Case Study of Building a Robotic Stage Coach
by John Kendall, July, 2006


First Session, "Stagecoach I"

On Thursday and Friday, Shomar and Khristian finalized plans for making their "LEGO-mation" stop-action movie with a Western theme. On Friday, we decided that the easiest way to incorporate a robotic vehicle was to make it into a stagecoach, with the sheriff and the deputy riding the stagecoach, and the two bad guys riding their horses, to rob the gold in the stagecoach.

I demonstrated the 3-in-1 Bot from the Jeff Elliott book (10 Cool Lego Mindstorms: Robotics Invention System 2 Projects), which doesn't have the look of the stagecoach because it only has three wheels. I looked through the three Mindstorms books I sent off for and the original wordless manual, which was bundled with the Lego kit. I debated back and forth Friday afternoon at home in the backyard about which model would work best (we want six toy horses in front and the sheriff and deputy riding on top), but also which one I could build with the most confidence.

The books by Joe Nagata (Lego Mindstorms Idea Book) and Dave Baum (Definitive Guide to Lego Mindstorms) both helped me to understand more of the programming involved and more of the insight into the physics involved (e.g. Baum's notes on pages 88-95, which explains if a wheeled vehicle is traveling straight ahead, if you send more power to the right front wheel, or otherwise acknowledge that the two sets of wheels, when turning, are making larger and smaller circles, then figuring out how to program it makes more sense).

I started trying to create a merger of Baum's "Tribot" (chapter 11) and "Onebot" (chapter 12), but within 5-10 minutes was unsure of the mechanics of it, b/c he did not provide detailed enough step-by-step instructions. So I fell back to plan B, and resorted to the Elliott book, which successfully guided me through my first construction, with its meticulous, detailed illustrations and instructions. I went to Chapter 4 for the "F1 Racer", hoping that with a two-axle chassis, I could modify the front bumper and not build the "spoiler" at the rear of the vehicle, but instead build on more of a stagecoach structure, such as long straight beams in front to attach the six horses.

I started around 10:00 p.m. and finally went to bed around 4:00 a.m., although I took several long breaks of reading the Sunday New York Times, taking a bath, etc., so it was perhaps 3 or 4 sessions of 45 min. to an hour by the time I stopped for the evening (morning). Several sections of time were also spent organizing the 700 pieces of Legos into more common-sensical Ziploc bags (image 12) and the 12 little circular Glad containers in the muffin tin (image 13). Although this slowed things down, it was ultimately less frustrating to be able to find the right piece more smoothly, and would speed any future constructions.
image 12 of first piximage 13 of first pix

So the first two building sessions of the F1 Racer chassis went fine, slow and tedious, but progressing (see images 01 through 06, of Stagecoach I).

image 01 of first piximage 02 of first pix
image 03 of first piximage 04 of first pix
image 05 of first piximage 06 of first pix

I actually built the bumper, thinking I might be able to use it in a modified form (image 07) for the stagecoach. I thought it would be good to have the "rack and pinion" steering of the front wheels, but it turns out that also makes them less sturdy and more troublesome, at least when I tried to adapt the model from an F1 racing car into an old-fashioned stagecoach.

image 07 of first piximage 08 of first pix

So the last several sessions in the wee hours of the morning went largely downhill, as I committed too much impulse design (a fancy phrase for using traditional Lego bricks piled on top of each other) and didn't reinforce the chassis with enough reliable support beams, connecting pegs, etc. For every three pieces I added, I had to put two back on, constantly reverse engineering as I saw the error of my ways.

I did come up with a functioning stagecoach of sorts (see images 09, 10 and 11). I used Baum's procedures for moving straight and a simple turn, although I had to use my capped pen to push the "run" button, deep inside the stagecoach (note for second version of stagecoach: build it so the brick is on top, with just one layer of beam reinforcement, so the kids can push the "run" button more easily).

image 09 of first piximage 11 of first pix
image 10 of first pix

Both axles lacked support, and either the wheels would come off or other problems occurred. I called it quits, but remembered one of my students quoting Thomas Edison: "I did not fail to invent the light bulb a thousand times. I learned how NOT to build it 999 times" or something like that. I now have a much better sense of what I should do, although I have to decide whether to still try to modify what I have already built (rear wheels which power the vehicle with one motor are reliable. I could perhaps replace the front wheels with one longer axle and not worry about the steering), or else take it apart, go back to the "3-in-1 Bot" and put a second axle with two wheels on it, to look more like a normal stagecoach.

I also found several large set Lego pieces from the Space Shuttle kit, which might make building the top of the stagecoach easier. Perhaps I give myself a 30-minute time limit or so, replace the front wheels with a single axle and try a simple re-finishing and see how sturdy or flimsy it is. I also bought six toy horses at A. C. Moore, so I can practice with their weight and logistics of turning.

Second Session, "Stagecoach II"

I am now tired of physical manipulation of all the Lego parts and achy from thinking so hard, but feeling very virtuous and satisfied, a direct opposite from ending my last session 14 hours ago. Sitting next to me as I type this is a very recognizable, sturdy and functioning Lego stagecoach, which has successful made several trips around the kitchen floor and a turn or two, without mishap or falling apart.

I did consider trying to recover last night's F1 Racer skeleton, but soon gave up, well before my 30 minute deadline. It was just too unstable, and I felt much more confident about the 3-in-1 Bot, where I knew I already had procedures in M.W. EX which worked.

So I opened the Elliott book to chapter five for the second time in 8 or 9 days, and put it together from scratch, after having disassembled it about 18 hours earlier. I took frequent digital snapshots for my attempts of engineering a second axle with two wheels (second image folder, # 02) and a long extension in the front, for the horses (4 of them, as it turns out, not 6, but also not 2, or not not any). (second image folder, # 01)

image 02 of second piximage 01 of second pix

The second axle didn't present a problem, and I gave it fatter race car wheels for additional stability, caring more about stability and reliability than historical accuracy. Images 03 and 04 depict those pages from the book, with the addition of the front tires and longer chassis. The next 8 photos, # 05A, 05B, etc., all record excruciating details of my attempts, ultimately pretty successful, but with several missteps, to create a solid extension of the front. By image # 05-C, I had figured out a way to extend a series of beams forward to attach the horses, and had stabilized them with the braces and pins, as I had learned from the F1 Racer and the 3-in-1 Bot. Then I took out the horses I had bought earlier, and began considering how to make sure I could attached them to the chassis and not have their hooves drag along the ground.
image 03 of second piximage 04 of second pix
image 05-A of second piximage 05-B of second pix
image 05-B.5 (right version) of second piximage 05.B (wrong version) of second pix
image 05-C of second pix

In image # 05-D, I realized that the two short braces used to reinforce and connect the beams could also be used to help attach the first two horses, with either twist ties or rubber bands, whichever seemed easier to secure them and look a bit more realistic. I tried inserting an axle to give the horse an additional "seat" (seems appropriate, given all the horses who have given humans a seat, that a humans should return the favor), and since that both worked, but also helped reinforce the beams, I inserted a second axle to provide additional "seating" for the second set of horses, and to provide an additional connection between the two forward beams. At this point, after using two L-shaped connectors to elevate the last beams, it seemed cleared that trying to attach a third pair of horses, as I had first hoped, seemed to be pushing my luck.

Image # "05-E_front" shows both the second supporting axle, but also two extensions to help attach the second pair of horses to. I realized that I could add a 45-degree angle socket to the ends of the axle and accomplish two objectives for the price of one. Image # 05-E_top shows the various "saddle" attachments from the aerial view. The four versions of Image # 06 (A, B, C & D) all illustrate the final version of "Stagecoach II," with the four horses attached with rubber bands and several Lego figures borrowed from other sets to take the place of the Sheriff and the Deputy, who are at school. The space shuttle set provided the little front piece ahead of the driver and "shotgun" rider, with chairs from other sets. Images A & D are from the same angle, with D only having the human figures added on.

image 05-E of second piximage 05-E (Top) of second pix
image 06-A of second piximage 06-B of second pix
image 06-C of second piximage 06-D of second pix

I plugged in the infrared tower and opened the MicroWorlds EX Robotics, then copied and pasted the procedures which worked from Erik Nauman's e-mail, using the two motors to drive each of the wheels. Stagecoach II traveled across the kitchen floor just fine, although I didn't try many turns, not wishing to push my luck. Perhaps we'll film some footage of the stagecoach in motion during first period tomorrow first, then try some turns.

My older daughter also told me her friend was not only willing to donate his robotics kit to the cause, but had an RCX brick as well. If we pick up the stuff tomorrow afternoon as planned, then I may not have to disassemble Stagecoach II to make the Lego-saurus for the Second Period's space travel project. Even if I do, this structure (unlike last night's ill-fated, hasty production with too many straight bricks and not enough beams and reinforcement) would not be difficult to remove the RCX brick without a lot of problems.

Return to Summer School Main Page

Return to 7th Grade Main Page