0000004197 19S 6SWS PR Praktikum: Smart Self-Adaptive Cyber-Physical Systems - A road towards Autonomous Systems (IN0012, IN2106, IN4257)   Hilfe Logo

LV - Detailansicht

Wichtigste Meldungen anzeigenMeldungsfenster schließen
Allgemeine Angaben
Praktikum: Smart Self-Adaptive Cyber-Physical Systems - A road towards Autonomous Systems (IN0012, IN2106, IN4257) 
Sommersemester 2019
... alle LV-Personen
Informatik 4 - Lehrstuhl für Software & Systems Engineering (Prof. Pretschner)
Zuordnungen: 1 
Angaben zur Abhaltung
In the recent years, the fast growth of cost-effective embedded systems with continuously increasing computation power, have created a solid foundation for emergence and expansion of the omnipresent Cyber-Physical Systems (CPSs), across different domains with increasing socio-economic influence. Modern CPSs need to be able to operate in constantly changing, uncertain and unanticipated environments, and yet remain efficient and reliable. Namely, these systems should be capable of learning, dynamically and automatically reconfigure themselves, and be able to cooperate with other CPS. In a nutshell: exhibit human-like, smart capabilities in an autonomic manner. Therefore, not only that the importance of self-adaptivity as a system's feature increases, but it becomes a fundamental approach for the systems to continue meeting their functional specifications, fulfilling their business objectives while perceiving the performance–despite all the run-time changes that the system encounters.
• Applicants can be Bachelor's or Master's students
• Good Python and C/C++ skills
• Preferably previous knowledge or experience with ROS
• Motivated, enthusiastic students, eager to broaden their knowledge and solve hard challenges
• Comfortable and willing to work in teams
• Interest in software and systems engineering, embedded systems, robotics
• Interest in playing with real hardware
In this practical course, we will aim to build different logic strategies for self-adaptation of smart CPSs. Our participant teams will compete to find the best approach for building a global semi-decentralized map in a cooperative way, and best ways to aggregate knowledge from two robots as two different sources of information. The final goal would be to assess if the ability to learn and store the past experiences of the systems as knowledge, that is later considered during the planning and decision-making process, improves the performance of the system. The students will use our in-house built ROS-based multi-robot system as a starting point for their implementations.

In the current implementation, we have two robots that keep their own maps of the static and the dynamic context. The static context is the room and other static elements of the room, where the robots are being deployed. The dynamic context is: first, the way the robots see each other, when they are within the range of their sensors; and second, the randomly appearing tasks for the robots in the room. The current maps are only depicting the current state of the context, without storing any past experiences as knowledge, that might potentially help for better future path planning.

There will be three phases in this practical course. Participants will form small teams that solve the phases collaboratively.
1. In the first phase, the teams will focus on getting familiar with the ROS-based multi-robot system, understand the problem that we are trying to solve, and sketch the first ideas for implementation which will be used in the second phase
2. In the second phase, the teams will implement the envisioned strategies/algorithms from the previous place, and make the first evaluation on a simulated ROS-based multi-robot system
3. In the last phase, the teams will make their final evaluation on real hardware, in particular, using two TurtleBot3 Burger (https://www.turtlebot.com/, https://github.com/ROBOTIS-GIT/turtlebot3)
Für die Anmeldung zur Teilnahme müssen Sie sich in TUMonline als Studierende*r identifizieren.
1. Kephart, Jeffrey O., and David M. Chess. "The vision of autonomic computing." Computer 1 (2003): 41-50.
2. Weyns, Danny. "Engineering Self-Adaptive Software Systems–An Organized Tour." 2018 IEEE 3rd International Workshops on Foundations and Applications of Self* Systems (FAS* W). IEEE, 2018.
3. Philippe, Lalanda, Julie A. McCann, and Ada Diaconescu. "Autonomic Computing Principles Design and Implementation." (2013).
4. O'Kane, Jason M. "A gentle introduction to ROS." (2014).
5. Broy, Manfred, María Victoria Cengarle, and Eva Geisberger. "Cyber-physical systems: imminent challenges." Monterey workshop. Springer, Berlin, Heidelberg, 2012.
Online Unterlagen
E-Learning Kurs (Moodle)
preliminary meeting: 04.02.2019, 11:00h, room 01.09.014