Motivation and background
The development of self-driving vehicles is an important and challenging task. The vehicles' perception systems are often based on deep learning and training these models requires a huge amount of annotated data. Annotating data is time consuming and our teams at annotell have developed an annotation platform to make the annotation workflow efficient, including all aspects from UI/UX to integration of interactive deep learning algorithms.
The objective of this thesis project is to create training data for a situation awareness function and also train and validate a model predicting the chosen situation. Example of situations that can be of interest, bus stopped at a bus stop (in contrast to a parked bus), predicting target movement intent or some other situation that is useful for a planning function
We expect the students to:
Perform a thorough literature study and select one or two suitable situations that is useful for a planning function
Create annotations for training of an AI to detect the chosen situation
Train and validate the AI and possibly update the annotations and iterate the training
Suggest an effective way to annotate (automatically or manually) selected situations
We seek two students with an interest in deep learning. We also expect the students to have experience with python and preferably pytorch.
Project start: 15th Jan 2022
Duration: 20 weeks
Last date of application: 5th of Dec
Isak Hjortgren, firstname.lastname@example.org.
Tommy Johansson, email@example.com