Year 10 Computer Science students enjoyed a day of hands-on physical computing when Queen's welcomed Aberystwyth University's Teaching Fellow, Martin Nelmes, who delivered a workshop on ‘Robotics and Artificial Intelligence’.
The day started by considering what 'intelligence' might mean and the difficult question of how we might rank intelligence - can we compare the intelligence of a tree with that of a military drone? The girls also watched footage of some robotics research being carried out at the University involving humanoid robot, iCub, who is teaching itself to perform simple tasks using heuristic software designed to replicate developmental learning processes. However, footage of a 2 year old human performing a more difficult task in a fraction of the time highlighted how far we still are from creating intelligent learning robots with the capacity and adaptability of humans - despite what Hollywood and science fiction might tell us.
The main focus of the day was getting the girls to programming their own Lego Mindstorms robots. Their first task was to build the robot including motors and sensors to allow it to interact with its environment. Next the girls worked in small teams to program their robot to follow a black line on the floor. Once this had been mastered they were ready to use their problem solving skills to manoeuvre their robot around an obstacle course. The girls soon learned that there is more to robotics than programming. Just because they had written the right program did not mean that their robot would behave as expected as difference between the performance of the motors and sensors could and did produce interesting, unexpected results. Martin described this as the point where programming and engineering come together - meaning that testing and refinement of their code was essential if each robot was to behave as expected.
The girls enjoyed the day and learned a lot about the amount of work, testing and perseverance needed to program a robot. A huge well done to Eleanor Hulse and Verity Lord who were the first to successfully get their robot across the obstacle course finish line.

Post date: 4th July 2017