Low Cost GNSS and Computer Vision Fusion for Accurate Lane Level Navigation and Enhanced Automatic Map Generation

Informations projet

INLANE

N° de convention de subvention: 687458

DOI

10.3030/687458

Projet clôturé

Date de signature de la CE 1 Decembre 2015

Date de début 1 Janvier 2016

Date de fin 30 Juin 2018

Financé au titre de

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies – Space

Coût total

€ 3 281 028,75

Contribution de l’UE

€ 2 642 935,88

Coordonné par

FUNDACION CENTRO DE TECNOLOGIAS DE INTERACCION VISUAL Y COMUNICACIONES VICOMTECH
Spain

Periodic Reporting for period 2 - INLANE (Low Cost GNSS and Computer Vision Fusion for Accurate Lane Level Navigation and Enhanced Automatic Map Generation)

Période du rapport: 2017-07-01 au 2018-06-30

Résumé du contexte et des objectifs généraux du projet

Nowadays lane navigation consists in preparing the exits and turns on difficult roads by showing the driver the lane and exit that he or she needs to follow. The information is visual (usually showing still images). This is informative, rather than being based on the vehicle’s real position on the road. Additionally, the system gives independent voice alerts as to whether you are correctly placed or not, and this in some cases can lead to confusion and loss of attention for the driver.

The current state of mapping is sufficient for the basic navigation applications, but the cartography data is facing a great challenge since the current data is not going to be enough for next-generation uses, such as lane level position in Advanced Driver Assistance Systems (ADAS), hyper-specific location-based services and self-driving cars.

The main objective of inLane is to develop a new generation, low-cost, lane-level, precise turn-by-turn navigation application through the fusion of EGNSS and Computer Vision technology. This will enable a new generation of enhanced mapping information with real-time updating based on crowdsourcing techniques. The resulting lane-level vehicle positioning will bring navigation to a new level of detail and effectiveness.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The project activity was focused on developing the tested and user aproved inLane prototype for lane level navigation and live map updates. For succesfully achiving this objective three different prototypes has been developep from alpha (aplication and functionality skeleton approach) to fully functional Gamma prototype that was tested and validated with endusers.

For each developemnt period three stages were defined and followed firstly definition of general requirements and functionality that was refined in each cycle followed by the development phase and validation phase to gather useful information for the subsequent cycle and reduce the deviation between technology and user expected functions.

In parallel, the strategy to involve end users in the project was defined, from the begining of the Project: the alpha prototype was tested with technical expert users, and a user survey to gather end users’ opinion and feed functional requirements and use cases were designed. Beta prototype was tested with technology background but not involve in the Project development and Gamma prototype was tested with potential final users with no specific background to validate the usability and usefulness of inLane outcome.

During the lifetime of the project, the consortium also focused on dissemination, exploitation, standardisation and management activities being very active in conress participation specially those related with ITS, Automotive sector and GNSS. The consortium organized a Final Event in Barcelona with large impact: with more tan 100 assistants from different sectors that had the opportunity to see the Project outcomes demos live. Each of these demonstrators is focused on one of the inLane’s core functionalities:

–Scene understanding
–Sensor fusion
–Local dynamic map generation
–Crowd-sourced map updates
–Camera to map alignment
–Lane-level navigation (live demo in a vehicle)
–Augmented reality for ADAS

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

1. Expected results and progress beyond state of the art
inLane, using a low-cost and therefore accessible model, will provide drivers with the following advantages, progressing beyond the state of the art and fostering a new generation of driver assistance navigation:
• Accuracy and integrity of the calculated position thanks to the use of EGNSS signals
• Road elements automatic detection using advance computer vision techniques
• The development of standards for coding new road data content classes
• Cartography data automatic generation and location with high precision and integrity
• Local Dynamic map information layer generation for assisting the driver
• Vehicle positioning in the lane (contrary to the most common approach in map matching of simply projecting the vehicle position on the centre of the segment)
• The relative lateral position of the vehicle on the carriageway, informing the user of how many lanes the carriageway has at its current cross section and on which one of these lanes the vehicle is
• A level of confidence on the position level that is based not only on GNSS measurements (as traditional integrity parameters do) but on the combination of all positioning information sources and the enhanced maps
• A level of reliability in terms of lane

2. Potential impacts
inLane will stimulate the use of European GNSS through demonstrating the ability to use EGNOS/EDAS and early Galileo services to obtain the accuracy needed to ensure a smooth, safe and precise vehicle positioning and cartography generation. This will be used for next generation driver assistance application such as lane level navigation that will be validated in this project and also to generate new, enhanced cartography. By working on the surface transportation area, the consortium intends to demonstrate how most modern EGNSS technologies can be applied in traditional sectors and become part of the tools accessible to everyone in their everyday life.

In particular, this project seeks to show to the mass market end-user that:
• Galileo or EGNOS based services bring enhanced accuracy, safety and integrity (that the GPS cannot typically bring) of the position and thus it usage into the driver assistance domain
• This enhanced safety and integrity can be a direct and significant benefit in terms of general public mobility, since the inclusion of this technology will be straight forward in vehicles, aftermarket devices and smartphones.

In doing so, inLane will also be fully in line with the Work Programme Objective and European Space Policy, which reflects the key strategic importance of space systems and space applications for Europe.

ITS is one of the leitmotiv themes in the European Transport Roadmap to 2050. Galileo will afford considerable advantages in many sectors of the economy. In road and rail transport, for example, it will make it possible to predict and manage journey times, or, thanks to automated vehicle guidance systems, help reduce traffic jams and cut the number of road accidents.

logo.png

inlane-architecture.jpg