:00 +0000https://danieledema.github.io/publications/20260605-distance_travelled_error/Abstract The linked workshop poster has no separate author-provided abstract. The summary below is AI-generated from the poster. Summary This workshop contribution introduces Distance Travelled Error as a trajectory-evaluation metric. It considers how to compare estimated motion using distance travelled as a reference for error assessment. Useful links Paper: https://norlab.ulaval.ca/icra_workshop_field_robotics/posters/2026/45.pdf Workshop programme: https://norlab-ulaval.github.io/workshop_field_robotics_icra2026/ Bibtex @inproceedings{distancetravellederror2026, author={Daum, Effie and De Martini, Daniele and Pomerleau, François}, title={A Novel Trajectory Evaluation Metric: Distance Travelled Error (DTE)}, year={2026}, booktitle={ICRA Workshop on Field Robotics}, url={https://norlab.https://danieledema.github.io/publications/20260605-quadruped_terrain_stability/Fri, 05 Jun 2026 00:00:00 +0000https://danieledema.github.io/publications/20260605-quadruped_terrain_stability/Abstract The linked workshop poster has no separate author-provided abstract. The summary below is AI-generated from the poster. Summary This workshop contribution investigates how a quadruped robot’s proprioceptive measurements can be used to infer terrain stability. It examines information available from the robot’s own motion and contact with the ground. Useful links Paper: https://norlab.ulaval.ca/icra_workshop_field_robotics/posters/2026/44.pdf Bibtex @inproceedings{quadrupedterrainstability2026, author={Fortin, Jean-Michel and De Martini, Daniele and Pomerleau, François and Giguère, Philippe}, title={Inferring Terrain Stability from Proprioceptive Data of a Quadruped Robot}, year={2026}, booktitle={ICRA Workshop on Field Robotics}, url={https://norlab.https://danieledema.github.io/publications/20260518-external_cameras_prior_maps/Mon, 18 May 2026 00:00:00 +0000https://danieledema.github.io/publications/20260518-external_cameras_prior_maps/Abstract Commonly available prior information, such as BIM models, floor plans, and remote sensing images, can provide valuable geometric and semantic context for autonomous robotic systems. In this paper, we treat observations from fixed external RGB cameras as Common Prior Maps (CPMs): wide-field views of the environment that initialize a semantic and geometric scene prior before any robot motion begins. We present an RGB-only framework for active, incremental 3D scene graph (3DSG) generation that seamlessly fuses observations from both onboard robot cameras and fixed external cameras within a single hardware-agnostic pipeline.https://danieledema.github.io/publications/20260518-rgb_active_scene_graphs/Mon, 18 May 2026 00:00:00 +0000https://danieledema.github.io/publications/20260518-rgb_active_scene_graphs/Abstract Current approaches to 3D scene graph generation rely on dedicated depth sensors, such as LiDAR or RGB-D cameras, for metric 3D reconstruction. This limits deployment to specialized robotic platforms and excludes settings where only RGB cameras are available, such as fixed external infrastructure. Existing pipelines also typically operate on passively collected observation trajectories, rather than selecting viewpoints based on the partially built scene representation, and therefore fail to effectively exploit the semantic and spatial information encoded within the graph during exploration.https://danieledema.github.io/publications/20251008-scene_graph_introspection/Wed, 08 Oct 2025 00:00:00 +0000https://danieledema.github.io/publications/20251008-scene_graph_introspection/Abstract This work investigates how semantics influence localisation performance and robustness in a learned self-supervised, contrastive semantic localisation framework. After training a localisation network on both original and perturbed maps, we conduct a thorough post-hoc introspection analysis to probe whether the model filters environmental noise and prioritises distinctive landmarks over routine clutter. We validate various interpretability methods and present a comparative reliability analysis. Integrated gradients and Attention Weights consistently emerge as the most reliable probes of learned behaviour.https://danieledema.github.io/publications/20240802-nerfoot/Fri, 02 Aug 2024 00:00:00 +0000https://danieledema.github.io/publications/20240802-nerfoot/Abstract This paper investigates the utility of Neural Radiance Fields (NeRF) models in extending the regions of operation of a mobile robot, controlled by Image-Based Visual Servoing (IBVS) via static CCTV cameras. Using NeRF as a 3D-representation prior, the robot’s footprint may be extrapolated geometrically and used to train a CNN-based network to extract it online from the robot’s appearance alone. The resulting footprint results in a tighter bound than a robot-wide bounding box, allowing the robot’s controller to prescribe more optimal trajectories and expand its safe operational floor area.https://danieledema.github.io/publications/20240520-tagic/Mon, 20 May 2024 00:00:00 +0000https://danieledema.github.io/publications/20240520-tagic/Abstract Image-based teleoperation offers significant flexibility and efficiency in several applications, such as teleoperated driving; still, it highly depends on reliable communication band-width and high Signal-to-Noise Ratio (SNR), which is hard to guarantee in uncontrolled environments. This poster tackles the challenge of reliable communication under limited bandwidth. We propose to leverage the context and task knowledge to guide the compression to favor task performance rather than image fidelity. In particular, we jointly designed source-channel coding with a task performer to present an end-to-end TAsk-Guided Image Communication (TAGIC) framework, which uses Soft Introspective Variational Autoencoder (S-IntroVAE) and prioritizes the task-critical image information with limited communication bandwidth in the low SNR region.https://danieledema.github.io/publications/20240416-watching_grass_grow/Tue, 16 Apr 2024 00:00:00 +0000https://danieledema.github.io/publications/20240416-watching_grass_grow/Abstract We describe a challenging robotics deployment in a complex ecosystem to monitor a rich plant community. The study site is dominated by dynamic grassland vegetation and is thus visually ambiguous and liable to drastic appearance change over the course of a day and especially through the growing season. This dynamism and complexity in appearance seriously impact the stability of the robotics platform, as localisation is a foundational part of that control loop, and so routes must be carefully taught and retaught until autonomy is robust and repeatable.https://danieledema.github.io/publications/20230101-semantic_explanations_workshop/Sun, 01 Jan 2023 00:00:00 +0000https://danieledema.github.io/publications/20230101-semantic_explanations_workshop/Abstract In this work, we investigate the use of semantic attention to explain the performance of a Graph Neural Network (GNN)-based pose estimation model. To validate our approach, we apply semantically-informed perturbations to the input data and correlate the predicted feature importance weights with the accuracy of the model. Graph Deep Learning (GDL) is an emerging field of machine learning for tasks like scene interpretation, as it exploits flexible graph structures to describe complex features and relationships in a very concise format.https://danieledema.github.io/publications/20221025-teleoperation_keyframe_poster/Tue, 25 Oct 2022 00:00:00 +0000https://danieledema.github.io/publications/20221025-teleoperation_keyframe_poster/Abstract The institutional repository explicitly lists no abstract. The linked item is a one-page poster without a separate abstract section; the summary below is AI-generated. Summary This poster proposes transmitting selected keyframes instead of every video frame in teleoperated driving. A predictor reconstructs the intervening observations to address bandwidth and delay, with preliminary evaluation in CARLA. Useful links Publication record: https://eprints.gla.ac.uk/277210/ Bibtex @inproceedings{teleoperationkeyframeposter2022, author={Diao, Y. and Dai, H. and Zhao, G.https://danieledema.github.io/publications/20221001-metaverse_codesign_poster/Sat, 01 Oct 2022 00:00:00 +0000https://danieledema.github.io/publications/20221001-metaverse_codesign_poster/Abstract The institutional repository explicitly lists no abstract, and the deposited poster is restricted. The summary below is AI-generated. Summary This late-breaking-results poster explores the joint design of sampling, communication and prediction for keeping physical and digital robot states synchronised. Useful links Publication record: https://eprints.gla.ac.uk/277211/ Related journal article: https://doi.org/10.1109/JSAC.2022.3221993 Bibtex @inproceedings{metaversecodesignposter2022, author={Meng, Zhen and She, Changyang and Zhao, Guodong and De Martini, Daniele and Zhang, Yichi}, title={Sampling, Communication, and Prediction Co-Design for Synchronizing the Real-Robot and Digital-Robot in Metaverse}, year={2022}, booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Late Breaking Results}, url={https://eprints.https://danieledema.github.io/publications/20211104-raven_workshop/Thu, 04 Nov 2021 00:00:00 +0000https://danieledema.github.io/publications/20211104-raven_workshop/Abstract In this paper, we introduce RaVAEn, a lightweight, unsupervised approach for change detection in satellite data based on Variational Auto-Encoders (VAEs) with the specific purpose of on-board deployment. Applications such as disaster management enormously benefit from the rapid availability of satellite observations. Traditionally, data analysis is performed on the ground after all data is transferred - downlinked - to a ground station. Constraint on the downlink capabilities therefore affects any downstream application.https://danieledema.github.io/publications/20210711-oxford_road_boundaries/Sun, 11 Jul 2021 00:00:00 +0000https://danieledema.github.io/publications/20210711-oxford_road_boundaries/Abstract In this paper we present the Oxford Road Boundaries Dataset, designed for training and testing machine-learning-based road-boundary detection and inference approaches. We have hand-annotated two of the 10 km-long forays from the Oxford Robotcar Dataset and generated from other forays several thousand further examples with semi-annotated road-boundary masks. To boost the number of training samples in this way, we used a vision-based localiser to project labels from the annotated datasets to other traversals at different times and weather conditions.https://danieledema.github.io/publications/20210612-radar_video_embeddings/Sat, 12 Jun 2021 00:00:00 +0000https://danieledema.github.io/publications/20210612-radar_video_embeddings/Abstract We learn, in an unsupervised way, an embedding from sequences of radar images that is suitable for solving place recognition problem using complex radar data. We experiment on 280 km of data and show performance exceeding state-of-the-art supervised approaches, localising correctly 98.38% of the time when using just the nearest database candidate. Useful links Preprint: https://arxiv.org/abs/2106.06703 Bibtex @inproceedings{radarvideoembeddings2021, author={Matthew Gadd and Daniele De Martini and Paul Newman}, title={Unsupervised Place Recognition with Deep Embedding Learning over Radar Videos}, year={2021}, booktitle={ICRA Workshop on Radar Perception for All-Weather Autonomy}, url={https://arxiv.https://danieledema.github.io/publications/20210101-smart_satellites_agu/Fri, 01 Jan 2021 00:00:00 +0000https://danieledema.github.io/publications/20210101-smart_satellites_agu/Abstract Satellite observations of the Earth’s surface provide a vital data source for diverse environmental applications including disaster management, detection of landcover change and ecological monitoring. Currently, sensors on satellites collect and downlink all data points for further processing on the ground. Limitations in downlink capacity and speed result in delays in data availability on the ground. This particularly affects applications such as disaster management which require data as quickly as possible for online event monitoring.https://danieledema.github.io/publications/20201019-sense_assess_explain/Mon, 19 Oct 2020 00:00:00 +0000https://danieledema.github.io/publications/20201019-sense_assess_explain/Abstract This paper discusses ongoing work in demonstrating research in mobile autonomy in challenging driving scenarios. In our approach, we address fundamental technical issues to overcome critical barriers to assurance and regulation for large-scale deployments of autonomous systems. To this end, we present how we build robots that (1) can robustly sense and interpret their environment using traditional as well as unconventional sensors; (2) can assess their own capabilities; and (3), vitally in the purpose of assurance and trust, can provide causal explanations of their interpretations and assessments.https://danieledema.github.io/publications/20200101-rainbench_workshops/Wed, 01 Jan 2020 00:00:00 +0000https://danieledema.github.io/publications/20200101-rainbench_workshops/Abstract Climate change is expected to aggravate extreme precipitation events, directly impacting the livelihood of millions. Without a global precipitation forecasting system in place, many regions – especially those constrained in resources to collect expensive groundstation data – are left behind. To mitigate such unequal reach of climate change, a solution is to alleviate the reliance on numerical models (and by extension groundstation data) by enabling machine-learning-based global forecasts from satellite imagery.https://danieledema.github.io/publications/20200101-physics_informed_precipitation/Wed, 01 Jan 2020 00:00:00 +0000https://danieledema.github.io/publications/20200101-physics_informed_precipitation/Abstract Under the effects of global warming, extreme events such as floods and droughts are increasing in frequency and intensity. This trend directly affects communities and make all the more urgent widening the access to accurate precipitation forecasting systems for disaster preparedness. Nowadays, weather forecasting relies on numerical models necessitating massive computing resources that most developing countries cannot afford. Machine learning approaches are still in their infancy but already show the promise for democratizing weather predictions, by leveraging any data source and requiring less compute.https://danieledema.github.io/publications/20190429-listening_closely/Mon, 29 Apr 2019 00:00:00 +0000https://danieledema.github.io/publications/20190429-listening_closely/Abstract The linked publication record lists a conference abstract, but does not provide its text. An author-provided abstract could not be located online. Summary This conference abstract explores radar-based terrain classification using information from auditory signals. Useful links Publication record: https://vbn.aau.dk/en/publications/listening-closely-to-see-far-away-radar-based-terrain-classificat-2/ Bibtex @inproceedings{listeningclosely2019, author={Williams, D. and De Martini, D. and Marchegiani, L. and Newman, P.}, title={Listening closely to see far away: Radar-based Terrain Classification from Auditory Signals}, year={2019}, booktitle={International Conference on Digital Image \& Signal Processing (DISP)}, url={https://vbn.