Fig 1.
Structure of GeoSOT global subdivision grids.
A dichotomy algorithm is applied in the three directions of longitude, latitude and altitude. The binary digit in the code is assigned 1 if the original data is higher than the mean value at the corresponding level; otherwise, the binary digit in the code is assigned 0. The 1st-level grid encoding order is shown in subfigure (a). The grids at each level below the 1st level should be encoded based on the upper-layer grid codes. Specifically, in the height direction, the encoding order extends from the lower level to the upper level, and the Z-order is adopted to continue encoding at the same level. The direction of Z-order coding is determined by the locations of the 1st-level grids. An example of the encoding order is shown in subfigure (b).
Table 1.
Air route type setting.
Fig 2.
Construction procedure of one air route by grids.
Fig 3.
Schematic of 5G signal coverage along a UAV air route.
The spatial coverage of the 5G signal is represented by massive regular grids. The size and position of each grid are defined according to the GeoSOT-3D coding schema. The air routes of UAVs are also constructed by consecutive grids. Blue grids demonstrate that the signal strength in the grids is strong enough to satisfy the flight requirements of UAVs. Green grids indicate that the 5G signal is weak, and there may be a safety risk when UAVs fly in these grids.
Fig 4.
UAV trajectory planning process in 5G signal field environment.
Fig 5.
Data size and number of track records at different subdivision levels.
Data size refers to the storage size of the airspace, 5G and UAV information under the same airspace using different levels of GeoSOT-3D grids and coordinates methods. The number of track records indicates the number of UAV tracks stored in the airspace.
Fig 6.
Time consumption for air route trafficability computation.
Time consumption represents the average time it takes for the UAV to search its neighborhood with or without a large database supporting index.
Fig 7.
Time and storage consumption for path planning.
In an identical airspace, with the same starting and ending points, this paper uses different environmental modeling methods to plan a UAV’s trajectory under the 5G field. The path planning time refers to the computing time required for a UAV to use the improved A* algorithm to obtain the optimal path. The storage number of the path grid refers to the number of trajectory grids generated in different environment modeling situations.
Table 2.
Airspace performance comparison between coordinates and GeoSOT-3D at the 20th level.