See TUMonline
Note: Registration for the lecture via TUMonline

After participating in the module, students will have an in-depth insight into the movement and behavior planning of autonomous vehicles. Students will be able to understand the basic principles of motion planning and generate dynamic models of different autonomous vehicles. After participating in the module, students will have an overview of the most important methods of behavior planning as well as global and local motion planning. At the same time, students will have understood the mathematical and technical fundamentals. Furthermore, students will be able to continuously reflect on the methods and knowledge they have learned with regard to sustainable ecological, social and economic development. Students are able to identify the most important challenges in movement planning and find solutions for them and approaches in movement planning. In addition, students are able to analyze real measurement data and develop suitable algorithms to solve these problems. As the module is open to different disciplines and raises transdisciplinary issues, students are able to understand the language of others, justify their own decisions and convince others with arguments in teams from their own discipline, but also in interdisciplinary teams.

The module "Autonomous Vehicles: Motion Planning & Decision Making" offers an in-depth insight into the methods and algorithms of motion planning and decision making for autonomous vehicles. We show how state-of-the-art technologies and algorithms enable the dynamic and agile movement of autonomous vehicles. Starting with the basics of motion planning and dynamics, you will be introduced step by step to methods of decision making as well as global and local motion planning. The lecture covers everything from graph-based methods to game-theoretic approaches to reinforcement learning to provide a comprehensive understanding. With the content of the lecture, you will be able to solve complex planning problems and develop algorithms that can make intelligent decisions in uncertain environments. Practical exercises and realistic scenarios will reinforce the theoretical knowledge.

The lecture "Fundamentals of Autonomous Vehicles" is recommended as a prerequisite, as it teaches the most important basics for understanding the motion planning of autonomous vehicles.

In the lecture, the course content is conveyed by means of a lecture and presentation (Power Point). More complex issues are derived and illustrated using a tablet PC. During the lecture, explicit questions are asked that require students to transfer their knowledge and where students are given the opportunity to speak up and discuss a possible solution. This is intended to deepen the challenging tasks of autonomous driving and enable the transfer from theory (mathematics) to practice (software). The lecture also explains simple situational examples that have to be mastered by autonomous vehicles. These example tasks can be actively solved by the students. These examples are primarily in the area of road vehicles (e.g. road intersection in the city center), whereby the students are subsequently able to analyze and evaluate further problems of other autonomous systems (e.g. robots in agriculture).

Each lecture unit is supplemented by a subsequent exercise unit. The exercise relates thematically to the topic presented in the corresponding lecture and thus deepens the content of the lecture. The exercise consists of calculation tasks (e.g. calculation of a trajectory), identification tasks (analysis of diagrams, e.g. transfer of tire forces), design tasks (e.g. which components are needed for decision-making in an autonomous vehicle) and identification tasks (e.g. definition of challenges and problems in certain driving situations). The tasks are worked on and solved together in the exercise and then discussed with the students. A correct and detailed solution is provided in writing and then made available to the students on Moodle.
A weekly consultation hour is offered to answer questions about the individual appointments and homework, which can be attended in person or online (appointment announced via Moodle).

The module examination takes the form of a written exam (duration 90 min, permitted aids: calculator). The basics of motion planning for autonomous vehicles are tested using short questions. By means of comprehension and transfer questions, the participants show, for example, that they have understood the individual algorithms of motion and behavior planning, can analyze real measurement data and can analyze the behavior of autonomous vehicles.

Pendleton et. al, Perception, Planning, Control, and Coordination for Autonomous Vehicles, Machines 2017, 5(1), 6; https://doi.org/10.3390/machines5010006
Latombe, J.-C. (1991). Robot Motion Planning. Springer US. https://doi.org/10.1007/978-1-4615-4022-9
Mobile Robot: Motion Control and Path Planning. (2023). In A. T. Azar, I. Kasim Ibraheem, & A. Jaleel Humaidi (Eds.), Studies in Computational Intelligence. Springer International Publishing. https://doi.org/10.1007/978-3-031-26564-8