High-Quality Assumed Gaussian Filtering for Nonlinear Systems

Research topic/area

Probabilistic machine learning, State estimation

Type of thesis

Bachelor / Master

Start time

-

Application deadline

30.06.2026

Duration of the thesis

6 months

Description

We consider the general state estimation problem for a discrete-time stochastic nonlinear dynamic system with noisy measurements. Specifically, we focus on Gaussian filters that approximate the true, in general complex, state Probability Density Function (PDF) by explicitly optimizing the shape of a Gaussian distribution after each processing step. This class of filters is known as Gaussian Assumed Density Filters (GADFs).

At ISAS, we have developed a broad spectrum of GADFs, ranging from Linear Regression Kalman Filters (LRKFs) and Progressive Gaussian Filters (PGFs) to Inverse Gaussian Process (IGP) interpolation methods. Compared to the former approaches, these IGP interpolation filtering techniques exhibit significantly improved performance, driven by deterministic sampling of the joint density of measurements and states, as well as the effective exploitation of these samples using machine-learning–based methods.

The goal of this thesis is to build upon the core ideas of these existing methods to design a novel algorithm and evaluate its performance against state-of-the-art techniques. The work will roughly comprise the following tasks:
● Literature research on nonlinear filtering methods,
● Familiarization with the Julia programming language,
● Development and design of a novel algorithm,
● Implementation and integration of the methods in Julia,
● Comparison with other state-of-the-art-methods.

Requirement

Requirements for students

• Strong self-motivation, reliability, and critical mind are expected.

Faculty departments

• Engineering sciences
  Electrical engineering & information technologies
  Geodesy & geoinformatics
  Informatics
  Mechanical engineering
  Mechatronics & information technologies
  Mobility and Infrastructure
  Mechanical Engineering
  Mobility Systems Engineering and Management
  Remote Sensing and Geoinformatics
  Information System Engineering and Management
  Computer Science
  Electrical Engineering and Information Technology
  Mechatronics and Information Technology
  Medical technology
• Natural sciences and Technology
  Mathematics
  Physics
  Mathematics in Technology
  Computational and Data Sience
  Physics
  Techno-Mathe­matics
• Economic & law sciences
  Information Engineering
  Business management

Supervision

Title, first name, last name

Jiachen Zhou

Organizational unit

Institut für Anthropomatik und Robotik (IAR) - Intelligent Sensor-Actuator-Systems (ISAS)

Email address

jiachen.zhou@kit.edu

Link to personal homepage/personal page

Website

Application via email

Application documents

• Grade transcript
• Certificate of enrollment

E-Mail Address for application
Senden Sie die oben genannten Bewerbungsunterlagen bitte per Mail an jiachen.zhou@kit.edu


Back