A tool for creating flight plan is needed to monitor forest fires on mountain trail through UAV. Existing tools for crating lfight plan for forest fires monitoring are created in a file format applicable to Ground Control Station. Additional time and manpower is consumed to apply the flight plan to the UAV. This paper presents AFPC(Automatic Flight Plan Creation), a tool that can automatically apply flight plan of UAV through digital twin. AFPC uses digital twins, so it can be easily expanded to enable UAV state monitoring and prediction using artificial intelligence. The evaluation of the AFPC compares the time and resources spent applying the flight plan created by the tool to UAVs with the related work.
A forest fire that fails to respond in the initial stage grows
into a large forest fire using surrounding leaves and
vegetation as fuel. Since large-scale forest fire cause great
damage to surrounding private houses and nature, early
detection and response is important. UAVs are suitable
for forest fire surveillance because they can
reconnaissance over large terrain. Fig. 1 shows the analysis by
the Korea Forest Service of the causes of forest fires that
occurred in Korea from 2012 to 2021 [
Based on this, the existing paper presented a tool for automatically generating a flight plan based on a mountain trail for forest fire monitoring through UAV. However, since existing tool creates a file in the format that can be used as the input to the GCS (Ground Control Station), time and resource are required to apply flight plans to UAVs.
This paper presents AFPC (Automatic Flight Plan
Creation), a tool that can automatically apply flight plan of
UAV through digital twin. AFPC combines digital twin
with existing flight plan creation tool, allowing UAVs to
perform automatically created flight plan. AFPC uses
digital twins, so it can be easily expanded to enable UAV
state monitoring and prediction using artificial
intelligence. The digital twin used in AFPC was constructed by
adding a DB module to the Ditto [
An Unmanned Aerial Vehicle (UAV) is an aircraft that
does not have a pilot on board and has a wide range of
activities. UAVs consists of various hardware devices
necessary for flight and software for automatic flight and
state control. Due to their ability to provide eficient
infrastructure and a wide range of services, UAVs are
increasingly being recognized as a major component of
the next generation of smart cities [
UAVs can be used for reconnaissance missions for forest fire monitoring. Since large-scale forest fires cause property and personnel injury, it is important to detect and respond to forest fires early through UAVs. UAVs can perform reconnaissance missions for forest fire detection through devices attached to UAVs while traversing the set flight plan. The flight plan of the UAV can be set using the flight plan creation tool. The created flight plan can be applied through GCS. GCS is software for monitoring and controlling UAVs and includes the function to apply lfight plans to UAVs.
Digital Twin was first introduced by Grieves [
The digital twin framework provides functions such
as data communication and object management. it can
enhance the reusability, compatibility, and
maintainability of digital twins. Ditto [
However, Ditto has a limitation of registering up to a maximum of 511 objects and the communicable data size is limited to 100KB. In addition, since Ditto does not manage past data, the Ditto used in this paper is used in conjunction with an additional DB module.
Ditana et al. [
Han et al. [
This paper presents AFPC (Automatic Flight Plan
Creation), a tool that can automatically apply flight plans
for forest fire surveillance to UAVs. AFPC directly
transmits flight plan information to the digital twin instead
of creating a flight plan file that can be applied to GCS
by applying digital twin to existing flight plan creation
tools. The flight plan stored on the digital twin then
allows UAVs to automatically read. The digital twin used
in AFPC was constructed by adding a DB module to the
Ditto [
Mountain Trail through APFC. 2. Second. The assigned flight plan is transmitted to the digital twin. 3. Third. The UAV reads the flight plan stored in the digital twin through the Raspberry Pi and applies it to the UAV system.
AFPC not only applies the flight plan in the environment to UAV, but also can collect UAV’s state data. This allows users to periodically manage the UAV through the digital twin included in AFPC. For example, users can verify whether the UAV is flying the flight path correctly and predict and handle errors that may occur in the UAV in advance. The method of monitoring UAV through AFPC is as follows: 1. First. The state of the UAV is transmitted to the
Raspberry Pi. 2. Second. The Raspberry Pi transmits the state of the UAV to the digital twin. 3. Third. The digital twin analyzes and responds to the current state of the UAV through user-defined algorithms. 4. Four. The Raspberry Pi applies the UAV control commands received from the digital twin to the actual UAV.
In additionally, the flight plan creation in previous work involved randomly creating waypoints along a specific path. In contrast, APFC applies an algorithm that reduces the number of randomly created waypoints based on distance. As a result, flight plans created by APFC require fewer waypoints to reach their destination compared to existing creation tool.
The environment used for the implementation and experiments of AFPC consisted of a PC with Windows 10 operating system, CPU Intel i7-11700, and 32GB RAM. Raspberry Pi 4B module was used as the onboard system of the UAV. In addition, Ditto 2.2V was used, which was built in a Docker environment.
The evaluation of AFPC compares the time and resources spent applying the flight plan to UAVs with related work. (a) in Fig. 3 compares the size of the flight plan file created by AFPC with the size of the flight plan ifle created by the related work. As a result of the comparison, AFPC created a file of about 4.2KB size when it created a flight plan containing about 110 waypoints, while related work created a file of about 33.8KB size. When using AFPC to create flight plans, flight plans can be applied to UAVs with approximately 8 times smaller data than related work.
(b) in Fig. 3 compares the time it takes to create a flight plan and apply the created flight plan to the UAV. As a result of the comparison, both AFPC and related work took approximately 0.2 seconds to create a flight plan. However, applying the created flight plan to UAVs took approximately 0.65 seconds for AFPC and approximately 5 seconds for related work. In summary, APFC could reduce time and resource consumption about 6-8 times compared to related work.
Additionally, We evaluate the AFPC through
simulation that the UAV flies according to the flight plan
crated by the AFPC. The simulation is performed in the
Gazebo [
Reconnaissance missions for forest fire surveillance using UAVs are important for early detection and response to forest fires. For forest fire detection, tool is needed to create flight plans for UAVs based on mountain trails. However, existing tools require application to UAVs via GCS, which requires additional time and manpower. This paper presents an AFPC that combines digital twins to automatically apply flight plan to UAVs. As a result of evaluating AFPC, APFC can reduce the time and resource consumption of about 6 to 8 times for applying flight plan to UAVs compared to related work. The simulation results evaluated that the UAV in the simulation flies correctly according to the flight plan applied through APFC.
This study was carried out with the support of ‘R&D Program for Forest Science Technology (Project No. 2021344A00-2223-CD01) provided by Korea Forest Service(Korea Forestry Promotion Institute). Following are results of a study on the “Leaders in Industry-university Cooperation 3.0” Project, supported by the Ministry of Education and National Research Foundation of Korea (No. 202209720001)