Master thesis project, 30 hp: Diffusion models for generative modelling of a posteriori probability measures in target tracking

Ready to find your path? Try new things. Share experiences. And discover what you truly enjoy.

Background

In order to provide an air-situational picture, a surveillance radar emits energy which is reflected by targets of interest as well as other objects in a surveillance volume, thereby giving rise to unlabeled sets of detections of unknown origin. Through repeated measurement of the surveillance volume and the use of target-tracking algorithms, with a foundation in Bayesian estimation, a situational picture emerges in the form of an a posteriori probability distribution describing the air-space. Fast identification and detailed estimates of target states - which is of great importance for overall system performance - relies on having accurate target motion models and being able to perform filtering with respect to these models. Unfortunately, models with higher accuracy also tend to be increasingly non-linear, resulting in the Bayesian filtering problem becoming intractably hard within the computational budget of the application. Alleviating this challenge is of high interest. Given the recent success of diffusion models for conditional image generation, it is believed that a similar generative paradigm could help alleviate the challenges of Bayesian filtering in target-tracking.

Project description

The goal of this project is to explore and evaluate the viability of diffusion models as conditional generative models for a posteriori probability distributions describing a single target trajectory. This involves training a neural stochastic differential equation to transform a conditional deep latent state into a target trajectory. It is of particular interest to tackle challenges in classical Bayesian filtering, namely non-linear state propagation as well as multimodalities in the a posteriori state distribution.

Your profile

You are in the end of your technical master's education in Engineering Physics, Engineering Mathematics, or similar, with an interest for advanced mathematics and numerical methods. Courses in Stochastic analysis and Bayesian statistics are meriting, as well as practical experience of deep learning. The degree project is suitable for one or two master students.

This position requires that you pass a security vetting based on the current regulations around/of security protection. For positions requiring security clearance additional obligations on citizenship may apply.

What you will be part of

Behind our innovations stand the people who make them possible. Brave pioneers and curious minds. Everyday heroes and inventive troubleshooters. Those who share deep knowledge and those who explore sky-high. And everyone in between.  ​

Joining us means making an impact together, contributing in our own unique ways. From crafting complex code and building impressive defence and security solutions to simply sharing a coffee with a colleague, every action counts. We encourage you to take on challenges, to create smart inventions and grow in our friendly and tech-savvy workspace. We have a solid mission to keep people and society safe.

Saab is a leading defence and security company with an enduring mission, to help nations keep their people and society safe. Empowered by its 22,000 talented people, Saab constantly pushes the boundaries of technology to create a safer and more sustainable world.

Saab designs, manufactures and maintains advanced systems in aeronautics, weapons, command and control, sensors and underwater systems. Saab is headquartered in Sweden. It has major operations all over the world and is part of the domestic defence capability of several nations. Read more about us here

Kindly observe that this is an ongoing recruitment process and that the position might be filled before the closing date of the advertisement. You can send us your application in either Swedish or English. If you have any questions about the job, don´t hesitate to contact:

Karl Hammar, supervisor +46 102167331

Jimmy Aronsson, supervisor +46 102788650

Benjamin Svedung Wettervik, supervisor +46 102161448

Per Gustavsson, recruiting manager +46 734379594