Saskatoon TechWorks Forum

I'm going to be building a robot and am wondering if others would be interested if I were to host a workshop for others to build a similar robot. Here's what I have in mind:

• Arduino-based (for ease of programming and so you can use your knowledge to build other things)
• Two drive wheels and an omni-wheel for forward/backward/steering
• IR line detector sensors (allows following a line on the floor)
• IR collision avoidance sensors (allows detecting and avoiding nearby obstacles like walls)
• Ultrasonic range finder mounted on a servo motor (for more advanced distance detection and mapping)
• Hopefully some other handy things like push buttons, LEDs, an LCD, and a mini breadboard to make hacking easier.
• Plenty of room to add additional sensors, remote control, etc.
• Total cost < $100

As an idea what the workshops would entail, I'm thinking along these lines:

1. Assemble kits and install pre-written software. Have a working autonomous robot that drives around randomly, attempting to avoid obstacles.
2. Intro to programming. Write simple code to run the motors. Write a routine that accepts 'W', 'A', 'S', 'D' commands from the computer over serial to make a simple RC car.
3. Make a line follower. Read from the IR line detectors and write a simple algorithm to drive the motors so that it follows a line.
4. Make a collision avoidance bot, like that from session 1. Use the collision avoidance sensors and the ultrasonic range finder and make an algorithm to detect obstacles and steer away from them.
5. Make a sumo bot. Use IR line detectors to avoid leaving ring. Use ultrasonic range finder and collision detectors to find and attack opponent. Write a routine to turn the servo back and forth, scanning for the opponent.
6. Sumo contest. Modify your algorithm and code and try to outsmart your opponent. We'll all be using the same hardware, so we'll be on a level playing field. Make your modifications at home and we'll have a tournament at the hackerspace.

Other things you could do with the bot could include solving a maze (e.g. micro mouse), drawing shapes with a pen (like a plotter or turtle), or many other things you could think up. With some simple additions, you could control it wirelessly using IR or Bluetooth. You could add more servos and have a robot arm. Add a vacuum and make your own Roomba. Add simple encoders to the wheels and experiment with mapping rooms. Add lasers or IR diodes and sensors and have laser tag tank battles.

If you are interested, please let me know here. Also let me know if there are specific projects you'd eventually like to do with your bot and I'll do my best to see if the robot could be used for that.

I'm going to order the first one myself, build it, figure out what needs to be done to get it working, and once I have parts and a plan, I'll organize the group build and workshop.

So, if you want your own cheap robot platform on which to learn and hack, let me know and I'll get things started.

/signed /followed /subbed

#takemymoney

Just for reference, the parts we used for my nephew are:
http://www.robotshop.com/ca/en/dfrobot- ... duino.html
http://www.robotshop.com/ca/en/dfrobot- ... ld-2a.html
an Arduino (Leonardo or Uno... can't remember)
Plus three: http://www.digikey.ca/product-detail/en ... ND/1681159
and associated resistors.

Sounds fun. I should be in.

I'm in.

Tim (Odemia) also said he is interested.

I've already ordered parts for my initial attempt they should be here by the end of May.

I also spent this past weekend reading a couple of books on building various robots using Arduinos. Most of them were similar to what I had in mind and were using similar parts. I also found their code was similar to what I had planned to write, so that's wonderful. Seems like, aside from building a decent platform, it should be a simple endeavour.

I'll let you know when I get the parts and as I make progress building it.

If any others are interested in joining us, please let me know so that I can plan accordingly. Given the timeline for acquiring parts, I expect to hold the workshop in August.

You might already have me on the list from last Tuesday's meeting, but either way. Count me in! I wanted to do an arduino starter project anyway and this looks perfect.

Thanks Drew, I'll let you know as things progress.

Speaking of which, I've received basically everything except for the motors. Hopefully the motors will show up this week and I can start working on my build this weekend. I've already gotten the OK from Rylan to use the laser cutter to make up the frames. I spent a short while on Monday trying out the Arduino Mega with the IR reflective sensors and the ultrasonic range finder and they work quite well. The IR obstable avoidance sensors are a bit annoying to calibrate (most of the range of the potentiometer isn't usable, and what is usable is very slim), but they worked well once I tuned them. Getting simple, dirty code to talk to them was a cakewalk, but I know it will be slightly more complicated when I want them to all be talking while still driving.

I also read over the book Arduino Robotics, which covered many of the same robots I intend to build. If others are interested in reading it, I'd recommend it, but it's not necessary. It confirmed for me that what I intend to do is quite reasonable and actually fairly simple. I think this will work out very well for beginners to robotics.

I'll keep you up on my progress as I get into my initial build.

Sounds good. If you are working at the space and want any help working on your beta - let me know. It sounds like fun and it would be a good chance to see how the laser cutter works. I ordered the book.

Progress update. I'm on my third revision of the frame design and it's quite close. I want to move a few things to make it a bit smaller and add a few more mounting points, but I have something that is sturdy, simple to build, and allows lots of room to drill holes and add additional bits.

On the first revision I hooked up the motors and wrote some dirty code to drive back and forth. I noticed a few things. First, with the gear ratio I've chosen, it moves about 1 ft/s and it has plenty of torque. Also, the tires slip on the hub if you try to drive it too hard (such as if the tire gets stuck on something), so that's good; it helps prevent stalling the motor and burning it out. Lastly, the motors are very inconsistent. One turned significantly faster than the other, so it wandered to one side. We'll have to add in some calibration to get the motors are roughly the same speed if we want to do any kind of dead reckoning.

I had encoder wheels and IR photointerrupters to try out some simple wheel encoders, but I overlooked the size of the photointerrupter and ordered something that was way too small. It might still be an option to add on, but I'm not going to worry about including it in this kit.

I wanted to have the wheel axis going through the center of the robot so that it turned in place, but with the torque in the motors, it would easily lift either the front or back off the ground while changing direction. I tried putting the caster wheel ball bearing on both front and back, but this has several downsides: less room to mount sensors right where they matter most, low clearance for having a wheel lifted off the ground, and slop where it tilts forward/backward as it changes directions which interfers with the sensor height. Instead, I've gone back to a traditional 3-wheel design with the drive wheels in the rear and one caster in the front. This seems to work well and provides lots of room for line sensors at the front.

I expect to have the final chassis built this weekend and then start some programming.