Validation of Mapping and Localization for Autonomous Vehicles (VAMLAV)
|Coordinator||ASTAZERO AB - AstaZero AB, Göteborg|
|Funding from Vinnova||SEK 5 749 416|
|Project duration||October 2019 - September 2021|
|Venture||Traffic safety and automated vehicles -FFI|
|Call||Traffic safety and automated vehicles - FFI - spring 2019|
Purpose and goal
The main objective is to create an open dataset over Rural Road at AstaZero for self-driving cars. The dataset is intended to include HD map data over the track Rural Road at AstaZero. In addition to the HD-map, the dataset will also include sensor data collected by cars that have driven multiple laps around Rural Road in different environment conditions to include repeatability. A part of the map generation is supposed to include crowdsourced data. The dataset will then be used to validate map generation and localization algorithms.
Expected results and effects
The dataset is intended to help the development stage for robust ADAS and self-driving car systems, especially validation of mapping and localization systems. The dataset will be made in a well-known and controllable geographic area and the fact that it will include repeatability, different enviroment variabels and a corresponding HD-map makes it one of a kind. The dataset will therefore enable the capability of more research within the fields: Mobility as a service, path planning, validation of positioning and mapgeneration, crowdsourcing and community building around data.
Planned approach and implementation
We schedule the project run over a two-year period. The first year is to develop and improve upon the already existing HD-map. While investigate the density of objects needed in the HD-map for positioning and map generation. Simultaneously we will start collecting sensor data from multiple laps around Rural Road in different weather conditions. Then crowdsourced maps will be created and with the help of new designed and geo-tagged anchor points validation of positioning and map generation will be done, most of the data distribution and collection will be done during year two.