Challenges

Deep learning for Earth observation (presented by EarthPulse)

The challenge will consist on training a neuronal network that classifies satellite images. It is thought as an introduction to Artificial Intelligence and computer vision for the Earth observation. The participants will have acess to the company’s library in order to ease the process a part from receiving all the needed knowledge in the preparation webinar that will be held by the CTO of EarthPulse..

Map matching for urban location (presented by Rokubun)

The GNSS signal, albeit specifically designed to be used outdoors, suffers from a high degradation of accuracy and availability in urban environments due to the presence of urban canyons (high buildings on the sides of the street). To solve for the loss of GNSS satellite visibility and increase of GNSS signal noise, additional sensors such as inertial measurement units (accelerometers, gyroscopes) or techniques (such as map-matching) can be used. The team will need access by creating a free account in the following services JASON (an on-cloud service to achieve high-accurate geolocation by uploading your GNSS campaign data) and MAPBOX (The Mapbox Map Matching API allows you to take recorded traces, such as the data from a GNSS unit or phone, and snap them to the OpenStreetMap road and path network).

Mission design software (presented by Sateliot)

Since the last 15 years a revolution in the space industry is on the go. Newspace approach has been taken by more and more players in the industry. Often these companies want to do things differently, but established methods, process and tools are not well prepared for the task. In this challenge we propose you to develop a mission design software/ toolkit for small satellites missions. It is up to you to determine which features do you want to include in it. You are free to use any existent code you have access to as long as the license allows it. There are multiple existing libraries and projects that can help you, for instance Orekit.

Smartphone application development showcasing the features of EGNSS and Copernicus (presented by EUSPA)

Develop and further investigate the possibilities of using an Android and/or iOS smartphone to increase the adoption of the EU free satellite services. We have more than 2 billion users with a Galileo compatible smartphone, so any of them has already the technical capability to use EGNSS and Copernicus for daily tasks.