Internship - Satellite health monitoring enhanced through transfer learning for constellation supervision à Toulouse

Airbus is a global leader in aeronautics, space and related services. In 2017, it generated revenues of € 67 billion and employed a workforce of around 130,000. Airbus offers the most comprehensive range of passenger airliners from 100 to more than 600 seats. Airbus is also a European leader providing tanker, combat, transport and mission aircraft, as well as Europe’s number one space enterprise and the world’s second largest space business. In helicopters, Airbus provides the most efficient civil and military rotorcraft solutions worldwide.

Our people work with passion and determination to make the world a more connected, safer and smarter place. Taking pride in our work, we draw on each other's expertise and experience to achieve excellence. Our diversity and teamwork culture propel us to accomplish the extraordinary - on the ground, in the sky and in space.

What if YOU starts your story with US?
We offer you to work in a world-leading company which is at the heart of a digital transformation, at the cutting edge of research and innovation.
We are looking for motivated students to join our teams at Airbus Commercial Aircraft or Airbus Defence & Space in Toulouse area.

You are familiar with technologies and you like to propose innovative solutions. In addition to the subjects that you will work on, you will have the opportunity to develop different skills such as teamwork, communication and project management.

Data analysis of sensor data (in orbit data, test data) has been implemented for 5 years within Space Systems. A portfolio of machine learning algorithms has been constituted and is being deployed on operational use cases. The novelty brought by satellite constellations lead us to consider transfer learning methods.
The objective is to allow the application of trained supervision algorithm to slightly different system. For example: how a degradation model of a battery learned on a satellite can be applied with same accuracy on another satellite whose battery is not operated the same way or experiences different environmental conditions.
The objective of the internship will be to identify, implement and test transfer learning method in the domain of time series analytics for satellite health monitoring and satellite testing.

The internships will start from March 2019 (date subject to some flexibility) and are for a duration of 4 to 6 months.

The intern will first understand existing methods in use in Space context, and understand the challenges they attempt to solve. Based on this current status, the intern will then explore method implementing transfer learning. Naive methods such as surrogate model of error between trained model and new dataset can be proposed, but it will be completed by a literature review. After selection of a set of methods the intern will then implement them in a Python environment, having access to distributed storage and computation environment delivered through Hortonworks or Palantir. Comparison of those methods will be performed on real dataset, either on ground testing or in orbit data. The internship will issue methodology and best practices to make use of transfer learning.

The internship will work in a team of 3 data scientists. He/She will have to exchange with end users in a multidisciplinary environment. He/She will be asked to participate to academic workshops organized by operational research community in Toulouse.

This job requires an awareness of any potential compliance risks and a commitment to act with integrity, as the foundation for the Company’s success, reputation and sustainable growth.

The candidate should be enrolled in a Bachelor/Master degree in mathematical engineering or applied mathematics field.

• Strong background in machine Learning and surrogate modelling.
• Strong basis in Python development.
• A first experience in pySpark and Hadoop and Spark environment or other distributed environment is a plus.
• Basis of software development environment is a plus (code versioning, ticketing solutions, testing Framework, continuous integration).

If you are interested in the challenge, then JOIN US!