Part 2/10
In order to make production lines flexible, future-oriented and economically attractive in today’s world, the automation of work steps is essential. The demand for robots is constantly increasing. It is assumed that the industrial use of robots will double in the next five years. This development brings with it new challenges in dealing with robots.
Programming robots requires a lot of know-how and trained staff, as commissioning is a time-consuming process. Even after the robot has been set up, errors can occur in production if production processes are not sufficiently optimized. If the product changes during production, further effort is required. In this context, learning robots are a great relief.
There are already AI-based industrial approaches for recognizing errors in production, learning from them and avoiding them in the future. However, they are not yet ready for everyday use and are still at a research level. Complex tasks such as optimizing a motion path or finding and grabbing an object in a chaotic environment can take several months to learn.
Looking at self-learning robots from industry, there is great potential in the social sector. Robotic arms are increasingly being used there to support physically impaired people. Due to high safety requirements, there has only been a small amount of research and a small range of solutions on the market to date. The difficulty lies in the fact that the tasks performed in everyday life can vary greatly. While the environmental conditions, the objects that can be manipulated and the interactions with humans are firmly defined in a production environment and hardly vary, these are constantly changing in everyday life. Traditional systems and AI approaches can hardly be used today. In the future, we will need flexible systems that can be constantly expanded.
This is where our current research project HIRAC can help. Together with the University of Applied Sciences Offenburg, we want to develop hardware-independent software that can, for example, control different robot arms from different manufactures.
kununu