lets write a little background information about software of ARIS

Differential Drive

In order to move around the course, we decided to use a two motor set up. Because of this, we were able to use differential drive for movement. This means that in order to turn, we count the number of encoder ticks per revolution, and use those ticks to measure a 90degree turn. With these numbers, we can multiple them by two in order to rotate 180degrees. When turning left, we lock the inside wheel and turn only the outside wheel, or right wheel. To turn right, we do the opposite. These movements are shown in the images above.

Course Mapping

Our idea for navigation of the course was to implement a course mapping technique. We divided each room into different sections, as shown in the figure above. These sections were based off of the number of turns needed to reach each checkpoint. Because this is a timed event, we goal was to cover the farthest distance with the least amount of turns and in the quickest time. Not only were the sections based off movements, but also on candle/decoy placement. If the decoy were placed in section A, then we would like A.R.I.S. to see the flame from the doorway. D would be the smallest section, therefore if the decoy were in D, there would be a small area to cover, therefore we'd like to do it in one sweep and move on to the next checkpoint.

We did not section the hall way off because we figured it wasn't necessary. If A.R.I.S. were in the hallway, the IR signature should be picked up with our 3-Tier IR detection system, regardless of position within the hallway.

This next image shows the possible routes of each room. Dependent on the identified location of the candle/decoy from the previous trial run, A.R.I.S. would know the fastest route to take. These routes are preprogrammed into A.R.I.S. and it is able to identify which route is fastest to candle/decoy.

OpenCV

Decoy
Flame

One of the techniques we used for IR source recognition and navigation was OpenCV. A.R.I.S. is equipped with a NoIR Raspberry Pi Camera, which detects an IR source, and using the data feedback, A.R.I.S. is able to drive to the source. As you can see from the images above, the IR decoy light is not constant, where as the flame is. The IR decoy has a random sampling rate.