Thesis - Manoeuvre planning and decision making in automated driving

Job-ID: T-TV Vehicle Solutions & Automated Driving-21894

Thesis - Manoeuvre planning and decision making in automated driving

______

Students  —  Thesis

EZ Chemnitz

This challenge awaits you:

Over the last few decades, automated driving (AD) has received a great deal of attention from both science and industry. Understanding the situation in road traffic is a basic prerequisite for carrying out collision-free, comfortable and rule-compliant automated driving manoeuvres. A driver constantly interprets how the perceived objects interact in the current situation, what properties and room for manoeuvre they have and will have. The interpretation largely depends on the driver's prior knowledge. This means that different objects in the driving scenarios can be related to each other, e.g. through relationships between vehicles, roads and lanes. For AD systems, a description of the object relationship can be realised with knowledge-based systems. This allows environmental situations to be described abstractly and semantically. Based on this, corresponding manoeuvres are planned, which must now be implemented by the automated vehicle in the movement planning.
The inference of manoeuvres based on the state representation to derive the optimal manoeuvre plan, taking into account constraints or temporal conditions, is a challenging task. The aim of this work is to implement an automated planning algorithm in which the necessary prior knowledge for the interpretation of a situation is first modelled (e.g. with predicate logic, PDDL, etc.). The problem can be solved using various C++ libraries, for example. At IAV, you have the opportunity to apply state-of-the-art methods based on symbolic state representation for the development of automated driving functions.

Your Tasks:

• Familiarisation with the IAV tool chain & development methodology
• Modelling and representation of the situation for a specific use case to solve an automated planning problem
• Conceptualisation, implementation and solution of the automated planning problem
• Implementation of test applications
• Analysis and documentation

The position can take place as part of a mandatory internship or a thesis.

Necessary Skills:

• Ongoing studies in computer science, mathematics, electrical engineering, information technology or a comparable degree programme
• Good knowledge and experience in programming with C/C++
• Experience in algorithms and software architecture
• Experience in the field of automated planning desirable
• Good knowledge of German or English
• Reliability, ability to work in a team, initiative and a high level of commitment, structured and independent way of working
  

We look forward to receiving your application including cover letter, CV, current performance record and relevant certificates.

You won’t just be working anywhere as a student at IAV. You’ll be right in the middle of it all. Real projects. Exciting future tasks. Completely integrated and side-by-side with IAV experts. Lots of responsibility and at the same time lots of freedom, combining university and work. The result is the best prospects for your professional development. And attractive compensation in accordance with our company wage agreements.
Diversity and equal opportunity are important to us. What matters to us is the individual, with his or her character and strengths.