AT209 Lab 5 Wind Speed vs Battery Life Measurement

In this lab, we used Mavic 2 Pro drones at different altitudes to conduct measurements on wind speed by comparing battery life of the drones hovering in place.

Methods:

The first step in this lab was to gather our supplies. The class members in lab got partners, and each partner group got a Mavic 2 Pro drone, landing pad, and two or more fully charged batteries from the Purdue Dispatch Center. We then headed to coordinates 40.44364180047488, -87.03272783607103. We set up the drones with landing pads in a row, spaced 5 meters apart. We each had different assigned altitudes so we lined up in order of increasing altitude, south to north. The altitudes for all of the partner groups were 3m, 25m, 50m, 75m, 100m, and 121m. My partner and I flew at 75m AGL. We set up and pre-flighted the drones on one battery, and when everyone was ready, we switched the batteries to a new fully charged one and started up at the same time. We then all took off and flew to our assigned altitudes while facing the drones to the East. When everyone was at their altitude everyone gave a thumbs up, and someone from our class said "Start!", and we all recorded what our battery percentage was at the start of the timer, along with how many charge cycles the battery has been through. Our aircraft continued to hover there. Once the first drone out of the class got to 25% battery, the member flying the lowest battery drone called out for all of the groups to stop, and the rest of the groups recorded how much battery they had at the end of the time. Our lab section's time period was 18 minutes and 01 seconds. We could then compare starting and ending battery percentages. I was part of the Tuesday Lab group, and this procedure was repeated with the Thursday Lab group.

After collecting all of the battery data, we also took data from the weather reporting station on the property of the test site. This data reported the wind speeds at 3m above the ground every 30 minutes. We were able to use this data to calculate an estimated wind speed at our different altitudes from the formula: V2= V1 * (ln(h2/Z0)/ln(h1/Z0)). We were then able to use Microsoft Excel to graph these data sets, having altitude on the x-axis and one graph with battery usage, and another with wind speed on the y-axis.

Results:

These were the data sets that we graphed from the data collected. The first two are a representation of what the wind speed was on Tuesday at each height. Each line on the left graphs represent the winds at different times, reported every 30 minutes by the weather station. The graphs on the right show the amount of battery percentage used to maintain hover throughout the time period. Thursday had two separate data measurements, which is why there are two lines on the Thursday graph. Setting these graphs up in this way allows to easily compare of the wind compared to the battery usage at different altitudes.

Discussion:

Before the lab was begun, it was a common belief that the trend of the wind speed would match the trend of the battery usage, meaning that the more wind there was, the more battery the drone would use to stay stationary over the ground. These results do not agree with our hypothesis. The graph from Tuesday's battery usage shows the opposite, that more battery was used at lower altitudes rather than at higher, windier altitudes. There are a few extraneous factors for this experiment. This list includes not knowing what the real winds were, not having enough data, the possibility of turbulent air near the surface, and battery health.

The wind speed data was measured at the surface, but calculated for the speed at different altitudes. This is making an assumption that the atmosphere is behaving exactly as it should, which is not usually the case. The winds at altitude could have been much different than what we predicted them to be.

We also collected a relatively small about of data points. There were only 2 days where we measured data and both for roughly a span of 15-20 minutes. For an experiment, this is not a lot of data. We could have gotten very different data by measuring battery life and wind on different days, so it would be better to get more measurements to have a more accurate average.

The turbulence of air close to the ground could have also affected battery life. This turbulence would be caused by obstacles like the trees surrounding the location. It is uncertain if the drone would use more battery as it fights this turbulence.

Battery health could have also affected the results of the battery percentage used. An older battery that has had many life cycles would not be as efficient with its life span as a newer battery would.

All of this being said, the results from the study were inconclusive as the data comparison did not have a clear correlation and there are many extraneous factors affecting the data measured.