Cognitive robotics is a driving force for the development of AI systems that can perform complex tasks autonomously and competently. With the help of knowledge and models, these AI systems are able to interpret their environment. For example, such robots can understand vaguely formulated instructions by correctly interpreting the context. They can perform complex everyday activities such as setting the table or cleaning up. Cognitive robotics allows them to learn by watching, reading and experiencing, and it lets them plan and explain their actions.
The skills to understand and develop such AI systems can best be taught using open robot control systems that contain a broad spectrum of AI technologies. The goal of IntEL4CoRo is to create and operate such a learning environment for cognitive robotics that enables competence-oriented teaching.
The learning environment will follow an immersive approach, containing robot control systems and photorealistic, physics-based simulation environments for experimentation. The control systems and simulation environments help to illustrate the concepts taught in the courses. They also facilitate the generation of suitable tasks for all related subject areas – students can complete these tasks by programming new components or correcting faulty parts of the teaching systems. In addition, the learning environment is designed for the work on final theses, team-based software development projects and robot competitions.
Configurations of the teaching system that can be transferred to real robots ensure practical relevance. Additional simulation scenarios in which autonomous robots assist and cooperate with humans provide a basis for discussion and the framework for ethical, legal and social issues (ELSI) of AI and cognitive robotics.
To encourage widespread use and collaboration, IntEL4CoRo has been implemented as an open-source and OER (Open Educational Resource) project.