Troubleshooting:

It doesn't go at all.

First be sure that the motor leads are touching the things they need to touch, and only those things. One should be touching the top of the battery and the other should be firmly pressed against our tin framework.

Check for shorts. Make sure that the only bit of metal touching the top of the battery is

Make sure that there is nothing keeping the weight at the top from spinning.

It falls over more than I'd like.

There are several ways you can counteract this.

Bend the motor back towards the center of the battery so its center of gravity is more in the middle.

Bend down the corners of the "front" of bottom platform.

It might be too vigorous, you can try letting the battery run down a bit so it doesn't jump as much.

Try a different surface. I found that a pad of paper was the most reliable. On a harder surface they'll bounce easier.

Get out a file and remove some of the weight from the top of the motor.

It falls over less than I'd like or doesn't act very drunkenly.

Give it a double whiskey neat and wait 10 minutes.

You might be using an under powered battery., especially if you're using a bigger motor. Try a fresh battery or a more powerful cell. (If you use a different battery you'll need to rework the tin holder.

Make sure the motor isn't shaking lose. If it is, a dab of glue or tape can take care of your troubles.

DOWNLOADS: 

CLICK HERE TO DOWNLOAD THE REPORT OF DRUNKEN ROBOT

A switch is used to turn the robot on or off. When it is on it is connected to a power supply of 4 AA batteries with 1.5 volts each for a total of 6 volts, this is considered the unregulated power. Unregulated power goes to a 5 volt regulator. Regulated power runs to the PIC, the H driver, the RJ11, and the 4 sensors. Unregulated power runs to the H driver as well.

The robot decides its direction based off of the outputs of the four sensors. The robot has 4 different states, and they are: Forward, Left, Right, and Turn Around. The state priority is in this order: Left, Forward, Right, and, lastly, Turn Around. The four sensors are placed close together at the front of the robot. Each sensor has a corresponding LED that lights up when the sensor is high. The left and right sensors are slightly farther back then the front two sensors and the front sensors are centered and side by side. The robot enters the Left state whenever the left sensor goes high until the front two sensors go high. When the two center sensors are high and the left sensor is low, the robot enters the Forward state. If neither of the previous conditions are true and the right sensor is high, then the robot enters the Right state until the front two sensors go high. If none of the sensors are high, then the robot enters the Turn Around state where it does a 180o right turn in place.

With the 4 previously mentioned states our robot is able to make turns, turn around from dead ends, correct itself on straight lines, and create random turns that ignore the left turn priority. The robot is able to do the second 2 mentioned abilities because of the positioning of the sensors. By having the left and right sensors extremely close to the front sensors, the robot is able to make very small left and right turns to keep itself on a straight line. The robot is able to randomly ignore state priority because of while loops used in the code. For example, the motor is coming to a 3 way intersection with left and straight directions in front of it. Under normal priority the robot should turn left at the intersection. When the robot approaches the intersection it may be caught in the Right state in order to correct itself on the straight line. The robot exits the Right state once the front 2 sensors have gone high, so there is a possibility that the robot is in the Right state as it enters the intersection. If this is true, the left sensor will be ignored until the front sensors go high and the robot will go through the intersection straight because the left option was ignored.

By being able to stay on the lines of the maze, follow turns, turnaround, and provide occasional random turn priority, the robot should be able to find its way through any maze eventually even if there are loops within the maze.

Overall, the final design worked as intended. However, several different versions of our wiring diagram were created before we had a working robot. This created some setbacks during the construction phase, however any problems that arose during this time were quickly found due to our methodical checking of the circuit being built at the time for any shorts or wrong connections.

Throughout the duration of this project we gained experience in building circuits that worked with each other to create a final outcome, and along the way we learned a few valuable lessons. Checking solder connections meticulously pays off and will save a lot of time in future work, and checking that everything works as intended on a system-by-system basis will further help with the overall construction of any soldering project. Once a system is checked for physical connections, the programming interface must be tested as well. We wrote a lot of very short programs to test each module for individual functionality before interfacing it with another section of the design, as its easier to debug one small section rather than one very large section consisting of several subsections.

In addition to checking for proper connections, ohmmeters are the quintessential debugging tool for a circuit physically as well as logically for code, and one should be kept nearby at all times. These two lessons saved a lot of time when creating the final version of our design, without running constant checks on our circuit with an ohmmeter we would not have been able to assemble our robot as quickly as we did.

The other important lesson learned came from the design of our circuit: DC power is a valuable asset when debugging a circuit, especially if the final design is battery powered. DC power allows the user to constantly run new tests without worrying about draining batteries and having to replace them constantly. DC power may draw more current, but in the end, if the circuit can handle the DC power it will be able to run on batteries.

C1, C5, C6 : 100uF/16V Electrolytic Capacitor

C3, C4, C7, C8: 10uF/16V Electrolytic Capacitor

C2 : 100nF Polyester Capacitor

Con1, Con2, Con3, Con4 : 2×4 Terminal

D1 : 1N4001 Diode

D2, D3, D4, D5, D6 : 1N5818 Schottky Diode

Q1, Q2, Q3, Q4 : IRF3205 Power MOSFET

R1, R2, R3, R4 : 1/4W Resistor

U1, LM7805CV Linear +5V Voltage Regulator

U2, U3 : IR2110 High and Low Side Driver

When one current way is off, namely its control signal is low, the boost up capacitor is charged up. When this way turns on, the boost up capacitor starts to bias the high side MOSFET until it fully discharges.So it is not possible to drive the motor in one way continuously without a PWM control signal. By using PWM control signals you can easily adjust the speed of motor and continuously run the load in one way. Same is also valid for the other way of current.

The MOSFETs used in this project are International Rectifier’s IRF3205 which can handle up to 115A drain current and 55V Drain to Source voltage. It has 0.008 Ohm RDS resistance. For lower currents (~0-5A) heat dissipation will be too low. But if you will use this board for high current applications you should connect a heatsink. On the other hand you can choose a different MOSFET that suits your needs.

As it is shown in the schematic, we input +12V DC supply voltage to the board. +12V is used for gate driving of MOSFETs. A LM7805 linear voltage regulator converts +12V to +5V DC for the logical supply of IR2110 which is suitable for microcontroller applications.