CN113759941B - Landing track control method for large-sized freight unmanned aerial vehicle - Google Patents

Abstract

The present invention discloses a landing trajectory control method for a large-scale cargo unmanned aerial vehicle. The method pre-designs the landing trajectory, establishes the geometric relationship of trajectory parameters, determines the trajectory parameters, establishes an aerodynamic model of the large-scale cargo unmanned aerial vehicle, combines the kinematics and dynamics equations of the large-scale cargo unmanned aerial vehicle, confirms the motion state of the large-scale cargo unmanned aerial vehicle and sets the landing trajectory, designs a longitudinal landing control equation, solves a bounded solution, utilizes feedback output, generates a state trajectory, optimizes the landing trajectory, and finally implements a landing trajectory detection and confirmation step to further confirm and verify the landing trajectory. The landing trajectory designed by the present invention has basically no deviation from the actual landing trajectory, and has important reference significance for the landing safety of large-scale cargo aircraft.

Description

A landing trajectory control method for large cargo drones

Technical Field

The invention belongs to the technical field of landing trajectory control of large cargo aircraft, and in particular relates to a landing trajectory control method for a large cargo unmanned aerial vehicle.

Background technique

During the landing phase of a large cargo UAV, as it decelerates and descends in altitude from the ground, its trajectory control is one of the important factors related to aircraft safety. The existing landing trajectory control method has limited trajectory accuracy. For large cargo aircraft, especially large cargo UAVs, which have a large weight difference before and after loading, and have high requirements for landing trajectories, the existing trajectory control methods often cannot meet their actual safety needs. In view of the above problems, the present invention provides a new landing trajectory control method to achieve higher trajectory accuracy and effectively improve the landing safety of large cargo UAVs.

Summary of the invention

The purpose of the present invention is to provide a landing trajectory control method for a large cargo UAV, which can achieve higher trajectory accuracy and effectively improve the landing safety of the large cargo UAV.

To achieve the above object, the present invention adopts the following technical solutions:

A large-scale cargo UAV landing trajectory control method, characterized in that the method comprises the following steps:

S1) Pre-design the three stages of the landing trajectory of large cargo drones: deep glide stage, pull-up stage and shallow glide stage;

S2) Determine the main trajectory parameters and geometric relationships. The main trajectory parameters include the deep glide angle, the shallow glide angle, the longitudinal height of the pull-up section, and the coordinates of the end point of the deep glide section. The geometric relationship reference height profile of the trajectory corresponding to the deep glide section, the pull-up section, and the shallow glide section is:

Deep glide section, _hs = _tanγ1 ( _x0 ), _γ1 is the deep glide angle, _x0 is the intersection of the deep glide section and the ground channel, pull-up section, x _E is the starting point of the pull-up segment, h _D is the height difference of the pull-up segment, σ is the attenuation rate of the pull-up segment, and for the shallow glide segment, h _q =tanγ ₂ (x _D ), γ ₂ is the shallow glide angle, and x _D is the coordinate of the touchdown point;

S3) parameters are determined, the deep glide angle is determined, the inclination change rate of the deep glide section is 0, and the reference lift coefficient can be obtained Determine the deep glide angle γ1 and the shallow glide angle to avoid exceeding the load limit due to too fast descent;

S4) Establish an aerodynamic model for large cargo UAVs, and the aerodynamic equations for lift coefficient and drag coefficient are:

_CL is the lift coefficient, _CD is the drag coefficient, and δ is the flap deflection angle;

S5) combining the kinematics and dynamics equations of the large cargo UAV, confirming the motion state of the large cargo UAV, and setting the landing trajectory according to the motion state of the large cargo UAV;

S6) Landing trajectory optimization: for the set landing trajectory, by designing the longitudinal landing control equation, solving the bounded solution, using the feedback output, generating the state trajectory, and realizing the optimization of the landing trajectory;

S7) Landing trajectory detection and confirmation: obtain sequence images through high-speed camera, obtain the coordinates of feature sites, and preliminarily locate the feature sites. Preprocess the images by setting the image pixel grayscale value, set the threshold, perform edge detection, use HOUGH transformation to further locate, calculate the change of feature sites, calculate the motion parameters of the large cargo UAV, further solve and obtain the various parameters of the actual motion trajectory of the large cargo UAV, and confirm the actual landing trajectory.

Furthermore, S1) the deep glide segment is a quasi-straight line trajectory, which starts at the end point of the level flight stage and ends at the starting point of the pull-up segment. In the pull-up segment, the longitudinal posture of the large cargo UAV is adjusted by adjusting the flap configuration, and the large cargo UAV changes from a head-down state to a head-up state and enters the shallow glide segment, which starts at the end point of the pull-up segment and ends at the touchdown point of the large cargo UAV.

Furthermore, S2) the deep glide angle is the angle between the extension line of the deep glide section trajectory and the horizontal channel, the shallow glide angle is the angle between the shallow glide section trajectory and the horizontal channel, and the longitudinal height of the pull-up section is the projection distance of the pull-up section trajectory of the large cargo drone on the vertical plane.

Furthermore, the dynamic equation described in S5) is:

m, v represent the mass and speed of the large cargo drone, γ represents the track inclination, h represents the altitude, x represents the horizontal position in the direction of the channel, and D and L represent the drag and lift, respectively.

Further, S6) the longitudinal landing control equation is:

y _d (t) = Cx _d (t), y _d (t) is the optimized output trajectory.

Furthermore, in the process of solving the bounded solution described in S6), for the landing trajectory of the large cargo drone, the output equation of its height and ground speed is defined as Determine the internal dynamic function ε(t) = [q, θ] ^T and solve the transformation matrix. For/> The variables in are used to obtain the linear transformation relationship of the original state variables, and the transformation matrix T is obtained. Then, through the transformation of the new and old coordinates, the bounded solution ε _d (t) is obtained, the internal dynamic equation is obtained, and then the optimized state trajectory is obtained.

Furthermore, the optimized state trajectory is:

Furthermore, the solution described in S7) is a method of utilizing camera data in combination with a mathematical model of landing parameters of a large cargo UAV to obtain an accurate solution of the spatial coordinates of feature points on each frame of the image, and then converting the spatial coordinates of the feature points to the runway coordinate system of the large cargo UAV to obtain high-precision coordinate information in the runway coordinate system, and outputting a trajectory curve based on the coordinate information of different feature points.

The present invention has the following beneficial effects:

The present invention pre-designs, confirms parameters, optimizes the trajectory, and detects and confirms the landing trajectory of a large cargo UAV, so that the designed landing trajectory is more consistent with the actual landing trajectory, and the actual application operability of the landing trajectory is stronger, which can effectively improve the landing safety of the large cargo UAV; an aerodynamic model of the cargo UAV is established, and the flap deflection angle is incorporated into the model when constructing the lift coefficient and drag coefficient aerodynamic equations, so that the regulating effect of the flap on the landing trajectory of the large cargo UAV is closer to the actual influence effect, which is more conducive to the accuracy of the landing trajectory design; the trajectory optimization process designs the longitudinal landing control equation, solves the bounded solution, uses feedback output, generates a state trajectory, and realizes the optimization of the landing trajectory. The optimization method also has certain enlightenment significance for the design and control of similar trajectories; in the trajectory detection and confirmation process, the actual landing trajectory is tracked and calculated, and compared with the designed trajectory, which further verifies the feasibility and effectiveness of the designed trajectory, and can greatly improve the landing safety of the large cargo UAV.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1: Schematic diagram of the landing trajectory of a large cargo drone;

The numbers in the figure are: deep glide path section-1, pull-up section-2, shallow glide path section-3, deep glide path angle-4, shallow glide path angle-5, pull-up section longitudinal height-6, approach and level flight section-7, touchdown and run section-8.

Detailed ways

The technical solution of this patent is further explained below in conjunction with the accompanying drawings and embodiments.

Example 1

As shown in the figure, the three stages of the landing trajectory of a large cargo drone are designed: deep descent stage 1, pull-up stage 2, and shallow descent stage 3;

The main trajectory parameters include the deep glide angle 4, the shallow glide angle 5, the longitudinal height of the pull-up section 6, the coordinates of the end point of the deep glide section, the geometric relationship between the corresponding trajectories of the deep glide section 1, the pull-up section 2 and the shallow glide section 3, the deep glide section, h _s =tan γ ₁ (x ₀ ), γ ₁ is the deep glide angle, x ₀ is the intersection of the deep glide section and the ground channel, x is the ground position corresponding to the location, the pull-up section, x _E is the starting point of the pull-up segment, h _D is the height difference of the pull-up segment, σ is the attenuation rate of the pull-up segment, and for the shallow glide segment, h _q =tan γ ₂ (x _D ), γ ₂ is the shallow glide angle, x _D is the coordinate of the touchdown point, and x is the ground position corresponding to the site;

The deep glide segment 1 is a quasi-straight line trajectory, which starts at the end point of the approach level flight segment 7 and ends at the start point of the pull-up segment 2. The pull-up segment 2 adjusts the longitudinal attitude of the large cargo drone by adjusting the flap configuration, and the drone changes from a head-down state to a head-up state, and enters the shallow glide segment 3, which starts at the end point of the pull-up segment 2 and ends at the touchdown point of the large cargo drone. The deep glide angle 4 is the angle between the extension line of the deep glide segment 1 trajectory and the horizontal channel, the shallow glide angle 5 is the angle between the shallow glide segment 3 trajectory and the horizontal channel, and the pull-up segment longitudinal height 6 is the projection distance of the pull-up segment trajectory of the large cargo drone on the vertical plane.

Example 2

Parameters are determined, the deep glide angle is determined, the inclination angle change rate of the deep glide section of the large cargo drone is 0, and the reference lift coefficient can be obtained mg is the gravity, q is the dynamic pressure, S is the reference area of the aircraft wing, so as to determine the deep glide angle γ1 and other parameters γ2 (shallow glide angle). The selection of the shallow glide angle should avoid exceeding the load limit due to too fast descent. The selection of the longitudinal height of the pull-up section should ensure that the pull-up section has a relatively suitable descent rate. The coordinates of the touchdown point are adjusted and changed with the actual length of the runway.

The aerodynamic model of a large cargo UAV is established, and the aerodynamic equations of lift coefficient and drag coefficient are:

q is the dynamic pressure, /> ρ is air density, x is air velocity, _CL , /> is the lift coefficient, C _D , /> is the drag coefficient, δ is the flap deflection angle, π is the constant, AR is the aspect ratio, and e is the trapezoidal ratio.

Combining the kinematics and dynamics equations of large cargo drones, Where m and v represent mass and speed, γ represents the track inclination, h represents the height, x represents the position along the runway, S represents the reference area of the aircraft wing, and q represents the dynamic pressure. Confirm the motion state of the large cargo drone and set the landing trajectory according to the motion state of the large cargo drone.

Example 3

Landing trajectory optimization: For the set landing trajectory, the longitudinal landing control equation is designed, bounded solutions are solved, and feedback output is used to generate state trajectories to achieve landing trajectory optimization;

The longitudinal landing control equation is:

y _d (t) = Cx _d (t), x _d (t) is the state variable, _ud (t) is the input variable, A and B are constants, and y _d (t) is the optimized output trajectory. When t → ∞, x(t) → x _d (t), y(t) → y _d (t), achieving accurate tracking of the output and finding bounded solutions inside the system;

For the aircraft landing trajectory, the altitude h and ground speed Vg

Output equation:

definition: _yi (t) represents the i-dimensional output, and the internal dynamic function is determined from the state variables as follows:

ε(t) = [q, θ] ^T , q is the pitch angle rate, θ is the pitch angle, and the transformation matrix T is solved.

for , find the linear transformation relationship of the original state variable x(t) and get the transformation matrix T.

Express x(t) and u(t) in the new coordinate system, and find the transformation of _yi (t) from the old coordinate system to the new coordinate system. x(t) is:

Find the bounded solution ε _d (t),

Combined with:

and

x(t) is expressed as:

Available:

The internal dynamic equation is obtained, and then the desired state trajectory can be obtained:

Achieve optimized landing trajectory design.

Example 4

Landing trajectory detection and confirmation: obtain sequence images through high-speed camera, obtain the coordinates of feature sites, and preliminarily locate the feature sites. Preprocess the image by setting the image pixel grayscale value, set the threshold, perform edge detection, and further locate it using HOUGH transformation. Calculate the change in the feature sites, calculate the motion parameters of the large cargo drone, and further solve the various parameters of the actual motion trajectory of the aircraft to confirm the actual landing trajectory.

The solution method is to use camera data, combined with a mathematical model of landing parameters of a large cargo drone, to obtain an accurate solution for the spatial coordinates of feature points on each frame of the image, and then convert the spatial coordinates of the feature points to the runway coordinate system of the large cargo drone to obtain high-precision coordinate information in the runway coordinate system, and output a trajectory curve based on the coordinate information of different feature points.

According to the body coordinates ( _XA , _YA , _ZA ) of the characteristic positions on the aircraft, the coordinates ( _XB , _YB , _ZB ) in the runway coordinate system are obtained through transformations such as rotation and translation.

According to the transformation matrix R and the translation parameters (X _j , Y _j , Z _j ), the motion trajectory of the aircraft can be calculated.

The six exterior orientation elements of the camera in the object space are obtained through the bundle adjustment algorithm. According to the rotation matrix composed of the angle elements, the spatial coordinates of the feature points in the body coordinate system are calculated to obtain the aircraft landing trajectory data, and then the actual motion trajectory of the aircraft is obtained. Through trajectory confirmation, the safety and reliability of the design of the landing trajectory of large cargo drones can be further ensured, and a verification and optimization method is provided for the design of more possible landing trajectories. The trajectory detection confirmation results show that the landing trajectory designed by the present invention is highly consistent with the actual landing trajectory, and the actual deep glide angle and shallow glide angle deviate from the designed landing angle by less than 0.2°. The height error at the time of touchdown is below 0.3m, achieving basically no deviation in the trajectory design. The trajectory control method disclosed in the present invention greatly improves the effectiveness of the landing trajectory design of large cargo drones, which is of great significance to the safety of landing of large cargo drones.

The above embodiments further elaborate and illustrate a large cargo drone landing trajectory control method disclosed in the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present application. For general technicians in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. The content of this specification should not be understood as a limitation on the present application.

Claims (4)

1. A landing trajectory control method for a large cargo drone, characterized in that the method comprises the following steps: S1) Pre-design the three stages of the landing trajectory of large cargo drones: deep glide stage, pull-up stage and shallow glide stage; S2) Determine the main trajectory parameters and geometric relationships. The main trajectory parameters include the deep glide angle, the shallow glide angle, the longitudinal height of the pull-up section, and the coordinates of the end point of the deep glide section. The geometric relationship reference height profile of the trajectory corresponding to the deep glide section, the pull-up section, and the shallow glide section is: Deep glide section, _hs = _tanγ1 ( _x0 ), _γ1 is the deep glide angle, _x0 is the intersection of the deep glide section and the ground channel, pull-up section, x _E is the starting point of the pull-up segment, h _D is the height difference of the pull-up segment, σ is the attenuation rate of the pull-up segment, and for the shallow glide segment, h _q =tanγ ₂ (x _D ), γ ₂ is the shallow glide angle, and x _D is the coordinate of the touchdown point; S3) parameters are determined, the deep glide angle is determined, the inclination change rate of the deep glide section is 0, and the reference lift coefficient can be obtained Determine the deep glide angle γ1 and the shallow glide angle to avoid exceeding the load limit due to too fast descent; S4) Establish an aerodynamic model for large cargo UAVs, and the aerodynamic equations for lift coefficient and drag coefficient are: _CL is the lift coefficient, _CD is the drag coefficient, and δ is the flap deflection angle; S5) combining the kinematics and dynamics equations of the large cargo UAV, confirming the motion state of the large cargo UAV, and setting the landing trajectory according to the motion state of the large cargo UAV; The kinetic equation is: m, v represent the mass and speed of the large cargo drone, γ represents the track inclination, h represents the height, x represents the horizontal position in the direction of the channel, and D and L represent the drag and lift, respectively. S6) Landing trajectory optimization: for the set landing trajectory, by designing the longitudinal landing control equation, solving the bounded solution, using the feedback output, generating the state trajectory, and realizing the optimization of the landing trajectory; The longitudinal landing control equation is: y _d (t) = Cx _d (t,), y _d (t) is the optimized output trajectory; S6) The process of solving the bounded solution, for the landing trajectory of the large cargo drone, the output equation of its height and ground speed is defined as Determine the internal dynamic function ε(t) = [q, θ] ^T and solve the transformation matrix. The variables in the equation are used to obtain the linear transformation relationship of the original state variables, and the transformation matrix T is obtained. Then, the bounded solution ε _d (t) is obtained through the transformation of the new and old coordinates, and the internal dynamic equation is obtained, and then the optimized state trajectory is obtained. The optimized state trajectory is: S7) Landing trajectory detection and confirmation: obtain sequence images through high-speed camera, obtain the coordinates of feature sites, and preliminarily locate the feature sites. Preprocess the images by setting the gray value of image pixels, set the threshold, perform edge detection, further locate by HOUGH transformation, calculate the change of feature sites, calculate the motion parameters of the large cargo UAV, further solve and obtain the various parameters of the actual motion trajectory of the large cargo UAV, and confirm the actual landing trajectory. 2. A landing trajectory control method for a large cargo UAV according to claim 1, characterized in that, S1) the deep glide segment is a quasi-straight line trajectory, which starts at the end point of the level flight stage and ends at the starting point of the pull-up segment. In the pull-up segment, the longitudinal posture of the large cargo UAV is adjusted by adjusting the flap configuration, and the large cargo UAV changes from a head-down state to a head-up state and enters the shallow glide segment, which starts at the end point of the pull-up segment and ends at the touchdown point of the large cargo UAV. 3. A large cargo UAV landing trajectory control method according to claim 1 is characterized in that, S2) the deep glide angle is the angle between the extension line of the deep glide segment trajectory and the horizontal channel, the shallow glide angle is the angle between the shallow glide segment trajectory and the horizontal channel, and the pull-up segment longitudinal height is the projection distance of the pull-up segment trajectory of the large cargo UAV on the vertical plane. 4. A large cargo UAV landing trajectory control method according to claim 1, characterized in that the solution in S7) is to use camera data, combined with a large cargo UAV landing parameter mathematical model, to obtain an accurate solution of the spatial coordinates of the feature points on each frame of the picture, and then convert the spatial coordinates of the feature points to the large cargo UAV runway coordinate system to obtain high-precision coordinate information in the runway coordinate system, and output a trajectory curve based on the coordinate information of different feature points.