The Spatiotemporal Data Analysis (SDA) research group does research on Artificial Intelligence, mainly deep learning and computer vision based methods to provide tools for developing sustainable smart cities.
AI for sustainable smart cities

CouSCOUS - Sustainable urban development emerging from the merger of cutting-edge Climate, Social and Computer Sciences

In the fight against climate change and coping with the pressures arising from urbanisation, cities worldwide make solutions modifying pollutant emissions and thermal comfort. The consortium answers to the possible challenges on air quality by helping cities to plan climate healthy future urban areas, considering future traffic flows and population structures. CouSCOUS combines the fields of artificial intelligence, atmospheric and social sciences in a way that has not been done is this extent before and therefore advances the state of scientific research in all disciplines involved and provides decision makers and city planners globally with novel relevant information and tools for creating future sustainable cities. Consortia will utilize various detailed data sources, from grid-level population data to climate and traffic data and analyses them. The consortia will work tightly both together as well as in co-operation with relevant stakeholders.

SDA develops deep learning methods for traffic flow prediciton and for providing recommendations for sustainable city planning. (Funder: Academy of Finland 2020-2024)

Relevant publications for this project:
Joanne C. Demmler, Ákos Gosztonyi, Yaxing Du, Matti Leinonen, Laura Ruotsalainen, Leena Järvi, Sanna Ala-Mantila (2021)
A novel approach of creating sustainable urban planning solutions that optimise the local air quality and environmental equity in Helsinki, Finland: The CouSCOUS study protocol

Artificial Intelligence for Urban Low-Emission Autonomous Traffic (AIforLEssAuto)

At present, road transport contributes a significant amount to the total carbon dioxide (CO2) emissions in the EU. Thus, cities look for practical strategies to make their transport system more efficient and sustainable. Electrification of road transport is the primary technological change needed to meet the carbon reduction targets. However, electrification is unlikely to be sufficient since the electricity production will not be carbon neutral in the near future. There is a second major technological transformation on-going in road transport—digitalization—bringing forth the advent of connected automated vehicles. Connected automated driving will transform traffic flow management into a proactive disaggregated and cooperative paradigm that, via appropriate strategies, may enable a decrease in CO2 emissions. However, the total joint effects of electrification and autonomy on CO2 emissions are not well understood. There are major potential cross-effects, such as an increase of vehicle-km travelled due, for example, to an autonomous car visiting a recharging station. Furthermore, the transitions will not be instantaneous but electric and combustion engines, and automated and human-operated vehicles will co-exist for a significant period, which is not typically taken into account in existing studies. When combined with new possible vehicle ownership models and policies, the impact of automation on urban traffic remains highly uncertain. In the future, digitalization and communication technologies may also enable much more flexible management of the existing infrastructure.

AIforLEssAuto brings together atmospheric and computer scientists and traffic engineers in active dialogue with municipal stakeholders with the ultimate aim to understand how autonomous electrified traffic should be organized during the transition period in order to reduce CO2 emissions. This is achieved by 1) building a framework of computational modelling tools to evaluate the CO2 emissions originating from electrified automated vehicles, and 2) developing artificial intelligence based control from vehicle-level to city-center wide traffic-level in which CO2 emissions are minimized. Such multidisciplinary approach has not been done to this extent before and thus AIforLEssAuto advances the state of scientific research in all disciplines involved and the novel combination will certainly lead to new, scientifically and societally, important breakthroughs.

SDA develops reinforcement learning methods for traffic simulating city level future traffic. (Funder: Academy of Finland 2022-2024)

UAS (Drone) University Collaboration Network

University network for developing drone related research and education in Finland. (Funder: Ministry of Culture and Education, 2021-2023)

Navigation and Situational Awareness

Computer vision

Computer vision provides crucial information for various applications requiring reliable navigation and situational awareness. SDA group develops novel methods for Simultaneous Localization and Mapping (SLAM) in challenging environments and 3D object tracking for improved safety for autonomous systems.

AIV - Artificial Intelligence for Industrial Vision

The project aims at developing cost-effective and reliable Visual Simultaneous Localization and Mapping (SLAM) and deep learning methods for 3D object tracking for improving industrial safety. Research is funded by a donation done by Konecranes for 2020-2021.

Relevant publications for this project:
Niclas Joswig, Juuso Autiosalo, Laura Ruotsalainen (2023)
Machine Vision and Applications
Improved deep depth estimation for environments with sparse visual cues

Videos about the pro­ject:
FCAI's video presenting AIV project
Win-win: Konecranes and FCAI collaborate to develop intelligent cranes
News about the pro­ject:
Published on 15.1.2020
Mas­ter's theses made about the pro­ject:
Monocular 3D Object Detection And Tracking in Industrial Settings
Evaluation of Deep Learning-based SLAM in Industry Vision
Non-Lambertian surfaces and their challenges in computer vision


CANDO II - Collaborative Augmented Navigation for Defence Objectives II

The CANDO II project will bring together navigation experts from SINTEF (Norway) and the University of Helsinki to build upon the results and lessons learned within the CANDO project (2018-2019) to design and produce a cutting-edge navigation system to address the situational awareness (blue force tracking) and navigation needs of dismounted soldiers with support for human-machine teaming in conducting building clearing and urban operations. Full exchange of relevant research background and foreground by both institutions will ensure the best possible technological outcome. The program goal of young researcher support will be addressed through the involvement of graduate students from the University of Helsinki in training activities. (Funder: NATO Science for Peace and Security 2021-2023)

Relevant publications for this project:
Laura Ruotsalainen, Aiden Morrison, Maija Mäkelä, Jesperi Rantanen, Nadezda Sokolova (2021)
IEEE Sensors Journal
Improving Computer Vision Based Perception for Collaborative Indoor Navigation

Global Navigation Satellite Systems (GNSS)

Availability of a global navigation and timing solution provided by GNSS has accelerated the appearance of various applications supporting sustainable development in the forms of improved safety, decreased environmental burden, new business opportunities and increased quality of life. However, the applications have demands on the PVT reliability, availability and accuracy that GNSS was not developed to support. The need for an ubiquitous GNSS that can adapt to the signal environment is self-evident. The break-throughs in the machine learning field, namely the development of accurate deep learning (DL) methods, have provided huge improvements in computation in many disciplines. However, GNSS and navigation field has been slow to introduce DL into research. SDA group provides novel machine learning methods for improving the performance of a GNSS receiver and providing ubiquitous adaptive GNSS.

Reason - Resilience and security of geospatial data for critical infrastructures

We will construct an automatized tool to predict the vulnerability of satellite navigation (GNSS) signals to ensure the continuity and resilience of location and timing services that are vital to a number of critical infrastructures, such as, search and rescue, electrical power grids, banking, security information systems, communications, and transport. We will utilize the FinnRef reference network and the GNSS Finland monitoring service and enhance it with temporary reference stations and signals of opportunity to monitor the GNSS signal quality in critical locations. Using the big data acquired with these networks, we will compute future trends to forecast the critical failures in both positioning and timing information. The research will also contribute to the robustness of the Finnish timing service. We will make this information available for the end users to enhance their capability for cyber security, situational awareness, and digital services.

SDA develops deep learning methods for detecting signal abnormalities and system failure and for mitigating their effects on critical infrastructure (Funder: Academy of Finland 2020-2023)

Relevant publications for this project:
Arul Elango, Sahar Ujan, Laura Ruotsalainen (2022)
2022 International Conference on Localization and GNSS (ICL-GNSS)
Disruptive GNSS Signal detection and classification at different Power levels Using Advanced Deep-Learning Approach

Arul Elango, Ahmed Al-Tahmeesschi, Mikko Saukkoriipi, Titti Malmivirta, Laura Ruotsalainen (2022)
WHITE PAPER: Protecting GNSS Against Intentional Interference


Advanced RFI Detection, Alerting and Analysis System II (ARFIDAAS2)

Advanced detection, alerting and analysis system for intentional GNSS interference. (Funder: European Space Agency, Navisp element 3 2021-2022)

5G signals

5G signals will provide opportunities for more accurate positioning, ranging between different actors and thereby improved cooperative positioning, and for transmitting information fast between different users e.g. in the autonomous traffic ecosystem.

Robust radio signal-based navigation

The availability and reliability of user equipment localization is a key enabler for numerous applications. Some examples include autonomous driving and location based advertisement. While most of current solutions are based on Global Navigation Satellite Systems (GNSS), maintaining localization accuracy in densely built-up areas may not be achieved. One of the modern radio networks requirements (such as 5G and beyond) is to be able to track the mobile user equipment at all time. Thus, it may be used for complementing or replacing GNSS signals in case they are degraded or denied. This research aims to utilize received 5G signals combined with deep reinforcement learning to improve the user localization accuracy in 5G and beyond radio networks.

Relevant publications for this project:
Ahmed Al-Tahmeesschi, Jukka Talvitie, Miguel López–Benítez, Laura Ruotsalainen (2022)
2022 International Conference on Localization and GNSS (ICL-GNSS)
Deep Learning-based Fingerprinting for Outdoor UE Positioning Utilising Spatially Correlated RSSs of 5G Networks

Past projects

5GIVE - 5G- assisted Ground-based Galileo-GPS receiver Group with Inertial and Visual Enhancement

The project aims to create methods fusing GNSS, sensors and terrestrial positioning signals for seamless navigation of ground based users, especially automated vehicles. Research at the Department of Computer Science (UH) will address the use of 5G-signals for improved positioning performance via two methods; by transmitting data for the high-accuracy hybrid PVT engine and by developing methods for ranging between vehicles in a platoon. Security and privacy requirement for the data will be also addressed for both methods. (Funder: European Space Agency, 2019-2021)

Relevant publications for this project:
Ashwin Rao, Martti Kirkko-Jaakkola, Laura Ruotsalainen (2021)
2021 Joint European Conference on Networks and Communications 6G Summit (EuCNC/6G Summit)
Dissemination of GNSS RTK using MQTT


Multiple complementary positioning sensors have to be used for managing varying positioning environments and for providing increased accuracy and integrity via sensor fusion. The main goals of the project are the compilation and evaluation of a concise and robust indoor / outdoor pedestrian navigation system that includes a Test-Bed collecting PNT related data and algortihms for computing an accurate and reliable navigation solution. (Funder: European Space Agency, 2019-2020)

CANDO - Collaborative Augmented Navigation for Defence Objectives

This project seeks to address the personnel protection goal of the counter terrorism key priority through provision of enhanced situational awareness and blue force tracking within urban and indoor environments. To achieve this goal, the project builds on the concept of 'Collaborative Navigation' (CN) whereby members of a platoon operating in the same area are able to act as navigation references and communication nodes for one another. The CN concept provides two primary benefits – 1) The enhancement of absolute and relative position information for every cooperating user, allowing extended indoor operation independent of satellite navigation, and 2) The dramatic enhancement of situational awareness information to every member of the formation, even those that are beyond direct communications range due to the relay nature of the network. (Funder: NATO Science for Peace and Security for SINTEF, Norway and Finnish Geospatial Research Institute, Finland. University of Helsinki for research done in Finland, 2018-2020)

Testing the computer vision and inertial sensor based cooperative navigation solution in the CANDO project.

RAAS - Research Alliance for Autonomous systems

RAAS ( is the leading interdisciplinary innovation ecosystem and service platform for autonomous systems R&D in Finland. Spatiotemporal Data Analysis group is heavily involved in the Situational awareness, Autonomous Navigation & Intelligent Control Research Task Force (RTF) 2018-2020.

White paper: Safety Challenges of Autonomous Mobile Systems in Dynamic Unstructured Environments: Situational awareness, decision-making, autonomous navigation, & human-machine interface…