Deep multi-object tracking for ground truth trajectory estimation
The aim of this project is to develop algorithms that will provide high-precision estimates of the trajectories of all dynamic objects in the vicinity of the host vehicle. The purpose is to obtain an efficient technique to extract estimates that can be viewed as ground truth, which is of utmost importance for the development and verification of both perception and control modules. The intended focus is on off-line techniques and on investigation of combinations of deep learning and conventional Bayesian methods, to enable extraction of as much information as possible from the data.
The project is expected to have an effect on the development and verification of environmental perception algorithms for self-driving cars. This will be done by designing tools and strategies to automatically obtain accurate estimates of vehicle properties and trajectories. From an academic perspective, the expected outcomes are: 1. a deep neural network for off-line multi-object trajectory estimation; 2. a deep neural network for off-line multi-sensor track-to-track fusion.
The work will be carried out by Yuxuan Xia, a PhD student at Chalmers University of Technology and two master students, under the supervision of Professor Lennart Svensson.