Multi-object tracking for automotive systems

Om kursen

The material includes the following content:
An introduction to single-object tracking including Kalman filters, probabilistic data association and nearest neighbor filtering.
  • An overview of vector-based multi-object tracking, including joint probabilistic data association (JPDA), global nearest neighbor (GNN) and multi-hypothesis tracking (MHT).
  • An description of the main components related to random finite sets (RFSs), including RFS densities, RFS models, PHD filters and performance metrics on RFSs.
  • An introduction to the state-of-the-art algorithms that make use of conjugate priors, with an emphasis on Poisson multi-Bernoulli mixture (PMBM) filters and labelled multi-Bernoulli filters.

Organization and examination
See course homepage on edX for details.

Prerequisites
We recommend that you have taken a course that teaches Kalman filtering, such as SSY345 - Sensor fusion and nonlinear filtering.

Mer information

Lennart Svensson Telephone: 031-772 1777 E-mail: lennart.svensson@chalmers.se

Föreläsare

Lennart Svensson E-mail: lennart.svensson@chalmers.se