Robotics in the Curriculum

All fools day it may have been … but not at the Open University headquarters today in Milton Keynes. The ‘Robotics in the Curriculum’ event was attended by a variety of people from the commercial and education worlds. Looking for ways to develop the use of robotics in education, there was plenty of experience to be shared. I was there showing off micromouse and the related schools competitions to  pretty well anyone who would stand still long enough to look interested…

The very professional looking stand was really that of Duncan Louttit of Swallow Systems. He is very active promoting schools micromouse competitions. These are introductory level event. There are two in particular that are popular. The F1 drag race requires an autonomous robot to perform a high-speed run along a 7m track, following a white line down the centre. This appears a simple, brute-force event but has a few subtleties to catch you out. The current speed record is 2.5 seconds and last years’s winners have apparently set themselves 2 seconds as their target. If you do some sums, this is quite a challenging target. Why so tricky? Well, the robot time is measured as it crosses the finishing line but it must not then overrun the 1.2m area beyond that or it incurs a penalty that will effectively disqualify it. Couple that with the need to steer the robot accurately at 4m/s or more and you have a competition that is easy enough for anyone to have a go while being tricky enough to require some real skills to win. The other popular event is the wall-follower contest. This takes place in the same maze as the  normal micromouse and has two classes – contact and non-contact. For the former, mice require physical contact with the walls to guide them around the maze. Non-contact mice may well find they bump into the walls but are designed to navigate the maze without having to touch a wall. This event attracts quite a few entries and produces some really interesting designs. It potential low-tech approach and ready availability of cheap materials makes if very attractive to the beginner or the hard-pressed school. A good design can be made for £10 with ample scope for development.

Also present were Derek Hall and Jim Chidley. Derek was pointing at Mouse X and reminding everyone that is is the current UK champion. Don’t worry, I have plans for that. Jim was showing off his kit mouse. This design is remarkably simple. It uses a pair of PicAxe processors and is a full maze-solving micromouse. There has been plenty of discussion about what a minimal maze-solver might require. Jim wasn’t satisfied with discussion so he went out and made one. It is an elegant design that anyone could build. Watch a video of it here:

Jim Chidley’s PICONE micromouse

I had a very interesting chat with Paul Foster of Microsoft who was demonstrating some of the huge variety of possibilities available with Microsoft Robotics Studio (MSRS). Having looked several times at this software suite, I learned more in ten minutes talking to Paul than I would have in ten hours staring at a monitor. For the kind of robotics involved in micromouse, MSRS has some limitations. Chief among which seems to be the reliance on some kind of Windows platform for the target system or good communications with a Windows host. However, Paul mentioned the mini-ITS and pico-ITX devices. The pico-ITX is essentially a PC on a board 100mmx75mm. This is almost the exact size of the average maze-solver and raises the interesting possibility of doing development work, including 3D simulation on your PC followed by implementation on an autonomous platform. Now, of course if the run-time of MSRS had a stripped-down version that executed in a simpler virtual machine – like JAVA or the Lego NXT – that would be interesting.

The First Lego league organiser was right next to us. That looks good – a world-wide competition that apparently attracts a couple of hundred entries form UK schools alone. You can’t beat a bit of corporate sponsorship to help these things along. We did agree though that it seems a real shame for there to be so many robot related activities and events that are al happening with little or no interaction. Essentially, they are competing for the attentions of hard-pressed teachers.

I have used some of the older Mindstorms kit and even though I found the programming interface to be a real pain to use, I can see why their kit is attractive. Out of the box, you get a bunch of hardware, most of which is familiar to the user and pretty intuitive to use. It is a rare individual who has never played with Lego. Coupled with that is a range of curriculum support material which is ready to use and, of course, the resources of a huge company to provide support. Teachers simply don’t have time to develop all these materials from scratch.  This is nothing new. Teachers have always relied on other people to produce at least the outline of a curriculum or scheme. A good textbook is often the core of many a teaching scheme and with good reason. Now, of course, the Goverment seems to want to mandate more and more of the content of the curriculum but with little evidence of understanding how that should be done. I think robotics in the curriculum will only really take off when there is a good range of relevant, ready-to-use material and a clear place for it in the school timetable. And that does not mean allowing the likes of Microsoft to dominate (sorry Paul).