Racing Drones: Does Practice Make You Faster?

Prepare the Experiment

1. Choose an open space where you can safely fly a drone.

 
Figure 3. Examples of open space where it is safe to fly a drone.

2. Plan the obstacle course. Think of obstacles to fly your drone around or through.
   • Make it difficult enough that you have fun trying something new, but do not make it too difficult, as you do not want to crash the drone! The easiest obstacle course would be flying the drone out to a certain landmark and back for a set number of times (Figure 4). Figure 5 shows an example of a very challenging obstacle course.
   • Whatever your obstacle course looks like, flying the course should take about 30–60 seconds. A shorter obstacle course might make the small inaccuracies that are introduced when starting and stopping the stopwatch look huge, and longer courses will drain the battery too fast, which might limit the number of trials you can perform.
   Therefore, you will need create an obstacle course that you think will fit the time frame and is not too simple or too challenging. Once you start, you can adjust if you see your course choices will not work.

Figure 4. An example of an easy drone obstacle course.

Figure 5. An example of a very difficult drone obstacle course.

3. This project describes how to collect data to answer the question: "Does practice result in faster completion of the obstacle course and/or one with fewer mishaps?" The Variations section lists a few variations that can be tested with a similar setup and procedure. If you are interested in one of the variations, remember that you might need to adjust the procedure and data table slightly.
4. In your lab notebook, make a data table like Table 1. It will help you organize your data. If you plan on testing several pilots, or several obstacle courses, make a table for each pilot and each obstacle course.

Collect Data

1. Before you head out to collect data, make sure the battery of the drone is fully charged. Charge and bring extra batteries if you have any. Bring your drone, lab notebook, pencil, stopwatch, obstacles (if needed), and helper to the test site.
2. Some weather conditions, such as gusty or strong wind, make flying a drone more difficult. Make sure all your trials are performed under similar weather conditions.
3. Set up your course, if needed, and explain it to your helper. Do not forget to indicate the take-off and finish area.
4. Ask your helper to perform the following three tasks for each trial:
   • Start the stopwatch when the drone takes off.
   • Be sure the drone follows the course, and count any mishaps (like touching the obstacle, flying under an obstacle instead of over, flying around a hoop instead of through it, etc.).
   • Stop the stopwatch when the drone flies over the finish area.
5. Perform your first trial. If you decide this course is not working, change it and start again with a new first trial for the new course. Do not change the course while you are collecting data on a course; trials 1–10 should all be from the same course and with the same pilot.
6. After each trial, note down the number of mishaps, the time it took to finish the course, and any comments that might be useful. When you crash the drone in flight, mark in the comments what obstacle led to the crash. Write "unfinished" as the time and fill in the number of mishaps, including the crash.
7. Continue with your next trial.
8. Repeat steps 6 and 7 until you have performed 10 trials, or until the drone is out of battery.

Analyze Data

1. Look at the data you collected; do you notice a trend in the data?
2. Graph your data. Make two graphs:
   • Number of mishaps versus trial number. This graph has the trial number on the x-axis (horizontal) and the number of mishaps on the y-axis (vertical).
   • Time required versus trial number. This graph has the trial number on the x-axis (horizontal) and the time it took to finish the course on the y-axis (vertical).
3. Looking at the data and the graphs, can you conclude that the number of mishaps increases or decreases as you get more practice? Does practice result in faster or slower completion of the obstacle course? Does the change happen in the first few trials, then level out, or do you see a similar change each time? Maybe your data is inconclusive, or does your data show that practice does not alter your performance to finish this course?
4. Optional: Consider exploring if more mishaps happen when you fly the drone faster. Do you see a trend when you graph the number of mishaps versus the time it takes to finish the course? Are the first few trials clustered in one area of this graph and the last few in another? If so, what does that mean? Can the data you have collected help you find out the answer to these questions? If not, what data would you need to find out?
5. What other questions do you have? Maybe you wonder if your conclusions are valid regardless of the difficulty level of the obstacle course. If so, what other data would you need to answer this question? If you are curious whether your conclusions are valid for most pilots, can you extend your experiment to include data that would allow you to answer this question? To protect the drone, make sure to keep the obstacle course is easy for inexperienced pilots!