Projects / Programmes
Adaptive deep perception methods for autonomous surface vehicles
Code |
Science |
Field |
Subfield |
2.07.07 |
Engineering sciences and technologies |
Computer science and informatics |
Intelligent systems - software |
Code |
Science |
Field |
1.02 |
Natural Sciences |
Computer and information sciences |
Autonomous boats, computer vision, deep methods, panoptic perception, modality fusion
Data for the last 5 years (citations for the last 10 years) on
March 27, 2024;
A3 for period
2018-2022
Data for ARIS tenders (
04.04.2019 – Programme tender,
archive
)
Database |
Linked records |
Citations |
Pure citations |
Average pure citations |
WoS |
136 |
6,642 |
6,320 |
46.47 |
Scopus |
172 |
9,127 |
8,681 |
50.47 |
Researchers (10)
no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
1. |
50367 |
Borja Bovcon |
Computer science and informatics |
Researcher |
2020 - 2021 |
19 |
2. |
30155 |
PhD Matej Kristan |
Computer science and informatics |
Head |
2020 - 2024 |
323 |
3. |
39227 |
PhD Alan Lukežič |
Computer science and informatics |
Researcher |
2020 - 2024 |
51 |
4. |
25049 |
MSc Dean Mozetič |
Computer science and informatics |
Researcher |
2020 - 2024 |
15 |
5. |
50843 |
Jon Natanael Muhovič |
Computer science and informatics |
Researcher |
2020 - 2024 |
23 |
6. |
21310 |
PhD Janez Perš |
Systems and cybernetics |
Researcher |
2020 - 2024 |
238 |
7. |
18198 |
PhD Danijel Skočaj |
Computer science and informatics |
Researcher |
2020 - 2024 |
308 |
8. |
34398 |
PhD Domen Tabernik |
Computer science and informatics |
Researcher |
2020 - 2024 |
49 |
9. |
20683 |
Aljoša Žerjal |
Physics |
Researcher |
2020 - 2024 |
50 |
10. |
55002 |
Lojze Žust |
Computer science and informatics |
Researcher |
2020 - 2024 |
22 |
Organisations (3)
Abstract
Autonomous robotics is a fast growing research discipline opening new scientific as well as technical challenges. Most of the research is invested in self-driving ground vehicles and space exploration, while marine robotics received much less attention. But with 90% of goods moved across the world in vessels, the interest in developing autonomy capabilities for unmanned surface vehicles (USV) has been increasing. A crucial element for autonomous operation is environment perception, which lags far behind the control and hardware research in USVs. In a closely related field of autonomous vehicles (AV), perception advancements have been primarily driven by deep models, which allow end-to-end learning of complex functions required for a reliable operation. But most of the perception methods developed for USV are hand-crafted or re-use parts of deep models pre-trained on general-purpose RGB-only datasets and fine-tuned on much smaller corpora of maritime images. These datasets are insufficient for developing and training complex models required for highly dynamic, illumination-varying marine environment, in which specular reflections, haze and mirroring are frequently observed. Thus an opportunity is missed for end-to-end training for complex maritime perception tasks due to the lack of sufficiently large and diverse multi-modal datasets that would reflect the behavior of typical USV sensor modalities in a maritime environment. Another issue with pre-trained models is a limited generalization ability. A sensor replacement or deployment of deep pre-trained methods at a new location (e.g., moving AV from city to a rural environment), typically requires re-capturing and re-annotation of a dataset for re-training the deep perception models, which is time consuming and costly. This is even more pronounced problem in a highly variable maritime environment. Our goal is to develop the next-generation marine environment perception methods, which will harvest the power of end-to-end trainable deep models. Research challenges essential for safe USV operation will be addressed: (i) general obstacle detection, (ii) long-term tracking with re-identification, (iii) implicit detection of hazardous areas and (iv) multi-modal sensor fusion. Particular focus will be placed on the adaptivity of the models and self-supervised tuning to new environments. New multi-modal datasets will be captured to facilitate the development of these next-generation models. The project will be composed of six work packages. Deep models for robust obstacle detection with scene adaptation capabilities (WP1); Segmentation-based tracking algorithms compatible with the deep obstacle detection architectures (WP2); New trainable deep sensor fusion methods for environment perception (WP3); We will construct large annotated multimodal USV datasets for training and objective evaluation of deep networks in realistic scenarios (WP4); Two work packages (WP5 and WP6) will contain support activities such as results dissemination and project management. Three partners will be involved in the project: the Visual Cognitive Systems Laboratory at the Faculty of Computer and Information Science, University of Ljubljana (ViCoS), the Laboratory of Machine Intelligence at the Faculty of Electrical engineering, University of Ljubljana (LMI) and SIRIO d.o.o research group. The ViCoS members will focus on adaptive semantic segmentation for detection and tracking, while the MVL members will focus on deep sensor fusion, autocalibration and hazard detection. Members of SIRIO d.o.o. research group have been developing unmanned surface vehicles for over a decade. They will be responsible for integration of the system on their USV, data-set acquisition and validation of selected methods on their USV. The combination of expertise in the field of computer vision and machine learning and the specific expertise in robotic boats development will guarantee the project success.