Projects

The Cadweiser Traffic Info pilot (Cadweiser Oy) uses generative AI to simplify traffic bulletins, converting text into speech and transmitting it via radio and communication channels. The goal is to learn how AI can improve the accessibility of bulletins in different languages and with customised settings.

The LLM CAM pilot (NinjaLabo) aims to develop a No Code Vision LLM camera platform that allows users to create smart city services using natural language without coding skills. Key phases include integrating YouTube Live and ChatGPT for initial functions, such as enabling users to inquire about free parking spaces or traffic conditions. The solution supports a community-driven platform that promotes collaborative, no-code smart city solutions.

The pilot with Dreamwaves addresses the lack of detailed accessibility information. The pilot was selected through the EIT Urban Mobility’s RAPTOR program for the challenge posed by Mobility Lab Helsinki (read more about the challenge here).

The solution is based on a mobile application that uses augmented reality and allows users to collect and annotate obstacles and issues on the streets, such as blocking objects like construction sites, e-scooters, potholes etc.

MarshallAI’s pilot project helps to understand, maintain and assess the demand for and future needs of bicycle parking services. Anonymous real-time data can be collected on the use of bicycle parking facilities using computer vision and video automation. The solution monitors the utilisation rate of bicycle parking facilities, bicycles parked at parking facilities for an extended period of time, as well as the number of bicycle thefts or attempted thefts.

The Norwegian Bikeloop AS‘s smart bicycle parking solution is a secure modular bicycle locker for eight bicycles. Booking bicycle parking is possible for anyone via their app. The bicycle locker design takes special groups into account, and the lights and voice guidance make the locker accessible.

In the Jätkäsaari area, utilising the smart infra from the Smart Junction project, piloting of C-ITS solutions with a couple traffic lights and vehicles. Demonstrations planned for 2025 after first developing it through simulations. More info about the use case of preventing rear-end collisions: https://www.smart-edge.eu/2023/07/25/use-case-preventing-rear-end-collisions-by-enhancing-road-intersection-safety/

In Gispo‘s experiment, the goal is to increase the potential use of the city’s 3D materials by enabling the utilisation of 3D data in QGIS for a wider user base within the city organisation. 

Webion participates in the experiment with the aim of developing the measurement of urban spaces. The technologies in the experiment can be used in the future to produce measurements of urban spaces that promote the development of digital twins of cities as well as maintenance. In the experiment, among other things, street-level imagery is refined using artificial intelligence. 

Sitowise is developing a new and easily scalable method in which data from various traffic infrastructure sources is analysed and classified automatically using commercial tools and artificial intelligence algorithms. 

The results of the pilot procurement will be used for the implementation of the CityGML 3.0 street environment test data in 2024.  

The project involves using light radar technology (lidar) to gather data on vehicles, bicycles, electric scooters, and pedestrians. The aim is to understand the impact of the changes on the Esplanadi on traffic flow and traffic behaviour.  

In a pilot conducted by Sitowise, the goal was to produce a 3D model of the urban environment, which can be utilized in examining various modes of transportation environments in Helsinki. The method used in the pilot was a continuation of Sitowise’s work created in the spring of 2022, where a tool was developed using the FME conversion process to transform road features from the General Area Registry (YLRE) into CityGML format. 

Zero Gravity was developing a way to combine existing City of Helsinki data on traffic areas with manual labelling for the training dataset as well as creating a segmentation model to extend to the rest of pilot area.

The three-month case study pilot was carried out in collaboration between Drover AI, Voi Technology, Vianova, VTT and Mobility Lab Helsinki. Drover’s PathPilot solution was installed on 25 of Voi’s e-scooters. In the first phase, it collected data about the infrastructure the e-scooters ride on. The second phase of the pilot introduced audio signals related to rider behaviour (such as alerts when riding on pedestrian streets, parking).

The case study produced a white paper analyzing the observed rider behaviour. The results were promising. For example, riding e-scooters on sidewalks was reduced during the test period. Continuation of testing with an expanded time and scope might be considered. Read more about the results.

“From the images collected by the scooters, we obtained accurate information about, among other things, potholes on sidewalks, which can be a safety risk for road users. Additionally, several situations were observed where, for example, a parked car blocked access to the bicycle lane,” says Juho Kostiainen, Project Manager of Mobility Lab Helsinki.

The pilot with Marjetas Academy is one of the agile pilots selected in Oct 2022 (read more in Finnish) which aim to support the development of Helsinki’s cityinfra platform.

The pilot with Normiopaste is one of the agile pilots selected in Oct 2022 (read more in Finnish) which aim to support the development of Helsinki’s cityinfra platform.

The URBANE project is aimed at developing a replication and scale-up model for successfully replicating smart green last-mile delivery solutions. It will use a PI-inspired innovation framework and digital twins with real life pilots in different cities around Europe. The activities in Helsinki are focused on developing and testing sustainable and viable last mile delivery solutions, and particularly autonomous systems. The Helsinki pilot partners include LMAD (Last Mile Autonomous Delivery, operating an autonomous system), DB Schenker (as the logistics operator) and Soben/TwinswHeel (as a provider of the autonomous platform) with Forum Virium Helsinki managing the pilot with the support of Helsinki/Mobility Lab.

In the pilot, xyzt.ai’s visual analytics platform is used for studying traffic flows and quality of data sets provided by GeoMobility. The objective is to better understand the quality and coverage of different data sets in the Helsinki area, and how they serve different needs and how they can be analyzed (e.g. routes, speeds and delays of traffic). For the companies, it provides a chance to build knowledge on regional data sources, challenges and needs.

The public summary report is available here as pdf.

The DVECE project pilots the use of innovative and dynamic visualisation tools within the context of traffic and mobility. The project is testing whether feedback data, ideas and suggestions concerning traffic and mobility can be collected through the City of Helsinki 3D model on the Decidim platform.

The project includes data collection pilots and the development of innovative data acquisition models especially in the context of Stara, the Helsinki City Construction Services and their operations.

The project’s objective is to create an operating model for companies to convey last mile delivery information and details to drivers as open data. It will promote the operating conditions and improve cost efficiency of logistics operations, improve the service levels for the customers and reduce the environmental effects through optimization (e.g. reduced time spent looking for parking spots).

The data is stored and available in OpenStreetMap and for demonstration purposes an example application app.gatesolve.com was created. It visualises entrances and routes of last 50 meters to them. In addition to the navigation application, another (open source) application, Open Logistics Map (OLMAP), is available for collecting such data. 

A final report for the project can be found from Tieke’s web site here.  

Relevant aspects developed and tested in Helsinki include MyData and piloting cases such as using data to address blind spots behind corners in the Jätkäsaari area.

The API documentation for the open data can be found here: https://urbsense.urbanplatform.city/api/public/docs/.

This pilot is one of the selected agile pilots from the call in the summer of 2020.

This project is one of the selected agile pilots. The solution will be tested in the spring of 2021 as the first phase of the agile pilot process.

This project is one of the selected agile pilots. The solution was tested in the spring of 2021 as the first phase of the agile pilot process, the second phase of the pilot is planned for spring of 2022.

This project is one of the selected agile pilots. The solution was tested in the spring of 2021 as the first phase of the agile pilot process.

This project is one of the selected agile pilots. The solution was tested in the spring of 2021 as the first phase of the agile pilot process.

This project is one of the selected agile pilots. The solution was tested in the spring of 2021 as the first phase of the agile pilot process. In the fall of 2021, a second phase of the pilot was done, extending the pilot to also include the testing of C-ITS messages based on the detections.

Professional drivers from logistics companies take part in the agile pilot by giving feedback about the feasibility and further development needs of the solution. They also validate and complement the data on loading sites. This pilot was one of the agile pilots done in 2020/2021.

Traffic measuring points for the main junctions for vehicle traffic to and from Jätkäsaari. The objective is to get a better understanding of the traffic to and from Jätkäsaari. In addition to just number of vehicles, also the time spent there (e.g. short visits for maybe dropping off people to the terminal) as well as estimates of the emissions they bring in. There will be two measuring periods, first one being in Nov/Dec 2020 and the other in the spring of 2021.

The camera-based devices automatically identify licence plates of cars, and derive technical vehicle information from Traficom’s database. The use of licence plate information is approved by Traficom and it complies with national privacy and data security requirements.

No personal information (such as car owners’ names, addresses etc) is queried in the process. Vehicle information is used to produce traffic statistics, and individual road users cannot be identified at any stage of the analysis.More information on Nodeon’s solution, including data privacy can be found here.

The pilot implemented a local distribution centre for parcels. The hub, operated jointly by six operators, is located next to Helsinki’s busy Baana pedestrian and bicycle corridor. It was opened on 9 June 2020. The pilot ran until the end of 2020 and during that time, companies could bring their parcels to the shared distribution centre instead of every delivery company driving into the congested city centre with their own vehicles. From there, the deliveries were coordinated resource-efficiently utilising electric bikes and other low-emission transport modes.

The LMAD project includes building a robot, a fleet management system that enables autonomous mobility and an order system that allows the robot to communicate with customers. In addition to the technical implementation, a service concept is created by means of co-creation and service design and tested in a genuine urban environment with genuine customers.

The data is stored and available in OpenStreetMap and for demonstration purposes an example application app.gatesolve.com is created. It visualises entrances and routes of last 50 meters to them. More information on the collected data and a preview service is available in the Materials and Data section.

The development done in this project is continued in 2021-2022 in the Logistics Accessibility Data project.

Vianova’s solution is based on MDS (Mobility Data Specification) communication protocol which enables bilateral and secure exchange of data between mobility providers and cities.

First, data is collected to detect road crossings. Later, the intention was to send alerts nearby drivers to prevent a potential collision but due insufficient amount of data and users, the pilot did not proceed to the next phases.

The technology can alert drivers even if the pedestrian is located behind an obstacle and the driver can’t see him. The incident data can also be used by the city itself to improve the city road safety infrastructure. Jätkäsaari Mobility Lab supports the pilot by helping in finding test users to use the application which is available both for Android and iOS. The application also utilises Helsinki’s open data to map whether road crossings are done at crosswalks or traffic lights. 

Key functionalities that were planned:

Pedestrian mode

– Road safety behavior stats to improve on-road behavior
– Real-time alert in case the user watches the smartphone when approaching a road
– Volume control – Adjusting headphones volume near roads

Driver mode

– Collision detection alert – Alert drivers on possible collision with pedestrians, even when the pedestrian is behind obstacles and the driver cannot see him
– Alerting drivers in possible collision situations
– Real-time notification when entering crowded pedestrian areas

City road safety dashboard

– Aggregated and statistical data for the city for better planning the physical infrastructure for pedestrians’ road safety (e.g. crosswalk planning, bicycle trails, traffic light optimization)

This data feeds to further research on possibilities to improve traffic fluency with the help of traffic signal optimization based on the application of AI. The area monitored by the first setup near the Clarion hotel is illustrated in this image of the installation, camera view and radar. Open data from the radars is available for third party development purposes, more information on it can be found here.

Project focuses on port-related traffic in Jätkäsaari, street canyons and 40K daily vehicles in Mäkelänkatu-Kumpula area, and domestic wood-burning in combination with adjacent main roads in Pakila.

The overall challenge is that measuring, analysis and dissemination of air quality data in any given location is very complicated and contaminated with error. Lack of real-time and reliable high resolution data has led to a situation where finding targeted interventions that take into account our citizens and their needs while also targeting specific pollution concentrations is missing for all European cities.

The project’s influencing methods are co-created with residents. At the same time, the project creates tools for both residents and data collection, thus facilitating local-level city planning.

Related open data and APIs: https://helsinki.liikennenyt.fi/opendata/

As continuation for a FinEst Smart Mobility innovation pilot, a traffic situational picture solution is implemented in West Harbour. The project produces open data and information for third parties, such as traffic events and road incidents in the area, traffic congestion, route recommendations as well as the number of vehicles arriving through the harbour gates.

Apart from enhanced visibility, the smart traffic sign includes technology that generates data on traffic volumes, speed, air quality and environmental conditions for transport and city planners.