Cogs 300 2008W Term 2: Lab #2 (External Representation)

Group members:-
Buğra Fırat
Rachel Teo
Theo Rosenfeld
Wendy Lin

For this lab, the task was to build and program a robotic vehicle that would navigate the maze using light sensors to detect walls as well as the endzone (the large rectangular region marked out in white tape, at the end of the maze). No internal representation of the map (e.g. a hard-coded path) was allowed in the robot's program. The robot also needed to be able to recognise the white endzone while also not mistaking the white lines scattered through the maze for the endzone, i.e. it had to do something more sophisticated than simply treating all white floors as though they were the endzone.

This video is NOT of our final, official run. Instead, this is a "for fun" run we did after the fact. Our "real" robot followed the right-hand wall, whereas for this run, we tweaked it to follow the left-hand wall instead. It made a significant difference, as our robot's algorithm wasn't able to negotiate some tricky parts of the left-hand wall (most notably, the small black triangles which it ran over; running over walls instead of avoiding them would require a restart of the run, if it were being officially recorded).

The robot's victory dance consisted of beeping Dschinghis Khan's "Moskau" and dancing to it!

-- Video recorded by Rachel Teo. --

Category:

Science & Technology

Tags:

• cogs
• 300
• robot
• external
• representation