Maze Solver

Goals of this project:
1. Use the Arduino board to build your prototype and use a barebone ATmega328p microncontroller for the final product.
2. Integrate the entire circuit onto a PCB board.
3. The DIY maze solver must be able to complete the maze through the optimal path.

1. Introduction

The left-hand or right-hand rule (http://arcbotics.com/lessons/maze-solving-home-lessons/):

One of the simplest ways of solving a maze with a robot is by using the Wall Follower algorithm, also know as the left-hand rule (or right-hand rule). Forget about the robot for a while, and suppose that you are a person inside a maze. Finding the exit could be done just by keeping one of your hands always touching a wall.  And by “always”, we really mean always here. It might take you a while and you might wind up taking all the wrong paths in the maze before getting to the end but by keeping your left hand on the wall of the maze you wind up taking every single left hand turn (sometimes turning around completely) and, as long as you keep taking steps forward, eventually you’ll get to the end of the maze.

2. Test the TCRT5000 IR pair

The datasheet of IR pairs available to you can be found here.

You must find the polarity of the pins by reading this datasheet carefully. It won't tell you how to make connectiosn to the pins directly but you will find answers from a few places on your own.

On the first page of the datasheet, the following figures tells you the polarity and the internal circuit of the IR pairs:

The name of the teminals are not given but only a few letters are given. You'll learn this in any 'Microelectronics' or 'Analog Electronics' classes. C: Collector of the BJT transistor, E: Emmitter of the BJT transistor, A: Anode of the diode, C: Cathode of the diode. Now you have enough information to make connections to this IR pair.

You may have done the robot car project from our ENGR 201 lab. If you did not or you want to review what you have done, go to this website: http://yilectronics.com/Courses/ENGR201L/ENGR201L_2019f/Lab5_The_Line_Follower/Lab5.html

The schematic for the core circuit was:

We can definitely re-use the most part of this circuit. The resistor at the LED side is at the ~100 ohm level (I used 560 ohm in the circuit below), the purpose is just to limit the current and protect the LED. However, on the BJT side, we need a 10k resistor as the 'pull-up' resistor. This resistance must be significantly higher than the ON mode of the BJT to take over all the voltage shares in this 'Voltage Divider'.

I tested this IR pair with the protection resistor and the pull-up resistor and it worked very well.

First, I powered it up and checked the IR emitter:

Then I use my figure to reflect the IR from the emitter to the receiver.

The demo video can be found below:

Task 2: Repeat the work in the demo video above, show your result in a VIDEO for the report.

3. Assemble and solder your IR sensor array

Solder the IR pairs onto a prototype PCB board. I solded four pairs in the following figure at the beginning and added an extra one later. I recommend at least 5 IR pairs for this module. (I ended up with using 6 of them).
My Samsung phone's camera can visualize part of the IR spectrum so you can see the purple light in the following figure.

To test if the receiver can detect the reflection, I added LEDs to the circuit.

This video proves the IR pairs are functioning.

Without an comparator, the intensity level of the LEDs are not binary, which means depends on the reflection intensity, the LED's brightness varies in a range. This is pretty bad for motor control because you cannot quantify the reflection intensity and use this intensity to control the motors' speed. We need to add comparators to the circuit.

With comparators, the LEDs will only have two states, ON and OFF. It is demonstrated in the following video.

Task 3: Repeat the work in the demo video above, show your result in a VIDEO for the report.

4. Build the main circuit of the car.

This car is controlled by a CPU but not simply a DC circuit. Use the control strategy you learned from the Elegoo Smart Car V3 to make the hardware connections.
The speed of the wheels should be controllable. Use PWM to power up the 'ENABLE' pins of the two motor drivers on the same chip. Use L293N as the motor driver.

The pin map of the motor driver can be found here:
http://yilectronics.com/Courses/ENGR201L/ENGR201L_2019f/Lab5_The_Line_Follower/Lab5.html

I used the batteries from the Elegoo car kit since they are rechargable. The following figure shows the complete circuit of the prototype. You may have noticed that I used 6 comparators which means I have 6 IR pairs at the bottom of the car.

Check out the following videos:

Task 4: Repeat the work in the demo video above, show your result in a VIDEO for the report. (Use the barebone ATMega 328 chip directly on the breadboard, do not use the Arduino Uno board on the car).

Task 5: Design the code to enable the car be able to complete the following maze.

5. Optimize the path

Task 6: Read the following slides and optimize the path of maze for the car. The performance of the car won' t be as good as the one sold by Pololu but it should be able to slowly complete and optimize the path for the maze.

Slides Tutorial
Demo video from Pololu