Analysis 1. Hardware 1.1.1 Robot Base The base will be a Lynxmotion CarpetRover 6" x 6" mobile robot platform powered by a pair of Hitec HS-300B servos modified for continuous rotation. See document: http://www.lynxmotion.com/kits.htm See document: servos.html 1.1.2 Microcontrollers A Lynxmotion "First-Step" Basic Stamp I controller will be used exclusively to run the drive servos. See document: lynxbs1.gif A Parallax Basic Stamp II mounted on a BS2-IC Carrier Board will be used for reading sensors and navigating the maze. See document: bs2csche.gif See document: bs2cb.gif See document: stamp2.gif See document: robotwir.txt 1.2 Sensors 1.2.1 IR Object Detector An Infra-Red Object Detection System was designed and fabricated on 03-11-99. The IR receiver module (RadioShack #276-137B) output was too susceptable to false triggering due to ambient lighting so it's output was integrated and fed into a comparator with an adjustable threshold reference voltage. A threshold was picked to eliminate false triggering. IR attenuation filters (slips,of paper) will be taped over each of the four receivers in varying thicknesses to equalize the sensitivity of each module so they all trigger at approximately 5 inches distance from the maze wall. See document: irsensor.gif A 74LS156 1 to 4 line Multiplexor was added to route the 40 KHz clock to 1 of 4 IR LED's using only 2 select lines. A 74LS153 4 to 1 line Demux. was added to select 1 of 4 IR Receiver Modules and route it's output to the integrator circuit using the same 2 select lines. See the Prototype section for a more detailed schematic. 1.2.2 Flame Detector During a period of indecisiveness over whether to buy a commercially made flame sensor, such as Acroname's IR flame sensor or Hammatsu's Ultraviolet flame sensor, See document: http://www.acroname.com/robotics/parts/R3-PYRO1.html See document: http://www.hpk.co.jp/products/ETD/UVtronE.htm I experimented with some IR phototransistors to see if it would be feasible to design our own. The first tried was a Radio Shack part. Unfortunately the ambient overhead flourescent lighting had more effect than a candle flame on this device. It was evident that both optical filtering and focusing would be required to make this feasible. I ordered every photo device in the Electronic Goldmine catalog along with some mirrored flashlight type reflectors and some Kodak 87C IR filter material. I also salvaged the IR lens material out of an old passive IR motion detector. 1.2.3 Wheel Optical Encoders I purchased some Reflective IR Sensor/Emitter units from the Electronic Goldmine (cat.# G8669) that looked ideal for this application. These are an IR phototransistor and IR LED paired together in a black plastic case, such that reflected IR light from a close proximity object will cause the phototransistor to conduct. These can be mounted on the underside of the Lynxmotion Carpetrover robot base, facing the inside surface of each tire. Circular paper disks colored with alternating flat black and flat white color bands can be glued to each tires inside surface. Once optimally positioned, the phototransistors should output pulses marking distance traveled by the spinning tires. Here is the related calculation: Tire Diameter = 2.875" Tire Circumference = PI * D = 3.14159 * 2.875" = 9.03207" Thus, Distance traveled = 9.03207 inches/revolution. Resolution of this system is determined by the number of white color bands on the disk. Thus for a disk with 8 white color bands seperated by 8 non-reflective black color bands, the resolution is 9.03207" / 8 or 1.129" per pulse. This is provided the Basic Stamp II controller can count every pulse without missing any. 2. Software 2.1 Robot Base A 3 bit bus will be used to direct 5 cardinal motions: Hold, Forward, Back, Right, and Left. A flowchart was completed on 02-21-99, and the PBasic program was written and debugged on 02-22-99. This would become the basis for motion control of our robot via the Basic Stamp I controller. 2.1.1 Flowchart See document: lynxflow.gif 2.1.2 Sourcecode See document: lynx.bas 2.2 Sensors See the next section "Prototyping" for info. on construction and further refinements.
Back to Robot Main Page