Will robots beat the human world champion football in 2050?

When I talk to journalists about robots, the most frequently asked question is: will the football robots beat the human world champion football in 2050? My answer is always: I don’t know, but the gradient is definitely pointing in the right direction.. For example, compare an MSL RoboCup match from 2005 with one from recent years. In 2005, most robots are stationary on the then still small playing field, both due to hardware problems and due to inadequate software. In recent years, full matches are played on a much bigger field with full teams of five players and the game already looks a lot like football with many passes, even in the free space of the field.In this story I want to focus on the answer to this question. By looking at evolutionary and revolutionary developments in recent decades. And by trying to identify steps that will be necessary to achieve the goal set for 2050. Where is the real challenge and where does (human) engineering make a difference in that development?