Industrial Partnerships

Project Title: Privacy Preserving Risk Assessment of Human &
Automated Driving using Sensor Fusion

The shift to automated driving will be gradual and steady, as each year cars are outfitted with more sensors and systems to improve driver performance and safety. These advances, collectively known as ADAS (advanced driver-assistance systems) include a number of developments from autonomous vehicle research that have already made a debut on public roads, such as adaptive cruise control, automatic parking, and much more. The 360Lab’s collaboration with Foyer Assurances Luxembourg investigates how these steps toward fully automated driving are already changing the automotive insurance market.

Project Title: Luxembourg in High Definition

To get a glimpse of the potential benefits of having access to a high-definition digital map of a city, one just has to observe the impact the Internet has had over the years on individuals and societies. Not only did the Internet democratize access to information, but it also created a new global marketplace and made all markets more efficient connecting buyers and sellers. It was also a key enabler for new businesses such as Uber and Airbnb, that have transformed transportation and travel respectively. The Internet ecosystem now affects the economy in ways that were completely unimaginable just a few years ago.

An HD digital map of a city will be a similar ubiquitous backbone of the future. While several applications of HD Maps such as those in Autonomous Vehicles, Autonomous Delivery Robots, and Smart City initiatives are already obvious, the map also serves as a foundation on which additional contextual information can be constantly layered from different sources, transforming the map into a living digital asset that will power the innovations of tomorrow.

The aim of this project is to create a high-definition digital base map of the city of Luxembourg and the digital cross-border testbed area. This digital base map will contain features that will have 15-20 cms absolute georeferenced accuracy and will enable autonomous vehicle navigation including side-walk navigation for delivery robots.

Nationally-funded Research Projects

FNR Bridges

Project Title: Connected Dynamic Insurance

Industrial Partner: Foyer Assurances

Cars are getting increasingly connected. In the past, insurance telematics products have been based on retrofitted tracking systems, commonly called “Black Boxes”, and used to calculate insurance premiums based on simplistic metrics, such as the distance driven. Those systems however have drawbacks. First, the users do not like to install such devices in their private cars, as they are usually perceived as invasive. Second, the cost of the device is usually significant, hence reducing the margin for the insurance provider who often covers their cost. In Europe, as of March 2018, every new vehicle has to be equipped with the so called “eCall” system, which allows to locate and communicate with a car if a crash is detected. Many car manufacturers are adding additional services on top of eCall to generate new revenue streams. This project will make use of such services to retrieve telematics data from connected vehicles (with the consent of the owner) to develop and test an insurance telematics product with our partner Foyer Assurances Luxembourg.

FNR Industrial Fellowship

Project Title: Machine Learning for Risk Assessment in Semi-autonomous Vehicles (MASSIVE)

Industrial Partner: Foyer Assurances

Individual PhD grant for François Robinet

In the next decade, the increased number of assistive systems and the advent of semi-autonomous vehicles are expected to lead to a reduction of both the frequency and magnitude of road accidents, which will force insurance companies to adapt their offering. One option is to use telematics to monitor drivers and reward safe-driving habits by lowering premiums. Previous attempts have generally limited themselves to using sensors that are readily available on customers’ smartphones, such as GPS, but the collection of such data cannot capture the complexity of real-world driving scenarios.

In this project, we collaborate with Foyer Assurances SA to research novel real-time risk assessment techniques by leveraging more advanced sensors. Our research group has recently built a prototype car for driverless mobility research, which is instrumented with a plethora of such sensors: cameras, LIDAR, differential GPS, driving control signals, etc. Our aim is to collect driving data in a wide range of realistic scenarios, and to combine it with historical insurance claim datasets recorded by Foyer. Using this data, we will be able to design human-understandable features for safe driving, and to train machine learning models to infer these features from raw sensor measurements. Our risk assessment pipeline will finally be set up on small embedded computers, to be tested and evaluated on board of a fleet of cars from volunteering employees from Foyer Assurances SA.

FNR Industrial Fellowship

Project Title: H​igh level U​nderstanding of 3D map ​GE​neration for autonomous navigation of cars (​HUGE​)

Industrial Partner: Civil Maps

Individual PostDoc grant for Arun Annayian

In this project we propose a localization system with forward looking monocular camera data with a given HD map generated by LiDAR and camera sensors. Localization using cameras is valuable because it is widely available on consumer vehicles for updating the map and it has RGB data which is important for classification of signs and lane markings. LiDAR has the advantage of directly reassuring depth whereas with cameras, the depth has to be inferred. By using a reference map created with camera and LiDAR, the depth of the camera can be more accurately inferred which improves both the localization of the vehicle and the precision of the extraction of the map object to be updated. Because it gives much denser data compared to sparse point clouds. In sparse point clouds it would be difficult to find an object unless it is densely represented. However RGB images are relatively easy to identify objects like sign boards and traffic lights irrespective of the angle and distance. Camera and LiDAR data can be fused sequentially. In this work the road contains the lanes, traffic sign boards, lights i.e landmarks are to be detected. To attain this we will run Inverse Projection Mapping (IPM) in frontal camera image plane in order to match with the HD map. Here lanes and sign boards are for lateral and longitudinal localization respectively. The proposed method enables the user to accurately localize in the already captured large scale maps using LiDAR sensor.

FNR Pathfinder

Project Title: DigitalUs (part 1)

Everyday tons of personal digital data is being generated and stored on social media and other platforms such as forums and review sites, and the rate is growing exponentially. It becomes increasingly important to manage, protect and interact with this data by providing a comprehensive overview of our digital self. The aim of this project is to validate a technology developed at SnT that allows to crawl and combine various data sources and find a suitable market for it.

FNR Proof-Of-Concept

Project Title: DigitalUs (part 2)

Every day, vast amounts of digital data is being generated on the Internet. This trend is increasing exponentially. A large amount of this data contains personal information and is publicly accessible. This data is generally distributed among many different data repositories (Social Media, Blogs, News Articles, etc.). Having access to a single repository does usually not provide any useful information, however combining all those data sources for one unique individual allows to create a public digital profile that includes commercially exploitable information. The AI- based technology that will be tested during this project allows to reliably match various public data-sources for any given person to compute a comprehensible digital profile. There are multiple approaches on how such a technology can be commercialized. A market study has been conducted as part of a Pathfinder project. The outcome suggests to start with a Business to Business (B2B) model as many companies are relying on personal data to provide new services. In this project we will therefore closely collaborate with a local insurance company, that would like to rely on digital profiles to propose a new kind of cyber-insurance to their customers.

EU-funded Research Projects

Project Title: TERMINAL

Luxembourg’s strong commuter culture and geographic location also make it a compelling laboratory for exploring the future challenges and opportunities of automated driving in Europe. The project TERMINAL, funded by the ERDF as part of the INTERREG V A Greater Region program, is already bringing together municipalities, researchers, and businesses from across the greater region to explore the possibility of an automated cross-border shuttle bus system. The three-year project will culminate in a six-month test of an electric automated minibus, which will operate a route across the German-French border in 2020.

As part of the project, the 360Lab will be collecting and analysing data from a number of Luxembourg’s own cross-border bus routes.

Visit the project website for more information.