Electric Imp as a core for educational tools

I recently visited a local Hackfest where many 2nd-3rd year engineering students were spending the day designing and building robots to solve some simple tasks. With a limited amount of time, none of the teams managed to complete the entire challenge. Quite a few of them got bogged down in the radio side, never quite getting the wireless control of their robots working. Additionally, students often had no idea what their robots were thinking once they deployed them in the arena. That made debugging like groping in the darkness. Consequently, many had little to show for their endeavours at the final demo.

As an observer, I was asked to offer some advice to the teams on approaches, teamwork, testing etc, but sadly not on the technology they should use. All the robots used had Arduino brainboards with various peripherals. Wireless operation was a necessary part of the challenge, but it wasn’t the key task. Nevertheless, it represented a huge proportion of the sweat given to the project. I know Electric Imp doesn’t yet have the eco-system that Arduino has, but it strikes me that it would be great for these sorts of projects. Has there been any thought given to educational tools based around Electric Imp? I can see a great deal of the roadblocks have already been solved, it’s really just making surely peripherals can be integrated with a minimum of fuss. I spoke to some of the students recently and most, if not all, mentioned proficiency/competence in Java or Javascript. I could see a suite of imp based products being winning over teachers as they allow students to focus on what’s important, avoiding all the dead-time involved with constantly having to plug the units back into a notebook for debugging/reprogramming etc.

great idea! I do see a problem; who would pay for it?

I think imp-based projects can get a higher level much more quickly than Arduino and the imp IDE is much better than the Arduino one with the exception of not having library support. As you said, if you have Javascript experience you can do a lot with the Agent and Device with just a little bit of a squirrel lesson.

TBH Though I love Imp, I don’t think in its current iteration it is the right platform for robotics (that need very low-latency control for example), if they are already into Arduino the “Yún” add’s all the radio stuff and Linux to the board.

http://arduino.cc/en/Products/ArduinoYUN

I built a bump & backoff robot using the imp as fun project with my 6yr old. What she liked best was the harwared design & build. There’s a wee video at the end of this post http://industrialinternet.co.uk/iot-2/making-a-robot-with-my-6yr-old/

Love it! I may just have to build something like that myself with my daughter

I still think something open-source with local network control would be best for experimental robotics especially if I need to do the digital equivalent of shouting STOP! :slight_smile:

We not-infrequently support university hackathons. I have consistently seen teams get very far during the course of a 1.5-2 day hackathon, even with zero experience with Electric Imp / Squirrel.

The value is exactly what @coverdriven mentioned - getting radios working as quickly as possible. I encourage people to follow the Getting Started Guide, and basically continue to use query parameters as a way of passing messages to the device (often through and app or webpage). It is far from the best implementation, but it works phenomenally for a weekend hack.

As for robotics - @back_ache hit the nail on the head. WiFi and HTTP requests can work great, but an ideal solution is local networking, whether bluetooth, wifi, or something else.