|You are here: RoboFlag Main Page > Frequently Asked Questions|
I've taken some of the more popular questions emailed to me, modified them so they're anonymous, and posted them and my response below. The idea is that if you have a concern that might not have been addressed yet, this is where you could look first.
Is Carleton planning on providing us with any money? If so, how much? Also, would be receiving the money ahead of time, or will it be afterwards, on a receipt basis?
Carleton will probably not provide you with any money, but the Department of Electronics will be providing you with some equipment. The list of this equipment is on the webpage. You will need to sign a couple of forms and put a deposit on the equipment, which you will receive back at the end of the term. You will be expected to spend your own money on the project, and I suggest that pool your resources to even out the amount of money each student spends.
What components (ie: resistors, caps, inductors, the standard BJTs), if any, we can use from the department stocks?
You can use any (and all) DoE equipment that you can legally obtain - as long as they don't go against the no-"kit" rules. There's a resistor cabinet in the lab, and you should feel free to use any/all equipment.
Is there going to be set meetings with you (i.e. once a week) so that we can ask any questions and keep you posted on our progress?
There will be set meetings with me, and they will be once a week. They will have to be scheduled around my thesis work at Nortel, so: Wednesdays and Sundays, after 7PM. Book your spot soon, because I'll deal them out on a first-come first-served basis.
How can I hand my assignments in? Where should I drop them off if I don't want to email them to you?
You can get your assignments to me in a number of ways:
Do you have any HandyBoard expansion boards? Can we borrow them?
I have two expansion boards, and they are available for borrowing. If you'd like an expansion board, please write me a quick proposal (memo format, detailing why you need it, why you should get it, etc) and I'll give them out to the best memo-writing teams. The proposal request deadline will be Oct. 2nd.
What are the limits towards pre-built motors, wheels, chassies, etc?
As detailed in the rules:
How will the robots detect each other? Will they carry a beacon or something?
Yes, each robot will use an identical beacon that will be mounted on the highest point of the robot. These beacons will not be allowed to be obscured in any way. This is detailed in the rules.
What will the range of the IR beacons be? Will robots be able to detect each other when they are the maximum possible distance apart (and within line of sight)?
Standard IR emitters and detectors do not have the required range for the size of the arena, and it's important that robots be able to detect each other at any range (when in line of sight). The parts that will be specified for the emitter and detector will be rather specialized, and will work throughout the entire maze. This is detailed here.
Can we have more than one grabbing device?
We're examining various ways of locating and identifying the walls, and we want to know for example if the walls will be matte (non-reflective), because we are thinking of perhaps bouncing IR waves off the wall (different frequency from the IR beacons atop the robots of course), but if the walls are painted flat black for example, then our system will be useless. Basically, as much information regarding wall colour, reflectiveness and things like that as you can accurately provide will be tremendously useful.
The walls will be painted matte black in order to reduce the amount of infrared light reflected (to stop the IR beacons from being detected everywhere). From my previous experience with this stuff, it reflects IR light with an intensity loss. IR navigation sensors should work fine.
We are trying to design the mechanical model of our robot, but we
still don't know:
Good question! I don't know any of the answers!
What is the maximum amount of time that a robot will have to play for? We need to know this so that we can plan how to design the power supply. Will we have time between games to switch battery packs if needed? How many games do you think will be played on the contest day? We would also need this info to figure out how many battery packs to bring.
The maximum game length will probably be 10 minutes. There will be ample time to switch battery packs and even modify code between games. A good estimate of the number of games played on contest day is outlined in the types of games section of the specifications.
Is the flag grab ring energized?
No. The ring is conductive, but not energized. You must be careful if you plan to run a small current through the ring in order to facilitate detecting if the ring has been grabbed. This is further explained in the rules, under the Flags section.
Can we use HVW motors? (high speed, worm, planetary, etc)
Maybe. These motors + gearboxes are small and convienient, but they wouldn't be my first choice for a robot that might see a little contact. The motors are 3V toys - they're noisy, cheap, unreliable and no two will be alike. If you run them at 6V you're probably okay. If you run them at a higher voltage, they'll break quickly. The gearboxes aren't much better. They feature plastic gears. These will strip if exposed to high-torque situations, and aren't as reliable as metal gears.
However, they're cheap and easy. http://www.hvwtech.com/gearsets.htm
How will RF communication work? Will we be given the MING encoder and decoder boards? Do we have to have RF communication working?
The RF communication specifications are listed in the rules, under RF comms. These specifications did not initially address the I/O port requirements for the RF communications, but I have since updated the rules.
You will be provided with a MING 300MHz AM Transmitter and Reciever, DigiKey part numbers TX-99V3-ND and RE-99V3-ND. You will need to purchase the daugherboards for encoding/decoding, DigiKey part numbers RE-01-ND and TX-01-ND.
Yes, you should be able to communicate with your teammate via RF.
Will the PIC programmer in the lab program a PICxxx?
Probably. The PIC programmer is a simple serial-connection (via the PC's parallel port) programmer that has five connections to each PIC. The connections are +5V, GND, ~MCLR, DATA, and CLK. You can find which pins these programmer wires must be connected to from the microchip datasheets (available from the http://www.microchip.com/ webpage.
The list of PICs the programmer can program includes:
Do we really need six IR emitters for the IR beacon?
You don't specifically need six emitters, but you must emit an IR beacon signal in all 360-degrees. That said, the IR emitters that I specify have a half-power-beam-width of 60-degrees, and in order to ensure a 12' detection distance I suggest you overlap the beams at the half-power angle. That, to me, suggests that you need 6 IR emitters.
Is there a limit to the motor driver chips/boards we can use?
In order to discourage teams from buying a golf cart motor control board or the latest and greatest motor-driver IC, I'm setting an artificial limit of 1A/motor if you're using something you're going to buy. What that means is that if your motors draw less than 1A when stalled, you should use the HandyBoard with the Texas Instruments SN754410 "Quadruple Half H-Driver".
If, however, you want more than 1A/motor, I want you to work for it. Feel free to design a killer H-bridge that can source 5A peak, uses huge MOS or IRFZ FETs, and will require it's own cooling fan. PCB it, breadboard it, do what needs to be done.
What constitutes a "dead end" in the arena?
This is a touchy subject. I consider a "dead end" to be a position in the arena when three walls are within 2 or 3 inches of a 12" robot. This means that there will be no tight passageways where only one robot can fit. If a robot can be approached from two different directions, I don't consider the arena to have a "dead end", as shown below.
---+ ----+ O| <- A dead end O| <- Not a dead end ---+ | ----+
This means that a "U" shape in the arena is legit, as long as it is wide enough for a couple of robots to sit side-by-side. Pay close attention to the suggested arena shown in the rules section!
What is the flashing flag beacon, and what can we use to detect it?
What frequency will the two flag beacons work at?
20Hz and 30Hz.
What can you tell us about detecting the flag?
This is a fairly difficult technical task. Your team PIC expert should be brushing up on their programming, thinking in assembly, and worrying about each and every microsecond.
Once you've got a plan to extend the PIC IR beacon (and you've gained some PIC experience), you should think about programming the PIC flag detector. This detector will obviously need to interface with a number of light detectors (which are detailed above) and will need to:
The problem lies in the sampling and decision-making. It's not difficult to perform a simple digital sample over time (say, xx samples over .1 seconds). If the signals were digital, it'd be easy enough to decide if there were two oscillations (each taking 0.05s, or 20Hz) or three oscillations (each taking 0.033s, or 30Hz). It's too bad that the physical light detector sensor probably won't be digital.
The ambient light value will probably float as the robot moves towards and away from the black arena walls. Can you detect a small 20Hz signal on top of a randomly fluctuating ambient light value? This is a lot harder than interfacing with the PNA4612M module (which does the IR demodulation for you). You'll definitely need to think (and test) the light detectors in order to come up with a good plan.
I'd also suggest you think about combining the electronic and digital solutions to the flag detector. A pure PLL solution isn't very practical (lock frequency ranges, noise issues, etc) and probably won't work. The PIC is an excellent processor to perform this type of digital sampling and filtering.
What will the arena floor colours be? How can we detect the different colours?
The arena zones will probably be something like a dark red and a light blue team zone (the 1' strip at each end of the arena), with matte black walls and arena floor. You will need to use some sort of reflectance sensor to detect how much the different colours reflect light. There will be ample time to tune your sensors to the three different colours used in the arena.
A reflectance sensor can be made with an IR emitter and detector pair (tuned to the same wavelength). Dark surfaces will absorb more light than light surfaces, and you can use this sensor to get an idea of the floor colour. You would probably want to have multiple emitter/detector pairs and perform an averaging. Good shielding will be important, and the distance to the floor will change the sensitivity of the device. You'll want to have the emitter/detector pair less than 1.5cm from the floor. The angle of the emitter/detector pair will also feature in the device sensitivity.
Can we tell the robot (before each match) what direction it needs to go?
No! You won't be allowed to give your robot ANY special instruction (except for what type of game it's currently going to play). You should download the code for the type of game, move the robot into the arena, and press "START". That's it.
The robots will be positioned with their "back" to the wall in their own zone, pointing their "front" towards the other team's zone. Robot placement within the zone is decided by the course coordinator or competition judges.