Hey guys, I only have experiance in visull programling like mindstorms. I was gonna buy the starter robot to start figgureing out how to do real programing but in the box, several advised i go with Arduino instead. I certainly dont want to learn a dead language, so, what do I need so I can use the Arduino instead to build the starter bot? Is the voltage requierments the same? Can it use the same yelloe chip and the same motor controler thingy. Is the rest of the start here bot part compatible with the Arduino board? If not, can you recomedn what parts are needed?
Arduino. The language it’s just like a simplified C, way more similar to Python and Java than the old style C. Once got the paradigm it’s easy to switch from Arduino to Java, Python, Javascript and other.
Arduino gets limiting when you get to deal with many listeners, data structures and advanced math, but it’s more limited the atmel chip on which you are running.
The tadpole is a excellent starting platform. I am signed up to teach an adult ed course starting next month (requires at least six students) and we will be using the tadpole educational machine. Chris at Rocket Barnd Studio makes a super product and is extremely helpful. The quality and accuracy of the laser cut acrylic is astounding and the Dagu controller has a very complete and well organized set of I/O headers. I have writen a dozen programs to exercise the robot and everything works as advertized.
I am currently debugging a quadrature encoder board for the tadpole that uses the magnetic disc from Dago and Hall effect chips. Having trouble getting the proper mechanical alignment of the sensor but results are promising. I highly recomed you get some form of rotational feedback.
I’m not sure this size robot will work at Trinity. It is much smaller than the candel!! This contest is what got me hooked on robots. Would be interested in hearing about your prior entries and current plans.
I also suggest the Arduino I know several computer languages, including some obscure ones like APL, LISP, and Forth. However, I tend to prefer for C family of languages for real down to dirt programming.
This may not be your dream board. As you build robots that have more AI, you’ll probably want something larger. But for now, an Arduino will do what’s needed to control a robot.
One thing to think about: I built my first digital computer from a kit. This was around the last half of the 1970s. The MPU was a 6502. It was a 8-bit micro with some add one to allow it to use memory. The first build had the processor board and a memory board in a huge wooden box. The capabilities of that box was less than the Arduino. The Challenger had 2k of RAM, but I believe it ran at kilohertz speed.
So don’t think that the Arduino is a wimp. It’s a good controller and it will allow you to learn C.
Next time i hit trinity, I’d 3d print a frame or fabricate out of aluminum. I’m not planning on taking the little bot to trinity. Im planning on learning to program from the little bot, then building something way more advanced for trinity. How many sensors can that board read and how many servoes/ motors can it handle? I’m good on the building/physics side or robotics.The electrical I can figure out. The software is my weekness.
Last time for the trinity competition (10 years ago), we had nothing but sensor problems. We used an RCX because none of us knew how to program. We had programmed it to use 3 ir sensors to navigate and servo mounted candle detector. On final testing we had a power problem that destroyed one of our rcx and 2 ir sensors.(the internal RCX batts aren’t supper reliable on speed so we had an external power source) We literally re wrote our entire program for the thing to navigate with only 1 rcx and 1 distance sensor. This involved writing a code for the robot to back through part of the course since our distance sensor was fixed on one side. By code, i mean visual blocks, but still hard to redo the whole way our bot drives in a few hours. We finally got a whole new navigation system set up and got it to work with our bot driving much slower. Without all the sensors we had to dead recon turns and pull up until it found wall.We are doing our final test and the light sensor failed. My team wanted to quit but I pulled out the junky lego sensor that couldn’t detect the flame but could detect the white circle. That year, using the circle was a penalty, but it meant we could compete. When it found the circle it just went full blast on compressed air can and swivled, hoping the flame went out.It didn’t know the orientation to the candle. We did great on our first 2 runs and had really good times, especially considering we basically built the whole programming in a day. On the last run either the ir sensor failed or a camera flash made it mess up, but our robot crashed in to a wall full throttle. Since the reprogram made out bot turn so slow, we hauled ass on the straits.(right into a wall)We did however do really well in the robot written test and actually placed. Long story short, i think we did pretty well considering half our equipment failed. The rules say this is not a build competition and no team has ever fully programmed their bot at the competition. We did lol.We came in 14th but with the times scores we had in the first to run we would have placed if we did it a 3rd time.
So, plans for next year?
Redundancy is key.I want to figure out how to make the robot realize a sensor is bad and jump to another program on the fly. I’m thinking 2 side ir sensors and one roving in front. (on servo) I also want to program it to use a roving sonic sensor. I’d like to make it be able to do the whole course on either. If the ir throws an odd reading i want to program the bot to switch to only using roving sensor mid trip.If that goes, switch to roving sonic. basically have a fail safe. If a sensor reads more then x or less then y, then start program 2 or 3 etc.
Im thinking 4 wheel steering.Its more complex, but way better handling.Let’s face it, speed is key. The winner in the comp we went to had a robot with 2 wheel steering that leaned and went 3 times the spees of any other bot. You can go faster around turns if you don’t need to worry about the nose or ass hitting a wall. Without totally giving away my strategy…i think it’s best to figure out orientation start location based on distance from walls. Loop the search and do just start the loop depending on where they drop the bot.Also, on the right 2 walls, we know where the walls could be, we just don’t know where they are.Simply have the bot check for walls and change its search through that room based on findings.If the wall is only one of 2 places, it’s pretty easy for the robot to check and thus know where the wall is and what program to run.As for the dog, im thinking ir should stop the bot, but id put a retractable bump sensor on front.If the sensor is right on front of the bot it will move the dog more than one cm and fail the run.If the sensor mount can retract back 3-6 inches, it gives the robot more time to stop.This would only be secondary to finding the dog with ir or sonar.Obviously when it finds the dog, the bot goes to plan b. But what do I know, I need to play around with something other than mindstorms to know if this is all possible.Im assuming the “if/then” are a bit easier in real programming.The old mindstrorms would only let you go 4 deep with ifs in ifs.
Thanks fo the reply. Im more concerned with how many sensors/mortors i can hook up to it.
trinidy now requiers sound start, leds when u find the flame…a start and stop button. Plus you need a lot of sensors just to navegate and find the flame.
for my mower project i will need tons of sensors to. When you start talking about dual motor drive, electronic brakes, the mower blade, you need a lot of motor ports.
One four wheel steering One four wheel steering mechanism uses the mechano style wheels. These are the wheels that have wheels along their rim, so they can go sideways or forward or or with the right programming anyway you want. The Dagu ComMotion board is designed for such things, and you can ask questions of the designer, OddBot right here at LMR.
