*7.1. Validation of Mixed Reality Simulation*

In this paper, for the MR simulation of the quadcopter, three flight tasks are performed. Figures 17–19 represent the real-time flight trajectories of the real quadcopter in Mission Planner (the left side) and the real-time flight trajectories of the virtual quadcopter in X-Plane (the right side).

**Figure 17.** Flight Task 1, flight trajectory of the real (**left**) and the virtual quadcopters (**right**).

**Figure 18.** Flight Task 2, flight trajectory of the real (**left**) and the virtual quadcopters (**right**).

**Figure 19.** Flight Task 3, flight trajectory of the real (**left**) and the virtual quadcopters (**right**).

It can be seen from Figures 10–12 that the flight trajectories from Mission Planner (real quadcopter) and X-Plane (virtual quadcopter) are similar, which means that the virtual quadcopter followed the real quadcopter. Note that in the real and virtual trajectories, there are some micro differences, due to tiny differences in the real and the virtual locations and these kinds of micro differences can be neglect in the simulation field. To validate the simulation, it is necessary to be more precise, thus, one should go through the data analysis using the Pixhawk's data log file and the X-Plane's data file.

The Mission Planner and X-Plane softwares have a data acquisition system that can record a wide range of parameters. The latitude, longitude, altitude, pitch angle, roll angle, and heading data are considered for data comparison. The data comparison procedure is shown in Figure 20. From the X-Plane's data file, we can get flight data of the virtual quadcopter. In the case of the real quadcopter, we can get the flight data log file from Mission Planner or Pixhawk's micro SD card in the real quadcopter.

**Figure 20.** Data comparison procedure.

Figures 21–26 illustrate the flight data comparison of the real quadcopter and the virtual quadcopter. Here, we compared the flight data from Flight Task 1. The latitude, longitude, altitude, pitch angle, roll angle, and the heading comparison of the real and virtual quadcopters from Flight Task 1 are shown in Figures 21–26, respectively. In Figures 21–26, the continuous line indicates the real quadcopter's data, and the broken line represents the virtual quadcopter's data.

**Figure 21.** Latitude comparison of the real and the virtual quadcopters.

**Figure 22.** Longitude comparison of the real and the virtual quadcopters.

**Figure 24.** Pitch angle comparison of the real and the virtual quadcopters.

**Figure 25.** Roll angle comparison of the real and the virtual quadcopters.

**Figure 26.** Heading comparison of the real and the virtual quadcopters.

It is crystal clear from Figures 21–26 that the real and the virtual quadcopters flight data are almost the same at every point of Flight Task 1. A comparison of the results shows that the latitude, longitude, and altitude of both quadcopters are nearly the same, which corroborates that the positions of both quadcopters are almost the same. The attitude comparison results (pitch angle, roll angle, and heading) of both quadcopters are also approximately equal. From the flight data comparison, we observed that on average a 400 ms time delay occurred in this MR simulation. Note that, if we observe the comparison results very precisely, we can see very tiny differences in the real and the virtual quadcopters' data which are due to the small time-delay. In the simulation field, we can neglect the small time-delay, and the tiny differences based on our allowable bandwidth.

Figure 27 shows the MR simulation of the quadcopter in real time (Flight Task 1). The stages 1 to 10 denoted (see Figure 27) that the full flight of the quadcopter (from take-off (1) to landing (10)) with precision close to a prescribed target was met. From Figure 27, we can see, by using our proposed architecture, that the virtual quadcopter in X-Plane (V series in the figure) followed the real quadcopter (R series in the figure) in real time. Therefore, here, the virtual quadcopter interacted with the real quadcopter in real time, and the MR simulation was carried out. In Figure 27, for the virtual part, we used Still Spot view in X-Plane, note that we could use different views on X-Plane including Chase, Panel, Circle, Free-Camera, and Linear Spot.

**Figure 27.** Mixed reality simulation of the quadcopter (R series = real part and V series = virtual part).

The comparison results from Figures 21–27 and the video in Supplementary Materials show that the performances of both quadcopters are almost identical in each and every point in the prescribed trajectory. Thereby, we could establish, herein, that using our developed architecture, the virtual quadcopter has interacted and followed the real quadcopter in real time. It proves conclusively that the MR simulation of a quadcopter is successfully achieved and validated.

By using our proposed MR simulation technique, we could see the clear view and the real-time performance of the real quadcopter on the simulation platform (X-Plane), which could possibly solve the visibility problems during a long-time mission.
