introductionhardwaresoftwareresu.htm

机器人应用的全套实例寻机小车

  well. Originally, in drive mode, the rear motor was constantly on. To slow the 
  car down, we sent a PWM signal to constantly turn the motor on and off. The 
  PWM slowed the car down to approximately half the original speed. Driving 
  slower, the sensors were then able to sense turns in the 
  track.</FONT></SPAN></P>
  <P class=MsoNormal style="LINE-HEIGHT: 150%" align=justify><U><SPAN 
  style="FONT-WEIGHT: 700; LINE-HEIGHT: 115%; FONT-FAMILY: Palatino Linotype"><FONT 
  size=4>Relationship of your design to available IEEE, ISO, ANSI, DIN, and 
  other standards</FONT></SPAN></U></P>
  <P class=MsoNormal style="LINE-HEIGHT: 150%" align=justify><SPAN 
  style="LINE-HEIGHT: 115%; FONT-FAMILY: Palatino Linotype"><FONT size=4>All 
  components used for our project has been approved by the IEEE compliance 
  standards and there are no standards that pertain to our 
  project.</FONT></SPAN></P>
  <P class=MsoNormal style="LINE-HEIGHT: 150%" align=justify><U><SPAN 
  style="FONT-WEIGHT: 700; LINE-HEIGHT: 115%; FONT-FAMILY: Palatino Linotype"><FONT 
  size=4>Existing patents, copyrights, and trademarks which are 
  relevant</FONT></SPAN></U></P>
  <P class=MsoNormal style="LINE-HEIGHT: 150%" align=justify><SPAN 
  style="LINE-HEIGHT: 115%; FONT-FAMILY: Palatino Linotype"><FONT size=4>There 
  are no relevant copyrights, patents, and trademarks for our project. 
  Line-following robots already exist without a patent and our car 
  implementation does not infringe upon any copyright or 
  trademark.</FONT></SPAN></P></BLOCKQUOTE>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
<A 
href="http://instruct1.cit.cornell.edu/courses/ee476/FinalProjects/s2007/bl222_wh84/bl222_wh84/index.html#top">Back 
to Top</A> 
<H1><FONT face="Palatino Linotype"><A id=hware></A>Hardware</FONT></H1>
<BLOCKQUOTE>
  <P class=MsoNormal align=justify><FONT face="Palatino Linotype" 
  size=4><B><U>H-Bridge Circuit</U></B></FONT></P>
  <P class=MsoNormal align=justify><FONT face="Palatino Linotype" size=4>We 
  implemented a full H-Bridge and a half H-Bridge to control the turning of our 
  motors. Since we needed our front motor to turn both directions for left and 
  right, we connected a full H-Bridge to the front motor, which allows us to run 
  the current in two different directions. Because we did not need our car to 
  drive in reverse, a half H-Bridge was sufficient to control the rear motor. A 
  schematic of our full and half H-Bridges are shown in figures 2 and 3 
  respectively.</FONT></P>
  <P class=MsoCaption style="TEXT-ALIGN: center" align=center><IMG height=635 
  src="IntroductionHardwareSoftwareResu.files/HBridge.jpg" width=1069 
  border=0></P>
  <P class=MsoCaption style="TEXT-ALIGN: center" align=center>Figure 2: H-Bridge 
  for Turning Motor</P>
  <P class=MsoNormal align=justify><FONT face="Palatino Linotype" 
  size=4>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The H-Bridge 
  design is essentially four switches, where the switch is a BJT. When the top 
  left and bottom right switches are turned on by a logic high from the MCU, the 
  current will flow from the battery source through the switches and will spin 
  the motor in one direction. When the top right and bottom left switches are 
  turned on, the current will flow the opposite direction through the motor. 
  This opposite direction of current flow causes the motor to spin the opposite 
  direction. When the two top switches are turned on and the bottom two switches 
  are turned off, it creates a short circuit across the motor. This short causes 
  the motor to spin in both directions and hence simulates braking or 
  deceleration. Although we could also stop the car by simply turning off all 
  the transistors, we noticed during testing that there would be a delay when 
  the steering motor turns off.&nbsp; Since we didn't want the car turning more 
  than it should, we added the braking logic so that there would be quicker stop 
  in the front motor.&nbsp; Switches on the same side (left or right) are never 
  turned. If they are turned on simultaneously, there will be a short circuit 
  across the power supply and will damage the circuitry as well as the power 
  source.&nbsp;&nbsp; </FONT></P>
  <P class=MsoNormal style="TEXT-INDENT: 0.5in" align=justify><FONT 
  face="Palatino Linotype" size=4>Since the motor will be switching on and off 
  in two directions, protection diodes are needed across the collectors and 
  emitters of the BJTs to prevent inductive kickback that could damage the motor 
  and transistors. Fortunately, the TIP102s and TIP107s we chose already had 
  protection diodes built-in so there was no need for external protection 
  diodes. To ensure that the noise generated by the motor does not affect the 
  MCU power and ground, we used opto-isolators to turn on the BJTs. Ports A3 - 
  A6 are connected to the inputs of the opto-isolators, which isolate MCU power 
  and ground from the H-Bridge power and ground. To prevent too much current 
  flowing from the MCU ports into the input of the opto-isolators, 330Ω 
  resistors were connected to them to limit the current. We chose an 
  opto-isolator packaged with four units on a single chip to facilitate circuit 
  population. Each of the four opto-isolators for the full H-Bridge is connected 
  to the base of one of the four BJTs.&nbsp; Figure 2 shows which MCU ports are 
  connected to which BJT gate.</FONT></P>
  <P class=MsoNormal style="TEXT-INDENT: 0.5in" align=justify><FONT 
  face="Palatino Linotype" size=4>The half H-Bridge was constructed much the 
  same way except it only used one NPN, one PNP, and two opto-isolators. Only 
  one control signal is required to turn both BJTs on and thus turn the motor 
  on. We did not implement braking for the rear motor because there was no need 
  for sudden braking. The input signal to the BJTs for the rear motor is 
  connected to Port B.3 on the MCU. The schematic of the half H-Bridge is shown 
  below in figure 3.</FONT></P>
  <P class=MsoCaption style="TEXT-ALIGN: center" align=center><IMG height=617 
  src="IntroductionHardwareSoftwareResu.files/Half_Bridge.jpg" width=640 
  border=0></P>
  <P class=MsoCaption style="TEXT-ALIGN: center" align=center>Figure 3: Half 
  Bridge for Rear Motor</P>
  <P class=MsoNormal><B><U><FONT face="Palatino Linotype" size=4>H-Bridge 
  Testing and Results</FONT></U></B></P>
  <P class=MsoNormal><FONT face="Palatino Linotype" 
  size=4>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
  Initially, we tried using NMOS transistors to construct our H-Bridge. In the 
  MOSFET H-Bridge design, the drains of the top two transistors were connected 
  to the positive end of the battery and sources of the bottom two transistors 
  were connected to the negative end of the battery. The sources of the top 
  transistors were connected to the drains of the bottom transistors and the 
  motor was placed between these two nodes. We applied 5V to the gates of the 
  top left and bottom right transistors to supply voltage to the motor. However, 
  the across the motor was consistently around 1V no matter how much voltage we 
  supplied to the H-Bridge. We had first supplied 5V to the H-Bridge with only 
  1V across the motor. We increased the voltage to the H-Bridge to 12V and the 
  we still measured 1V. We then realized that 5V was insufficient to put the 
  switching NMOS into saturation. Instead, the NMOS transistors were still in 
  the linear region and thus limited the current flow and thus limited the 
  output voltage. We raised the input voltage to the NMOS gates to 7V with a 5V 
  supply to the H-Bridge. This time, the NMOS transistors were in saturation and 
  the H-Bridge gave an output voltage of 4.6V. Although we had gotten our MOSFET 
  H-Bridge to work, the 7V needed to turn on the transistors posed a problem 
  since the MCU ports could only provide 5V when turned high.&nbsp; Also, when 
  we applied 7V to the MOSFETs, they became very hot and this would pose an 
  unacceptable risk to the user.&nbsp; Of course we could use heat sinks but 
  that would increase costs and make construction more inconvenient.&nbsp; 
  </FONT></P>
  <P class=MsoNormal><FONT face="Palatino Linotype" 
  size=4>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Due 
  to this problem, we decided to try using BJTs instead of MOSFETs for our 
  switching transistors. The two high side transistors were PNPs and the two low 
  side transistors were NPNs. We picked TIP102s for the NPNs and TIP107s for the 
  PNPs because they were both Darlington configurations, which have high current 
  gains, and they had protection diodes built-in. Furthermore, these BJTs came 
  in TO-220 packages which are easy to use on the protoboard.&nbsp; The 10kΩ 
  resistor between the output of the opto-isolator and the base of the BJT 
  limits the current that goes to the gate.&nbsp; According to the TIP102 
  datasheet, the max base current is 1 A but we decided to be more cautious and 
  use a large resistance value.&nbsp; </FONT></P>
  <P class=MsoNormal><FONT face="Palatino Linotype" 
  size=4>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; We 
  tested the BJT H-Bridge by applying 5V to the top left and bottom right 
  opto-isolators and also supplying a separate 5V to the H-Bridge power supply. 
  This time, the voltage across the motor was approximately 4.6V so we were able 
  to see at least what we were supplying to the H-Bridge across the motor. We 
  then connected the H-Bridge to the motor, and the H-Bridge successfully 
  powered the motor. Since we wanted our car to run without being connected to a 
  power supply, we decided to power the H-Bridges using batteries. We initially 
  thought of using four AA batteries in series to provide 6V to the H-Bridges. 
  However, when we tested the car by supplying 6V to the H-Bridges using the 
  power supply, the car had difficulty fully activating the turning and driving 
  at the same time. The two motors needed at least 7V for them to run 
  simultaneously without problems. Therefore, we decided to power the H-Bridges 
  using a 9V battery. We initially tested running our car off of one 9V battery 
  but the car had difficulty driving because the battery was not providing 
  enough current. Thus, we connected a second 9V battery in parallel to supply 
  the H-Bridge with more current. With increased current, the car ran 
  normally.</FONT></P>
  <P class=MsoNormal><B><U><FONT face="Palatino Linotype" 
  size=4>Phototransistors</FONT></U></B></P>
  <P class=MsoNormal><FONT face="Palatino Linotype" 
  size=4>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; To detect 
  black lines on a white surface, we chose the QRB1114 phototransistor for its 
  low cost and easy implementation. The QRB1114 is a package consisting of an 
  LED and a phototransistor BJT. The transmitter is the LED, which emits 
  infrared light that reflects off a surface.&nbsp; The amount of reflected 
  light is measured by the BJT, which sinks more current to ground if it detects 
  more reflected light.&nbsp; White surfaces reflect light well whereas black 
  absorbs most of the light and thus very little is reflected back to the 
  receiver. A schematic of the sensor circuit is shown below in figure 4.&nbsp; 
  Note that the schematic for the photosensors is somewhat misleading.&nbsp; The 
  light from the LED does not directly activate the BJT, which is what happens 
  in an opto-isolator.&nbsp; Rather, the light reflects off a surface and 
  activates the BJT.</FONT></P>
  <P class=MsoNormal>&nbsp;</P>
  <P class=MsoNormal align=center>&nbsp;<IMG height=730 
  src="IntroductionHardwareSoftwareResu.files/photosense.jpg" width=640 
  border=0></P>