Test Plan - lukehsiao/RobotSoccer GitHub Wiki
Soccer players are judged and analyzed on criteria such as speed, dribbling ability, shot accuracy, and endurance. For our robot soccer competition, speed is is a critical metric that can give our robot a significant advantage in the competition. A fast robot will allow us to exploit the inherent reaction latency associated with artificial intelligence systems.
From our Functional Specification Document, we already identified a metric to measure this:
| Metric # | Req. # | Metric | Ideal | Margin | Units |
|---|---|---|---|---|---|
| 7 | 7,6 | Maximum speed of robot in one direction | >200 | >100 | cm/s |
For this test, we will measure the maximum speed of the robot when it is traveling in a single direction. This is similar to the standard 40-yard dash statistics used by sports leagues worldwide.
- Robot
- Workstation Computer with Vision Algorithm running and broadcasting location data to the Robot.
- Ceiling-mounted camera to record a real-time video, from which speed can be calculated
Like the 40-yard dash, we will measure the robot's maximum speed going in the forward direction. We will use tape to mark a 1 meter distance on the soccer field underneath the ceiling-mounted webcam. Then, we will begin recording video through the webcam. By using a simple Python script, we will command the robot to move straight forward at maximum speed, and then stop it after it passes through the marked meter and stop the recording.
Then, we can use the video recording to measure the precise time it took for the robot to travel a meter of distance, and calculate its speed using the formula:
speed = distance/time.
We can then repeat the process several times and determine the average.
We expect our robot to perform just under 200cm/s ideal due to our experience with watching it move. However, we've ever performed an actual measurement.
| Trial # | Speed |
|---|---|
| 1 | 101.52cm/s |
| 2 | 101.63cm/s |
| 3 | 101.52cm/s |
| 4 | 101.73cm/s |
| 5 | 101.42cm/s |
The statistical data from our trials is shown below:
- Mean: 101.56 cm/s
- Standard Deviation: 0.12 cm/s
We noted that our maximum speed of 101.56 cm/s was just barely above our marginal value, and only half of our ideal value. Although we meet our minimum specifications, we have realized through doing this test that to really have a competitive edge in the competition, we may need to be able to increase the speed of our robot.
Because we have not had actual competitions to compare speeds with our competitors robots, we will not know for sure until after our competition. If our opponents speed is still slow in comparison to our own, we can consider this metric met. Otherwise, we will have to identify ways to improve the speed of our robot.