facebook tracking

Master Thesis - Statistical modelling for sensor fusion in Bayesian framework

Background

Autonomous driving systems and advanced driver assistance systems require a description of their surrounding environment. This description consists of both static and dynamic objects of the surrounding environment. The description of these objects is usually provided by using sensor detections and fusing detections from several sensors to estimate different attributes of the objects in the surrounding environment.

Historically, sensor detections could be modeled by continuous Gaussain variables where the variance would reflect the uncertainty of a sensor detection. Nowadays, with more capabilities added to modern sensors, some can provide us with categorical information such as the class of a detection as well as continuous values.

In the sensor fusion context this implies that two different sensors could convey information regarding the same physical entity in two different ways. For example, while one sensor can give us a detection regarding the length and width of an object, another sensor can give us class probabilities for that object. In this new context there is a need for new models which can incorporate both types of sensor information into a single model of an object and consequently a single model for multiple objects. In addition, we need new methods to update such a model upon receiving either (or possibly both) type of information from a sensor.

 

Project description

In this master thesis you will focus on:

- Modelling the probability density of an object (possibly multiple objects) such that it can be updated by both categorical class measurements and continuous size/shape measurements.

- Using camera detections to update the model parameters.

- Evaluate the estimated model.

 

Requirements and qualifications

We are looking for two students, preferably with good knowledge of:

- Statistical modeling

- Bayesian filtering

- Knowledge of C++ programming is a plus.

Further information and contacts

Final application date: December 15th 2019. Please send in individual applications. If you wish to partner with someone, simply note that in your application.

Planned start: Beginning of 2020, with some flexibility.

Duration: 30 ECTS

For questions regarding the project, please Contact Maryam Fatemi (maryam.fatemi@zenuity.com, 0731258048)

 

Or, know someone who would be a perfect fit? Let them know!

Gothenburg, Sweden

Lindholmspiren 2
417 56 Göteborg Directions

Making safe and intelligent mobility real.

At Zenuity, we lead the global movement of crafting tomorrow's mobility with the software platform of choice. Our mission is to “Make safe and intelligent mobility real, for everyone, everywhere”. This statement marks our conviction and dedication to bring autonomous driving out on the streets for real and is at the center of everything we do.

We could not dream of achieving this without our great teams of very talented people. We are on this journey together and our agile way of working is reflected throughout our entire organization; it is part of our culture and how we work, develop and grow together.

Teamtailor

Applicant tracking system by Teamtailor