KR20200008281A - Apparatus for following lane on road by unmanned aerial vehicle and method the same - Google Patents

Abstract

The present invention discloses an apparatus for following a road lane of an unmanned air vehicle and a method thereof. The apparatus enabling an unmanned air vehicle for autonomous flight by following a road lane comprises: a photographing unit photographing a road from the bottom of the unmanned air vehicle; a GPS module measuring a flight trajectory and altitude of the unmanned air vehicle; a posture measuring unit measuring a flight posture of the unmanned air vehicle; a control unit extracting a photographed image inputted from the photographing unit, the flight trajectory and altitude of the unmanned air vehicle inputted from the GPS module, and a road lane based on a posture angle inputted from the posture measuring unit, converting the extracted road lane into NED coordinates, and calculating and outputting a speed command to follow a waypoint based on the converted road lane; and a drive unit driving the unmanned air vehicle to follow the road lane according to the speed command of the control unit.

Description

Road lane tracking device of unmanned aerial vehicle and its method {APPARATUS FOR FOLLOWING LANE ON ROAD BY UNMANNED AERIAL VEHICLE AND METHOD THE SAME}

The present invention relates to a road lane tracking device and a method thereof of an unmanned aerial vehicle, and more particularly, to detect a location of a road lane from a photographed image to track a path based on a lane of a road, and to determine a smooth curved path. The present invention relates to a road lane tracking device for an unmanned aerial vehicle and a method thereof.

Recently, unmanned aerial vehicles (UAVs) such as drones, quadcopters, and helicams equipped with cameras have been commercialized, and are being used in various fields such as camera photography.

These drones are moving beyond the broadcasting and military fields, which are mainly used in the past, to enter the era of general users using cameras for shooting cameras.

However, the unmanned aerial vehicle according to the prior art is inefficient in that it must be a user because it can be controlled wirelessly through a remote controller (RC) and a smart user terminal, and also by the user manually controlled from the ground due to the characteristics of the unmanned aerial vehicle In case of accidents, accidents frequently occur due to immaturity of the user, and thus, expensive equipment is damaged or causes safety accidents.

Therefore, the unmanned aerial vehicle can be detected autonomously and automatically tracked without the user's control on the ground, so that it can be applied to a realistic and applicable unmanned aerial vehicle that can prevent safety accidents and expensive drones from being damaged. Technology is desperately needed.

Background art of the present invention is disclosed in Republic of Korea Patent Publication No. 2017-0022872 (published on March 2, 2017, unmanned aerial vehicle having an automatic tracking function).

The present invention has been made in view of the necessity as described above, an object of the present invention according to an aspect is to detect the position of the road lane from the captured image to follow the path based on the lane of the road to determine the position, smooth curve path It is to provide a road lane tracking device and a method of the unmanned aerial vehicle to follow the.

Road lane tracking device of the unmanned aerial vehicle according to an aspect of the present invention, the shooting unit for photographing the road from the lower portion of the unmanned aerial vehicle; GPS module for measuring the flight trajectory and altitude of the unmanned aerial vehicle; A posture measuring unit measuring a flight attitude of the unmanned aerial vehicle; Based on the captured image input from the photographing unit, the flight trajectory and altitude of the unmanned aerial vehicle input from the GPS module, and the attitude angle input from the attitude measuring unit, the lane of the road is extracted, and the extracted lane is converted into a NED coordinate system. A control unit for calculating and outputting a speed command to follow a path point based on the converted lane; And a driving unit which drives the unmanned aerial vehicle to follow the lane of the road according to the speed command of the controller.

In the present invention, the photographing unit is equipped with an ultra wide-angle fisheye lens having a field of view (FOV) of 180 degrees.

In the present invention, the controller extracts the line segment from the photographed image and then performs curve fitting based on the points of the line segment to extract the lane of the road.

In the present invention, after performing the curve fitting, the controller extracts the lane by performing the second curve fitting based on the points of the line segment extracted within the set ROI.

Curve fitting in the present invention is characterized by performing based on the RANSAC (RANdom SAmple Consensus) algorithm.

In the present invention, the ROI is characterized by being set to a different size according to the altitude of the unmanned aerial vehicle in a rectangle including one lane.

In the present invention, the control unit, by applying the scale factor according to the attitude angle of the unmanned aerial vehicle input from the attitude measurement unit and the altitude of the unmanned aerial vehicle input from the GPS module to the selected points in front of the extracted lane based on Equation 1 below And converting the position to a coordinate system position.

[Equation 1]

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the attitude angle of the image pickup unit, and s is the scale factor between the image position of the image and the position of the actual NED coordinate system. scale factor).

In the present invention, the controller generates a curved path based on three path points for following the curved lane and an incident angle at which the unmanned aerial vehicle enters the path point, and generates a normal speed command and a tangential speed command based on the curved path. It is characterized by generating a speed command by the sum of.

In the present invention, the incident angle is characterized in that the control unit calculates from the flight trajectory of the unmanned aerial vehicle input from the GPS module.

In the present invention, the tangential speed command is represented by the product of the tangential direction vector by the magnitude of the speed that the unmanned aerial vehicle is going to go, and the normal speed command is a projection of the t-axis error between the unmanned aerial vehicle and the path with the normal vector. The error obtained by multiplying Pgain (Kp) is characterized by the above-mentioned.

According to another aspect of the present invention, there is provided a road lane tracking method of an unmanned aerial vehicle, including: receiving, by a controller, a photographed image photographed from a photographing unit mounted on a lower portion of an unmanned aerial vehicle flying on a road; Extracting, by the controller, the line segment from the captured image; Extracting a lane by performing curve fitting based on the points of the line segment extracted by the controller; Converting the extracted lane into an NED coordinate system; And generating, by the controller, the speed command based on the path points converted into the NED coordinate system.

In the present invention, the photographed image is an image photographed by an ultra wide-angle fisheye lens having a viewing angle of 180 degrees.

The present invention includes the steps of extracting a line segment within a region of interest set after the controller performs a curve fitting; And extracting a lane by performing a quadratic curve fitting based on the points of the line segment extracted in the region of interest.

Curve fitting in the present invention is characterized by performing based on the RANSAC (RANdom SAmple Consensus) algorithm.

In the present invention, the ROI is characterized by being set to a different size according to the altitude of the unmanned aerial vehicle in a rectangle including one lane.

In the present invention, the step of converting to the NED coordinate system, by applying a scale factor according to the attitude angle of the unmanned aerial vehicle input from the attitude measurement unit and the altitude of the unmanned aerial vehicle input from the GPS module with respect to selected points in front of the extracted lane. It is characterized by converting to the position of the NED coordinate system based on Equation 2 below.

[Equation 2]

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the attitude angle of the image pickup unit, and s is the scale factor between the image position of the image and the position of the actual NED coordinate system. scale factor).

In the present invention, the step of generating a speed command, when the controller is a curved lane, generates a curved path based on the three path points for tracking and the angle of incidence that the unmanned aerial vehicle enters the path point, and the normal direction based on the curved path The speed command is generated by the sum of the speed command and the tangential speed command.

In the present invention, the incident angle is characterized in that the control unit calculates from the flight trajectory of the unmanned aerial vehicle input from the GPS module.

In the present invention, the tangential speed command is represented by the product of the tangential direction vector by the magnitude of the speed that the unmanned aerial vehicle is going to go, and the normal speed command is a projection of the t-axis error between the unmanned aerial vehicle and the path with the normal vector. The error obtained by multiplying Pgain (Kp) is characterized by the above-mentioned.

In accordance with an aspect of the present invention, a road lane tracking device and a method thereof of an unmanned aerial vehicle may detect a location of a road lane from a captured image and follow a smooth curved path in order to follow a path based on a road lane. By doing so, you can follow the lane path and autonomously fly without invading the lane.

1 is a block diagram schematically illustrating a road lane tracking device of an unmanned aerial vehicle according to an exemplary embodiment of the present invention.
2 to 5 are images illustrating a process of extracting a lane in a road lane following device of an unmanned aerial vehicle according to an embodiment of the present invention.
6 is a view for explaining the conversion of the NED coordinate system in the road lane tracking device of the unmanned aerial vehicle according to an embodiment of the present invention.
7 is a view for explaining the following of the curved path in the road lane tracking device of the unmanned aerial vehicle according to an embodiment of the present invention.
8 is a flowchart illustrating a road lane following method of an unmanned aerial vehicle according to an exemplary embodiment of the present invention.

Hereinafter, a road lane tracking device and a method thereof of an unmanned aerial vehicle according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of the lines or the size of the components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described below are terms defined in consideration of functions in the present invention, which may vary depending on the intention or convention of a user or an operator. Therefore, the definitions of these terms should be made based on the contents throughout the specification.

1 is a block diagram schematically illustrating a road lane tracking device of an unmanned aerial vehicle according to an embodiment of the present invention, and FIGS. 2 to 5 are lanes of a road lane tracking device of an unmanned aerial vehicle according to an embodiment of the present invention. Figure 6 is an image showing a process of extracting, Figure 6 is a view for explaining the conversion of the NED coordinate system in the road lane tracking device of the unmanned aerial vehicle according to an embodiment of the present invention, Figure 7 according to an embodiment of the present invention A diagram for explaining the following of a curved path in a road lane following device of an unmanned aerial vehicle.

As shown in FIG. 1, a road lane tracking device of an unmanned aerial vehicle according to an exemplary embodiment of the present invention may include a photographing unit 10, a GPS module 40, a controller 20, and a driver 30.

The photographing unit 10 is equipped with an ultra-wide angle fisheye lens having a field of view (FOV) of 180 degrees, so that when the unmanned aerial vehicle is flying on the road, the photographing image captured on the road from the lower part of the unmanned aerial vehicle to the controller 20 is used. to provide.

The GPS module 40 measures the flight trajectory and the altitude of the unmanned aerial vehicle and provides the same to the controller 20.

Here, the flight trajectory of the unmanned aerial vehicle allows the control unit 20 to calculate the coefficient of the third order polynomial by reflecting the incident angle along with the three path points as boundary conditions when generating the curved path.

In addition, the altitude of the unmanned aerial vehicle may be applied to set a region of interest when the control unit 20 extracts a lane from a captured image, and a scale factor applied when converting coordinates extracted from the captured image into a NED coordinate system. Available to get).

The attitude measuring unit 50 measures the flight posture of the unmanned aerial vehicle and provides the same to the controller 20.

Here, the flight attitude of the unmanned aerial vehicle may be applied when the controller 20 calculates the attitude angle of the photographing unit 10 based on the flight attitude of the unmanned aerial vehicle and converts the coordinates in the captured image into the NED coordinate system.

The control unit 20 is based on the captured image input from the photographing unit 10, the flight trajectory and altitude of the unmanned aerial vehicle input from the GPS module 40, and the attitude angle input from the attitude measuring unit 50 The lanes are extracted, the extracted lanes are converted to the NED coordinate system, and the speed command is calculated and output to follow the path point based on the converted lanes so that they can be followed.

That is, the controller 20 extracts a line segment as shown in FIG. 2 to extract a line, which is a basic feature of the lane, from the input photographed image, and then uses a random SAmple consensus (RANSAC) algorithm based on the extracted points of the line segment. Curve fitting may be performed to extract lanes.

Thereafter, the controller 20 may set an ROI from the extracted lane and extract the line segment from the ROI as shown in FIG. 4.

Here, the ROI is a quadrangle including one lane and may be set to different sizes according to the altitude of the unmanned aerial vehicle.

After extracting the line segment in the region of interest as described above, the controller 20 performs a second curve fitting based on the points of the line segment extracted in the region of interest, extracts the lane as shown in FIG. 5, and extracts the extracted lane as shown in FIG. 6. Can be converted to NED coordinate system (absolute position).

For example, the controller 20 extracts three points in front of the lane to convert to the NED coordinate system, and then acquires left and right three pairs of image information using the y-axis information of the three points, and Equation 1 We can calculate the position of the NED coordinate system for each point by substituting for.

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the pose angle of the imager, and s is the scale factor between the image position of the image and the position of the actual NED coordinate system. .

At this time, the attitude angle of the photographing unit can be obtained based on the flight attitude of the unmanned aerial vehicle inputted from the self-measuring unit 50, and the scale factor between the image position and the position of the actual NED coordinate system is input from the GPS module 40. Can be obtained by using the altitude of the unmanned aerial vehicle.

In this way, the path points to be followed by the unmanned aerial vehicle for the lane on the road may be converted into positions in the NED coordinate system.

Thereafter, the controller 20 generates a speed command for following the path points based on the points of the NED coordinate system and outputs the speed command to the driver 30 to follow the autonomous flight.

At this time, following a path point made in a lane made of a straight line may not invade the side lane even with a conventional point navigation flight, but in a curved lane, a lane may be invaded according to the distance between path points.

Therefore, in this embodiment, in order to prevent this, a curved path is generated to follow a smooth curved path using given path points, and based on this, the normal speed command (Vn, cmd) and the tangential speed command (Vt, cmd) are generated. The speed command (Vcmd) generated by the sum of) can be generated to follow the curve path.

For example, since at least three path points are required to generate the curved path, three pairs of lane positions converted in FIG. 6 may be used. As shown in FIG. 7, three points, path points P _i-1, P _{i and} P _{i + 1} , may be obtained through the average of the right and left points.

Thus, the curved path connecting the three points of the path can be defined by the third order polynomial d (s). Here, four boundary conditions are required to obtain all the coefficients of the third order polynomial. In the present embodiment, Equation 2 is based on the incidence angle of the unmanned aerial vehicle entering three path points and P _i-1 points. The coefficient of the difference polynomial can be determined.

Here, the incident angle of the unmanned aerial vehicle may be calculated based on the flight trajectory of the unmanned aerial vehicle input from the GPS module 40.

On the other hand, the tangential speed command (Vt, cmd) can be calculated and generated as shown in equation (3), the connection direction speed command (Vt, cmd) of the path is a direction vector in the tangential direction to the magnitude of the speed that the unmanned aerial vehicle wants to go It can be expressed as the product of.

In addition, the normal speed command (Vn, cmd) may be represented by multiplying the P gain (Kp) by the error obtained by projecting the t-axis error between the unmanned aerial vehicle and the path with the normal direction vector, as shown in equation (4).

Therefore, the control unit 20 violates the side lane by outputting the speed command Vcmd generated by the sum of the normal speed command (Vn, cmd) and the tangential speed command (Vt, cmd) as shown in Equation 5 on the curved path. This allows you to follow the curve path smoothly.

The driver 30 drives the unmanned aerial vehicle to fly on the road according to the speed command output from the controller 20.

Therefore, the control apparatus of the unmanned aerial vehicle extracts the lane from the photographed image on the road input from the photographing unit 10 and converts the lane to the NED coordinate system, and autonomously flies the path along the lane based on the converted lane. Not only does it drive the drive unit 30 by following the speed command so that it can follow, but also creates a curved path in the curved lane so that it does not invade the lane by the speed command by the sum of the tangential speed command and the normal speed command. Make sure you follow the path.

As described above, according to the road lane tracking device of the unmanned aerial vehicle according to the embodiment of the present invention, in order to follow the path based on the lane of the road, the lane of the road is detected from the photographed image to determine the position, and the smooth curve path By allowing to follow the lane path can be autonomous flight without invading the lane.

8 is a flowchart illustrating a road lane following method of an unmanned aerial vehicle according to an exemplary embodiment of the present invention.

As shown in FIG. 8, in the road lane tracking method of the unmanned aerial vehicle according to the exemplary embodiment of the present invention, first, the control unit 20 has an ultra-wide angle of 180 degrees from the photographing unit 10 mounted to the lower part of the unmanned aerial vehicle. The fisheye lens is adopted to receive the captured image (S10).

The controller 20 receiving the captured image in step S10 extracts a line segment as shown in FIG. 2 to extract a line which is a basic feature of the lane from the input captured image (S20).

After extracting the line segment in operation S20, the controller 20 extracts a lane by performing a curve fitting through a random SAmple consensus (RANSAC) algorithm based on the extracted points of the line segment (S30).

Since only a part of the lane is sampled through the curve fitting through the RANSAC algorithm, as shown in FIG. 3, the fitting information may deviate significantly from the far lane.

Therefore, the controller 20 sets the region of interest from the lane extracted in step S30 and extracts the line segment from the set region of interest as shown in FIG. 4 (S40).

Here, the ROI may be a rectangle including one lane and may be set to different sizes according to the altitude of the unmanned aerial vehicle input by the controller 20 from the GPS module 40.

After extracting the line segment in the ROI in operation S40, the controller 20 performs a second curve fitting based on points of the line segment extracted in the ROI to extract the lane as illustrated in FIG. 5 (S50).

After extracting the lane in step S50, the control unit 20 converts the extracted lane into a NED coordinate system (absolute position) as shown in FIG. 6 (S60).

For example, the control unit 20 extracts three points in front of the lane to convert to the NED coordinate system, and then obtains left and right three pairs of image information using the y-axis information of the three points, and Equation 6 We can calculate the position of the NED coordinate system for each point by substituting for.

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the pose angle of the imager, and s is the scale factor between the image position of the image and the position of the actual NED coordinate system. .

At this time, the attitude angle of the photographing unit 10 can be obtained based on the flight attitude of the unmanned aerial vehicle inputted from the self-measuring unit 50, and the scale factor between the image position and the position of the actual NED coordinate system is determined by the GPS module ( Can be obtained using the altitude of the unmanned aerial vehicle input from 40.

In this way, the path points to be followed by the unmanned aerial vehicle for the lane on the road may be converted into positions in the NED coordinate system.

After converting the lane to the position of the NED coordinate system in step S60, the control unit 20 generates a speed command for following the path points based on the path points converted into the NED coordinate system to follow the autonomous flight (S70). .

Following a path point made from a straight lane may not invade the next lane even with a conventional point navigation flight. However, in the case of a curved lane, a lane may be involved depending on the distance between the path points.

Therefore, in this embodiment, in order to prevent this, the sum of the normal speed command (Vn, cmd) and the tangential speed command (Vt, cmd) based on the curve path can be used to follow the smooth curve path using the given path points. Create a speed command (Vcmd) to be created to follow the curve path.

For example, since at least three path points are required to obtain a curved path, three pairs of lane positions converted in FIG. 6 may be used. As shown in FIG. 7, three points, path points P _i-1, P _{i and} P _{i + 1} , may be obtained through the average of the right and left points.

Thus, the curved path connecting the three points of the path can be defined by the third order polynomial d (s). Here, four boundary conditions are required to obtain all the coefficients of the third order polynomial. In this embodiment, three equations are obtained based on the incidence angle of the unmanned aerial vehicle entering three path points and P _i-1 points. The coefficient of the difference polynomial can be determined.

Here, the incident angle of the unmanned aerial vehicle may be calculated based on the flight trajectory of the unmanned aerial vehicle input from the GPS module 40.

On the other hand, the tangential speed command (Vt, cmd) can be calculated and generated as shown in Equation 8, the connection direction speed command (Vt, cmd) of the path is a direction vector in the tangential direction to the magnitude of the speed that the unmanned aerial vehicle is going to go It can be expressed as the product of.

In addition, the normal speed command (Vn, cmd) can be represented by multiplying the P gain (Kp) by the error obtained by projecting the t-axis error between the unmanned aerial vehicle and the path with the normal direction vector, as shown in equation (9).

Therefore, the control unit 20 violates the side lane by outputting the speed command Vcmd generated by the sum of the normal speed command Vn and cmd and the tangential speed command Vt and cmd on the curved path. This allows you to follow the curve path smoothly.

As described above, according to the road lane tracking method of the unmanned aerial vehicle according to the embodiment of the present invention, in order to follow the path based on the lane of the road, the lane of the road is detected from the photographed image to determine the position, and the smooth curve path By allowing to follow the lane path can be autonomous flight without invading the lane.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and those skilled in the art to which the art belongs can make various modifications and other equivalent embodiments therefrom. I will understand.

Therefore, the true technical protection scope of the present invention will be defined by the claims below.

10: recording unit 20: control unit
30: drive unit 40: GPS module
50: posture measuring unit

Claims (19)

Shooting unit for photographing the road from the bottom of the unmanned aerial vehicle;
A GPS module for measuring a flight trajectory and an altitude of the unmanned aerial vehicle;
A posture measuring unit measuring a flight attitude of the unmanned aerial vehicle;
A lane of a road is extracted based on a captured image input from the photographing unit, the flight trajectory and altitude of the unmanned aerial vehicle input from the GPS module, and an attitude angle input from the attitude measuring unit. A control unit for converting to a NED coordinate system and calculating and outputting a speed command to follow a path point based on the converted lane; And
And a driving unit for driving the unmanned aerial vehicle to follow a lane of a road according to the speed command of the control unit.
The apparatus of claim 1, wherein the photographing unit is equipped with an ultra wide angle fisheye lens having a field of view (FOV) of 180 degrees.
The apparatus of claim 1, wherein the controller extracts a line segment from the photographed image and extracts a lane of a road by performing curve fitting based on points of the line segment.
The road of claim 3, wherein the controller extracts a lane by performing a second curve fitting based on points of the line segment extracted within a region of interest set after performing the curve fitting. Lane following device.
5. The apparatus of claim 4, wherein the curve fitting is performed based on a random SAmple consensus (RANSAC) algorithm.
5. The apparatus of claim 4, wherein the ROI is a quadrangle including one lane and is set to a different size according to the altitude of the unmanned aerial vehicle. 6.
The scale factor of claim 1, wherein the controller is further configured to determine a scale factor according to the attitude angle of the unmanned aerial vehicle input from the attitude measuring unit and the altitude of the unmanned aerial vehicle input from the GPS module with respect to selected points forward from the extracted lane. The road lane following device of the unmanned aerial vehicle, characterized in that for converting to the position of the NED coordinate system based on Equation 1 below.
[Equation 1]

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the attitude angle of the image pickup unit, and s is the scale factor between the image position of the captured image and the actual NED coordinate system position. scale factor).
The method of claim 1, wherein the controller generates a curved path based on three path points for following a curved lane and an incident angle at which the unmanned aerial vehicle enters the path point, and generates a normal speed based on the curved path. And a speed command is generated by the sum of the command and the tangential speed command.
The apparatus of claim 8, wherein the incidence angle is calculated by the controller from the flight trajectory of the unmanned aerial vehicle input by the GPS module.
The method of claim 8, wherein the tangential speed command is expressed as a product of a tangential direction vector multiplied by a magnitude of a speed to which the unmanned aerial vehicle wants to go,
And wherein the normal speed command indicates a t-axis error between the unmanned aerial vehicle and a path by multiplying Pgain (Kp) by an error obtained by projecting a normal vector with a normal direction vector.
Receiving, by the control unit, a photographed image photographed from a photographing unit mounted on a lower portion of an unmanned aerial vehicle flying on a road;
Extracting, by the controller, a line segment from the captured image;
Extracting a lane by performing curve fitting based on the points of the line segment extracted by the controller;
Converting, by the controller, the extracted lane into a NED coordinate system; And
And generating, by the control unit, a speed command based on the path points converted into the NED coordinate system.
The method of claim 11, wherein the photographed image is an image photographed by an ultra wide-angle fisheye lens having a viewing angle of 180 degrees.
The method of claim 11, further comprising: extracting, by the controller, the line segment within a region of interest set after performing the curve fitting; And
And extracting a lane by performing a second curve fitting on the basis of the points of the line segment extracted in the region of interest.
15. The method of claim 13, wherein the curve fitting is performed based on a random SAmple consensus (RANSAC) algorithm.
The method of claim 13, wherein the ROI is a quadrangle including one lane and is set to a different size according to the altitude of the unmanned aerial vehicle.
The method of claim 11, wherein the converting into the NED coordinate system comprises: the attitude angle of the unmanned aerial vehicle input from the attitude measuring unit with respect to the selected points forward from the extracted lane, and the altitude of the unmanned aerial vehicle input from the GPS module. A road lane tracking method of an unmanned aerial vehicle, characterized by converting to a position of the NED coordinate system based on Equation 2 by applying a scale factor according to Equation 2 below.
[Equation 2]

Where u and v are distances from the center of the image, f (u, v) is the fisheye lens model function, R is the attitude angle of the image pickup unit, and s is the scale factor between the image position of the captured image and the actual NED coordinate system position. scale factor).
The method of claim 11, wherein the generating of the speed command comprises: generating a curved path based on three path points for following when the controller is a curved lane and an incident angle at which the unmanned aerial vehicle enters the path point; And generating the speed command based on the sum of the normal speed command and the tangential speed command based on a curved path.
18. The method of claim 17, wherein the incidence angle is calculated by the controller from a flight trajectory of the unmanned aerial vehicle input from a GPS module.
18. The method of claim 17, wherein the tangential speed command is expressed as a product of a tangential direction vector multiplied by a magnitude of a speed to which the unmanned aerial vehicle wants to go,
And said normal speed command indicates a t-axis error between the unmanned aerial vehicle and the path by multiplying Pgain (Kp) by an error obtained by projecting the normal vector with the normal direction vector.

Images