Micromouse Team
Fall 2008

---

Micromouse Design 1.0

Team Members: Chris Gibson, Ambrose McJunkin, Tim Montague

Parts Involved:

� 2 DC Motors
� LCHB-SN754410 Dual H-Bridge Module
� M9S08QG8 Microcontroller
� Gyroscope
� 3 Ultrasonic Range Finders
� Aluminum Frame, Wheels
� Rotary Encoder

The motor will be driven by the Dual H-Bridge Module. The microcontroller will communicate directly with the H-Bridge module to move the robot and change the direction of rotation of the wheels, making it turn right or left. The gyroscope will be used to keep the robot following a straight path. Also, the gyroscope will assist in helping the robot make accurate and consistent 90� turns. The encoder will keep track of wheel rotations, and from that data the microcontroller will calculate the distance the robot has covered, lending to our algorithm. Two of the Ultrasonic Range Finders will detect changes in maze geometry on the sides of the robot, alerting the robot to adjacent paths. The last will be used to alert the robot to any immediate obstacles in front of it, such as a dead end or L turn in the maze, and provide input to the algorithm in the microcontroller that an alternate direction of travel must be sought.

Building it:

It�s necessary to get the robot built and performing simple movements before we can test any algorithms. This means deciding on the robot�s hardware components will be the first big step in our design process. We would like to get the robot moving as soon as possible, so we are going to try to use the board, microcontroller, and chassis from the EE101 robot. We believe this will speed up the building process and ordering overhead of the project. We are going to replace the original DC servos with DC motors in order to increase the robot�s speed. The motors we choose will fit well in the EE101 chassis with room for the H-bridge and the digital gyroscope. The sensors will also be easy to mount on the original frame. Any questions we have will get hammered out within the next week and hopefully we will start putting it together by the next week.

The Algorithm:

The maze-solving algorithm is currently a work in progress. We made a simulator with a virtual maze and robot to help develop the algorithm. The robot can be moved around the maze and can simulate different sensors. With this tool, the algorithm can be developed completely without a running robot. Then once the robot is built and running, we can port the algorithm to the microcontroller with a minimum of difficulty. This will help with our limited schedule. A number of different algorithms are currently being discussed, including A* search. We intend to read some papers, implement different algorithms and test them in our simulator until the best one is found.

The Encoders:



There is an encoder on each wheel of the robot which can be used to tell distance traveled. Each wheel has a encoder wheel glued to it which consists of 38 white and black stripes. Each stripe cooresponds to .5cm of movement, so when 36 stripes have been passed, the robot has moved into the next square of the maze (each square is 18cm across). The stripes are detected with a QRB1134 phototransistor.

Back to Main