See TUMonline

The lecture aims to put these instrumental ways of thinking about autonomous vehicle systems into a broader perspective and to take a reflexive and critical look at the relationship between hardware, software and the final application. After participating in the module, students will be able to:
- Recall the characteristics of a useful autonomous vehicle system.
- Identify which technologies (software, hardware) can be used to develop new autonomous vehicle systems.
- Apply the approaches to the design of a useful autonomous vehicle system.
- Characterize an operational design domain (ODD) for an autonomous vehicle system
- Evaluate different software architectures of autonomous vehicle systems
- Develop concepts of autonomous vehicle systems

We are currently developing autonomous vehicle systems and robots with the aim of achieving a higher level of automation and being able to perform certain tasks without humans. Numerous questions arise during this development: Which (partial) tasks must my system actually be able to solve autonomously in the end? What challenges do autonomous systems face in your application? Which software modules are necessary to achieve the desired functions? What level of assurance does my system need in order to be used in the final application? This module goes into thematic depth on the technical content of autonomous vehicle systems with the goal of identifying which challenges can arise in the development of new, autonomous vehicle systems and which technologies can provide a corresponding solution approach.
The following course contents are planned:
I. Introduction
- Overview of the course content
- Overview of autonomous vehicle systems
- Definition and description of the use of the autonomous system in an ODD (Operational Design Domain)

II. characteristics of a useful autonomous vehicle system
- Relevance to real world problems
- Efficient and effective solution
- Feasibility of deployment and scalability
- Compliance with ethical and safety standards
- Arguments for hardware selection (sensors, actuators, computational platforms).

III. approaches for the design of useful autonomous vehicle systems
- User-centered design
- Problem-oriented design
- Context-oriented design
- Software-oriented design 1: Selection of an appropriate software architecture: from classical perception-planning-control architecture to end-to-end approaches and Autonomy 2.0
- Software-oriented design 2: Selection criteria for autonomous driving functions
- Software-oriented design 2

IV. Case studies on useful autonomous vehicle systems
- Application areas of autonomous vehicle systems and their fundamental needs and challenges.
- Passenger transportation: autonomous vehicles on our daily roads
- Healthcare: autonomous vehicles for medical transportation
- Agriculture: autonomous vehicles for farming and crop management
- Civil protection: autonomous vehicles for search and rescue missions
- Environment: autonomous vehicles for monitoring and cleaning up oil spills

V. Design challenges and solutions
- Identifying and addressing risks of autonomous vehicle systems.
- Technical challenges (e.g., sensor accuracy, robustness of algorithms)
- Societal challenges (e.g., public trust, ethical considerations)
- Economic challenges (e.g., cost of deployment, scalability).

VI. future developments in the development of useful autonomous vehicles.
- Integration of new technologies (e.g., 5G networks, blockchain).
- Interdisciplinary collaboration (e.g., engineering, sociology, psychology)
- Development of standards and regulations for deployment

We recommend prior attendance of a module that covers the basics of autonomous systems or robotics such as "Software Development for Autonomous Driving" or "Autonomous Systems".

The teaching methods in the module are divided into two parts. In the first part of each lecture, the teaching content is conveyed by means of lectures and presentations (Power Point). In the process, more complex issues are derived and vividly illustrated using a tablet PC. Within the lecture, certain topics are deepened by live demonstrations of simulations. For each lecture 1-2 scientific papers (6-8 pages) will be provided in preparation. This paper has to be prepared before the lecture in order to enable a deeper discussion in the second part of the lecture.
In the second part, jointly prepared questions will be answered and their answers discussed. Students are expected to actively participate in class discussions, which requires that they are familiar with the papers provided. The format of the course is mostly interactive in the second part.

The module examination takes the form of a written, scientific paper in which a scientific, application-oriented question on the topic of autonomous vehicle systems is dealt with using the scientific methods of engineering (qualitative or quantitative). The students can propose their own questions or choose from a wide range of proposed topics. For this purpose, a research question is created (e.g. "What is a useful autonomous system"), a literature search is conducted, the research question is answered methodically, and a concluding discussion on the quality of the results is held.

Pendleton et. al, Perception, Planning, Control, and Coordination for Autonomous Vehicles, Machines 2017, 5(1), 6; https://doi.org/10.3390/machines5010006

Ashesh Jain, Luca Del Pero, Hugo Grimmett, Peter Ondruska Autonomy 2.0: Why is self-driving always 5 years away? https://arxiv.org/abs/2107.08142

A. Faisal, T. Yigitcanlar, M. Kamruzzaman, and G. Currie, “Understanding autonomous vehicles: A systematic literature review on capability, impact, planning and policy,” JTLU, vol. 12, no. 1, 2019, doi: 10.5198/jtlu.2019.1405.