CN111984036A - Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera - Google Patents

Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera Download PDF

Info

Publication number
CN111984036A
CN111984036A CN202010892049.8A CN202010892049A CN111984036A CN 111984036 A CN111984036 A CN 111984036A CN 202010892049 A CN202010892049 A CN 202010892049A CN 111984036 A CN111984036 A CN 111984036A
Authority
CN
China
Prior art keywords
unmanned aerial
aerial vehicle
target
tracking
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010892049.8A
Other languages
Chinese (zh)
Other versions
CN111984036B (en
Inventor
刘志宏
王祥科
杨凌杰
丛一睿
李�杰
俞先国
陈浩
黄华平
王冠政
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN202010892049.8A priority Critical patent/CN111984036B/en
Publication of CN111984036A publication Critical patent/CN111984036A/en
Application granted granted Critical
Publication of CN111984036B publication Critical patent/CN111984036B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/12—Target-seeking control

Abstract

The invention discloses a method for tracking a fast moving target by a fixed wing unmanned aerial vehicle based on a pan-tilt camera, which comprises the following steps: step S1: the whole process is divided into three states of a tracking state, an adjusting state and a transition state, and the unmanned aerial vehicle flies according to a controller designed in an earlier stage when the unmanned aerial vehicle is in the tracking state; step S2: when in the adjusting state, the unmanned aerial vehicle switches the spiral direction, and meanwhile, the holder rotates from one side to the other side to continuously track the target; the transition state is to select to enter a tracking state or an adjustment state according to the current state of the unmanned aerial vehicle, so that the unmanned aerial vehicle is prevented from being far away from a target. The method has the advantages of simple principle, wide application range, capability of improving the tracking performance of the method on the fast moving target and the like.

Description

Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera
Technical Field
The invention mainly relates to the technical field of unmanned aerial vehicles, in particular to a method for tracking a fast moving target by a fixed wing unmanned aerial vehicle based on a pan-tilt camera.
Background
The unmanned aerial vehicle executes the target tracking task, which is one of important applications of the unmanned aerial vehicle and is also the basis for defining follow-up tasks. Compared with other types of unmanned aerial vehicles, the fixed-wing unmanned aerial vehicle has the characteristics of long flight time and high speed, and can be used for continuously tracking a ground target. In the whole tracking process, the unmanned aerial vehicle captures a target on the ground through the airborne vision sensor, and obtains the coordinates of the feature points on the image by means of a target detection algorithm. On the basis, the position of the target relative to the unmanned aerial vehicle is solved by the aid of internal and external parameters of the sensor in the traditional method, and then the control law is designed in a three-dimensional space. However, the resolving link error is large, and the flying speed of the fixed-wing unmanned aerial vehicle is high, so that the tracking effect is greatly influenced. Further, as the ground target is mostly in a moving state, how the fixed wing drone keeps the target in the field of view through the self movement and the deflection of the pan-tilt is also one of the key problems to be solved.
Because the fixed-wing unmanned aerial vehicle has the limit of the lowest flying speed and the flying direction has no arbitrariness, the difficulty in tracking the moving target is higher. Reference [1]]A target tracking algorithm is provided, and the fixed-wing unmanned aerial vehicle tracks a moving target by estimating the GPS coordinate, the speed and the orientation of the target and combining a low-cost micro holder. Reference [2]]By estimating the target speed, L is applied1The fast adaptive estimator designs a tracking controller, and a single camera with a holder and a high-bandwidth wireless link are provided for the fixed-wing unmanned aerial vehicle for transmitting videos and commands, so that the moving target is tracked by the fast adaptive estimator. Reference [3]A framework is proposed by considering different lighting, computational constraints and clarityAnd due to factors such as insufficient degree, the steps of target identification, target positioning, state estimation and unmanned aerial vehicle guidance are adopted, and the fixed-wing unmanned aerial vehicle can track the ground moving target. It should be noted that, in these control methods, target positioning is performed first, and then the design of the controller is completed by using state estimation and the like, so as to realize the tracking of the fixed-wing drone on the moving target. However, since the moving speed of the fixed-wing drone is fast, if the positioning error is large, the tracking error will be greatly affected. Based on this, reference [4]]An included angle between the orientation of the holder and the flight direction of the unmanned aerial vehicle is enabled to tend to 90 degrees, the controller is designed by directly utilizing the change rate of the target characteristic points through a visual servo method based on images, and the effectiveness of tracking control when the ground target moves slowly is verified. Although the method omits a target positioning link, the method only aims at tracking the target with a slow movement speed. Therefore, how to design a tracking strategy by combining the maneuvering characteristics of the unmanned aerial vehicle is very important for tracking the fast moving target.
The currently used method for tracking and controlling the moving target by the fixed-wing unmanned aerial vehicle mainly has the following two defects:
(1) when the unmanned aerial vehicle performs hovering tracking on a ground target, the position of the unmanned aerial vehicle relative to the ground target needs to be obtained through target positioning, and then state information of the target (such as the movement speed of the target) is obtained through state estimation. For example, the tracking control method for the unmanned aerial vehicle proposed in reference [1-3] is to complete the tracking of the fixed-wing unmanned aerial vehicle on the moving target according to the ideas of target positioning, state estimation and guidance control. However, the target positioning needs to use internal and external parameters of the vision sensor, and the parameters are easy to generate errors in the calibration process, and in addition, the fixed-wing unmanned aerial vehicle is in a high-speed motion state, so that the introduction of errors is easy to cause target loss. In addition, reference [1] needs to obtain the GPS coordinates of the target by means of GPS information, and this method cannot be used when the drone is in a GPS-denied environment. References [1-3] are respectively: [1] dobroodhov V N, Kaminer I, Jones K D, et al, vision-based tracking and motion estimation for moving targets using small UAVs [ C ]//2006American Control reference. IEEE,2006:6pp. [2] Li Z, Hovakiman N, Dobrookhov V, et al, vision-based tracking and motion estimation a using UAV [ C ]//49 IEEE reference on Decision and Decision (CDC) IEEE,2010: 2505. 2510 [3] Wang X, Zhu H, Zhuang D, et al, vision-based detection and motion estimation of a moving targets mounting J [ UAV ]// motion system [ UAV ]// J156, Journal J.156. 9. origin J..
When the controller is designed when the unmanned aerial vehicle orbits around the target, the visual servo control method based on the image is adopted, and the controller is directly designed on the image plane. The method does not need target positioning and state estimation, thereby reducing the influence of positioning errors on the tracking effect of the unmanned aerial vehicle and improving the performance of the unmanned aerial vehicle for tracking the moving target.
(2) When the fixed-wing unmanned aerial vehicle tracks the moving target, the moving target is tracked only by considering the increased degree of freedom of the holder camera, but the target with higher speed is tracked without considering the maneuverability of the unmanned aerial vehicle. After the fixed-wing unmanned aerial vehicle is provided with the cloud platform camera, if the target is in a motion state, the cloud platform can deflect according to the offset position of the target feature point relative to the image center, and therefore the target is kept in a view field. The above references [4-5] are all accomplished under the condition that the moving speed of the target is moderate or slow, that is, the fixed wing drone only needs to fly along a specific hovering direction (clockwise hovering or counterclockwise hovering), and then the target is kept in the field of view by utilizing the moving characteristic of the pan-tilt, so as to realize tracking. However, due to the limitation of mechanical characteristics of the pan-tilt, when the moving speed of the target is too fast, the target will be separated from the view field of the camera after the pan-tilt deflects to the extreme position, and may be continuously far away from the target, so that the unmanned aerial vehicle cannot continue to track. Reference [4] is: [4] orthogonal S A P, Hespan J P, video-Based Target Tracking with a small UAV, Optimization-Based Control protocols [ J ] Control Engineering Practice,2014,32:28-42 [5] Yang L, Liu Z, Wang X, et al.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides the method for tracking the fast moving target by the fixed-wing unmanned aerial vehicle based on the pan-tilt camera, which has the advantages of simple principle and wide application range and can improve the tracking performance of the fixed-wing unmanned aerial vehicle on the fast moving target.
In order to solve the technical problems, the invention adopts the following technical scheme:
a method for tracking a fast moving target by a fixed wing unmanned aerial vehicle based on a pan-tilt camera comprises the following steps:
step S1: the whole process is divided into three states of a tracking state, an adjusting state and a transition state, and the unmanned aerial vehicle flies according to a controller designed in an earlier stage when the unmanned aerial vehicle is in the tracking state;
step S2: when in the adjusting state, the unmanned aerial vehicle switches the spiral direction, and meanwhile, the holder rotates from one side to the other side to continuously track the target; the transition state is to select to enter a tracking state or an adjustment state according to the current state of the unmanned aerial vehicle, so that the unmanned aerial vehicle is prevented from being far away from a target.
As a further improvement of the process of the invention: the automaton model is divided into three states, called tracking states respectively
Figure BDA0002655744500000041
Adjustment state
Figure BDA0002655744500000042
Transition state
Figure BDA0002655744500000043
And determines a conversion relationship between them.
As a further improvement of the process of the invention: in a tracking state, the fixed-wing unmanned aerial vehicle performs spiral tracking around a target, an image-based visual servo control method is adopted, and the design of a controller is represented as follows:
Figure BDA0002655744500000044
wherein u isψSee the previous work on the invention. At upAnd utIn the design formula of (2), λuAnd λvA positive constant, the goal is to force the object on the current image toward the center of the image.
As a further improvement of the process of the invention: in the state of adjustment, when theta is equal to1When the condition is met and the unmanned aerial vehicle still flies along the current hovering direction, the posture of the holder quickly reaches saturation, so that the target moves to the image edge until the target disappears; let the drone change hover direction, on the other hand let the pan-tilt rotate to the other side of the drone, defining the process as
Figure BDA0002655744500000045
Status.
As a further improvement of the process of the invention: make the unmanned plane with the maximum yaw rate
Figure BDA0002655744500000046
The adjustment is carried out, and meanwhile, the tripod head also has the maximum deflection angle speed
Figure BDA0002655744500000047
And
Figure BDA0002655744500000048
adjusting; the desired angle for the pan-tilt to deflect to the other side is
Figure BDA0002655744500000049
And
Figure BDA00026557445000000410
these two angles need to be determined to ensure that the target can be detected again; considering that the larger the angle at which the drone is deflected to the other side, the larger the area swept by the camera and thus the faster the target can be detected. Thus, if the current yaw angle of the pan/tilt head is
Figure BDA00026557445000000411
(<0) Then, then
Figure BDA00026557445000000412
Is gammapThe elapsed time is:
Figure BDA00026557445000000413
as a further improvement of the process of the invention:
Figure BDA00026557445000000414
the determination needs to be dependent on
Figure BDA00026557445000000415
And the position of the drone relative to the target; o is1Denotes Θ1Position of the drone just when satisfied, O2Representing the position of the drone after a time of Δ 2; the corresponding optical axes of the cameras at the two positions are O1B1And O2B2And is less than G1O2B2Representing half the camera field angle.
As a further improvement of the process of the invention: in the transition state, marking the mark whether the target is detected as NtWhen the object is detected N t1, otherwise Nt0; order to
Figure BDA0002655744500000051
As a further improvement of the process of the invention: when theta is higher than theta2When the conditions are met, the unmanned aerial vehicle leaves
Figure BDA0002655744500000052
A state; o is1，O2And O3Indicating the position of the drone, O2H2And O3H3Normal, O, representing the unmanned aerial vehicle trajectory at that point2C1And O3C2Representing the camera optical axis; when unmanned aerial vehicleAt O1Then the target is detected for the first time, when it is located at the edge of the image; if the next state is
Figure BDA0002655744500000053
The cradle head needs to make the target tend to the center of the image by adjusting the posture, so that the posture of the cradle head continues to increase; when the target is at O2When it reaches near the center of the image, Θ1May be satisfied again, causing the drone to re-enter
Figure BDA0002655744500000054
Status. O is3The same applies to the case.
Compared with the prior art, the invention has the advantages that:
1. the invention discloses a method for tracking a fast moving target by a fixed wing unmanned aerial vehicle based on a pan-tilt camera, which is a tracking strategy based on three state transitions, and comprehensively considers the motion characteristics of the unmanned aerial vehicle and the pan-tilt so as to improve the tracking performance of the fixed wing unmanned aerial vehicle on the fast moving target. Firstly, the whole process is divided into three states of a tracking state, an adjusting state and a transition state, and the unmanned aerial vehicle flies according to a controller designed in an earlier stage when the unmanned aerial vehicle is in the tracking state; when in the adjusting state, the unmanned aerial vehicle switches the spiral direction (if the current spiral is clockwise, the spiral direction is changed into anticlockwise spiral), and meanwhile, the tripod head rotates from one side to the other side to continuously track the target; the transition state is to select to enter a tracking state or an adjustment state according to the current state of the unmanned aerial vehicle, so that the unmanned aerial vehicle is prevented from being far away from a target. In short, the motion performance of the unmanned aerial vehicle and the motion performance of the holder are integrated, and the fixed-wing unmanned aerial vehicle can track the fast moving target.
2. According to the method for tracking the fast moving target by the fixed wing unmanned aerial vehicle based on the pan-tilt camera, when the controller of the unmanned aerial vehicle in the process of circling around the target is designed, the three-dimensional space position of the target does not need to be calculated. The control process adopts a visual servo method based on images, takes the geometric center of a target on the images as a characteristic point, and directly designs a control law on an image plane. The design method realizes the direct mapping from the change rate of the characteristic points to the control of the whole system, and the target position is not required to be calculated through the calibration of the sensor parameters, so that the error in the parameter calibration process is avoided, and the tracking performance is improved.
3. According to the method for tracking the fast moving target by the fixed wing unmanned aerial vehicle based on the pan-tilt camera, the maneuverability of the unmanned aerial vehicle is increased on the basis of improving the tracking performance of the unmanned aerial vehicle by using pan-tilt deflection, so that the performance of the unmanned aerial vehicle is further improved, and the fast moving target is tracked by the unmanned aerial vehicle. Wherein, the system keeps the current motion state unchanged in the tracking state; in the adjusting state, the unmanned aerial vehicle and the holder are maneuvered simultaneously, so that the current hovering direction of the unmanned aerial vehicle is changed, and the situation that the unmanned aerial vehicle cannot find a target is avoided; and during the transition state, the unmanned aerial vehicle selects to enter a tracking state or an adjustment state according to the current state of the system, so that the unmanned aerial vehicle continuously tracks the fast moving target.
Drawings
FIG. 1 is a schematic diagram of a fixed wing drone of the present invention changing hover direction by a deterministic finite automaton.
FIG. 2 shows an example of the present invention in which an unmanned aerial vehicle tracks an objectpAnd thetatSchematic diagram of the variation of (1).
Fig. 3 is a schematic diagram of a position relationship between the drone and the target in an adjusted state in a specific application example of the present invention.
Fig. 4 is a schematic diagram of a fixed wing drone in a specific application example of the invention.
Fig. 5 is a schematic diagram of three states of the present invention in a specific application example.
Detailed Description
The invention will be described in further detail below with reference to the drawings and specific examples.
As shown in fig. 1, the method for tracking a fast moving target by a fixed-wing drone based on a pan-tilt camera according to the present invention includes the steps of:
step S1: the whole process is divided into three states of a tracking state, an adjusting state and a transition state, and the unmanned aerial vehicle flies according to a controller designed in an earlier stage when the unmanned aerial vehicle is in the tracking state;
step S2: when in the adjusting state, the unmanned aerial vehicle switches the spiral direction (if the current spiral is clockwise, the spiral direction is changed into anticlockwise spiral), and meanwhile, the tripod head rotates from one side to the other side to continuously track the target; the transition state is to select to enter a tracking state or an adjustment state according to the current state of the unmanned aerial vehicle, so that the unmanned aerial vehicle is prevented from being far away from a target. In short, the motion performance of the unmanned aerial vehicle and the motion performance of the holder are integrated, and the fixed-wing unmanned aerial vehicle can track the fast moving target.
As the target speed increases, the target may be lost if the fixed wing drone is hovering around only one direction. As shown in FIG. 1, O1' M denotes a motion trajectory of the object, O1O2Representing the trajectory of the unmanned aerial vehicle, O1′O2' is O1O2Projection on the ground. Due to the limitation of mechanical characteristics of the holder, when the target speed is too fast in the left image, in order to enable the target speed to tend to the center of the image, the posture of the holder of the unmanned aerial vehicle can quickly reach saturation in the turning process, and finally the target can be far away from the unmanned aerial vehicle.
To avoid this, the present invention designs a deterministic finite automata model. The model can change the hovering direction of the unmanned aerial vehicle when the unmanned aerial vehicle turns, and the effect is shown in the right diagram of fig. 1. Finally, the unmanned aerial vehicle is prevented from being far away from the target, so that the tracking capability of the unmanned aerial vehicle on the moving target is improved.
The invention divides the automaton model into three states, which are respectively called tracking states
Figure BDA0002655744500000071
Adjustment state
Figure BDA0002655744500000072
Transition state
Figure BDA0002655744500000073
And determines a conversion relationship between them. Some symbols are defined as follows:
TABLE 1 notation and corresponding explanations
Figure BDA0002655744500000074
(1) Tracking states;
in the state, the fixed-wing unmanned aerial vehicle carries out spiral tracking around a target, a visual servo control method based on images is adopted, and the design of a controller is represented as follows:
Figure BDA0002655744500000081
wherein u isψSee the work at the outset of the present invention (reference [5]]Theorem iii in (1). At upAnd utIn the design formula of (2), λuAnd λvA positive constant, the goal is to force the object on the current image toward the center of the image.
At the controller Ψ1Under the effect of, unmanned aerial vehicle can let it tend to the image center when tracking the target. However, when the drone is directly in front of a passing target, | θpAnd thetatThe value of | will increase. As shown in FIG. 2, N2Indicating that the drone is just in front of the target, OB is tangent to the trajectory of the drone, a1Indicates the position of the assumed object, A2Indicating the current position, OA, of the moving object1And OA2Both represent the optical axis. The invention defines theta corresponding to the two positionspIs composed of
Figure BDA0002655744500000082
And
Figure BDA0002655744500000083
corresponding to thetatIs composed of
Figure BDA0002655744500000084
And
Figure BDA0002655744500000085
the following relationship holds:
Figure BDA0002655744500000086
if the target speed is too fast,
Figure BDA0002655744500000087
and
Figure BDA0002655744500000088
the rapid increase causes the pan tilt attitude angle to reach the limit, resulting in the left diagram in fig. 1. Therefore, the present invention defines θ separatelypAnd thetatHas a threshold value of
Figure BDA0002655744500000089
And
Figure BDA00026557445000000810
order to
Figure BDA00026557445000000811
When theta is higher than theta1When not satisfied, mean that unmanned aerial vehicle is when the turn in the left picture of fig. 1, the cloud platform can not deflect to the limit. Therefore, in the process, the target can be kept at the visual field center through the deflection of the holder, and the unmanned aerial vehicle can be always positioned at the visual field center
Figure BDA00026557445000000812
Status.
(2) Adjusting the state;
when theta is higher than theta1When satisfied and unmanned aerial vehicle still flies along current hover direction, the cloud platform gesture will reach saturation very fast to lead to the target to move to the image edge until disappearing. Therefore, the invention enables the unmanned aerial vehicle to change the hovering direction on one hand and enables the holder to rotate to the other side of the unmanned aerial vehicle on the other hand. The process is defined as
Figure BDA0002655744500000091
Status.
Since the target in the process willThere is a brief disappearance, so it is desirable to minimize the time for this disappearance. According to this requirement, the invention provides for the drone to have a maximum yaw rate
Figure BDA0002655744500000092
The adjustment is carried out, and meanwhile, the tripod head also has the maximum deflection angle speed
Figure BDA0002655744500000093
And
Figure BDA0002655744500000094
and (6) adjusting. The desired angle for the pan-tilt to deflect to the other side is
Figure BDA0002655744500000095
And
Figure BDA0002655744500000096
these two angles need to be determined to ensure that the target can be detected again. Considering that the larger the angle at which the drone is deflected to the other side, the larger the area swept by the camera and thus the faster the target can be detected. Thus, if the current yaw angle of the pan/tilt head is
Figure BDA0002655744500000097
Then
Figure BDA0002655744500000098
Is gammapThe elapsed time is:
Figure BDA0002655744500000099
then, the process of the present invention is carried out,
Figure BDA00026557445000000910
the determination needs to be dependent on
Figure BDA00026557445000000911
And the position of the drone relative to the target. As shown in FIG. 3, O1Denotes Θ1Just full ofPosition of unmanned aerial vehicle in full time, O2Indicating the position of the drone after a time of at. The corresponding optical axes of the cameras at the two positions are O1B1And O2B2And is less than G1O2B2Representing half the camera field angle. Because the pan-tilt is arranged at the controller psi1Can keep the target near the center of the image, and the invention approximately considers O1B1Pointing to the target. Furthermore, the present invention recognizes drone slave O1To O2Yaw rate in the process is all
Figure BDA00026557445000000912
By combining the flying height and flying speed of the unmanned aerial vehicle and the attitude of the unmanned aerial vehicle and the holder, O can be estimated through simple geometric relationship1And O2The position of the unmanned aerial vehicle relative to the target is located at the position.
In fact, when the target performs uniform linear motion, the unmanned aerial vehicle is at O1And O2The position of the point relative to the target is not affected by the direction of movement of the target. Thus, in solving for
Figure BDA00026557445000000913
The specific direction of the movement of the target (such as westward movement) can be specified, and then the included angle between the directions of the movement of the target and the direction of the westward movement can be obtained by combining the azimuth angle of the unmanned aerial vehicle. When the unmanned aerial vehicle is positioned at O2Then, select G1The points are located on the trajectory of the target motion, so that the mapping of the target trajectory on the image intersects the image edges at a midpoint. In this situation, it can be ensured that even a slight offset of the target can be located in the camera field of view. Further, the determination can be made by simple geometric relationships
Figure BDA00026557445000000914
The size of (2).
In summary,
Figure BDA0002655744500000101
the controller of the time may be expressed as:
Figure BDA0002655744500000102
(3) transition state
Marking the mark whether the target is detected as NtWhen the object is detected N t1, otherwise Nt＝0。
Order to
Figure BDA0002655744500000103
When theta is higher than theta2When the conditions are met, the unmanned aerial vehicle leaves
Figure BDA0002655744500000104
Status. However, if the unmanned aerial vehicle directly enters at this moment
Figure BDA0002655744500000105
The status will be problematic. As shown in FIG. 4, O1，O2And O3Indicating the position of the drone, O in plan view2H2And O3H3Normal, O, representing the unmanned aerial vehicle trajectory at that point2C1And O3C2Representing the camera optical axis. When the unmanned aerial vehicle is positioned at O1The first time an object is detected, it is now located at the edge of the image. If the next state is
Figure BDA0002655744500000106
The cradle head needs to adjust the posture to enable the target to approach the center of the image, so that the posture of the cradle head continues to increase. When the target is at O2When it reaches near the center of the image, Θ1May be satisfied again, causing the drone to re-enter
Figure BDA0002655744500000107
Status. O is3The same applies to the case. Therefore, the drone may be far from the target in this situation.
Based on this, the present invention wants to enter the central area before the target (u)1∈(-ut,ut),v1∈(-vt,vt)，utAnd v t1/10 may be taken as the width and height of the image, respectively. ) The controller is still psi1And define the state as
Figure BDA0002655744500000108
Further, by setting θpIs determined whether to enter
Figure BDA0002655744500000109
Order to
Figure BDA00026557445000001010
It is noted that
Figure BDA00026557445000001011
Should be less than
Figure BDA00026557445000001012
Thereby preventing the phenomenon in fig. 4 from occurring again. At this time, when theta1When satisfied, due to
Figure BDA00026557445000001013
Unmanned aerial vehicle can not enter immediately
Figure BDA00026557445000001014
But continues to fly in the tracking state until the transition condition is satisfied. However, if in the process | θp| always greater than
Figure BDA00026557445000001015
Then theta3Will not be satisfied, thereby making the unmanned plane always in
Figure BDA0002655744500000111
Status. In this case, the drone will not change hover direction, so that the situation in fig. 1 occurs where the target disappears. To solve this problem, the present invention still desires to have the controller Ψ at this time2Can be used for function. Thus, according to equation (54), the present invention constructs another condition for directly converting the current state to
Figure BDA0002655744500000112
The conditions are expressed as follows:
Figure BDA0002655744500000113
thus, if the drone has not entered the tracking state when passing directly in front of the target's motion, Ψ2It will function directly.
In summary, the deterministic finite automata model can be represented as fig. 5. Defining the current state as SnThen the controller of the model can be expressed as:
Figure BDA0002655744500000114
meanwhile, the algorithm pseudo code of the model can be expressed as:
Figure BDA0002655744500000115
Figure BDA0002655744500000121
the above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (7)

1. A method for tracking a fast moving target by a fixed wing unmanned aerial vehicle based on a pan-tilt camera is characterized by comprising the following steps:
step S1: dividing the whole process into tracking states
Figure FDA0002655744490000011
Adjustment state
Figure FDA0002655744490000012
Transition state
Figure FDA0002655744490000013
Determining the conversion relation among the three states, and flying the unmanned aerial vehicle according to a controller designed in the earlier stage when the unmanned aerial vehicle is in the tracking state;
step S2: when the unmanned aerial vehicle is in the adjusting state, the unmanned aerial vehicle switches the spiral direction, and meanwhile, the holder rotates from one side to the other side to continuously track the target; the transition state is to select to enter the tracking state or the adjustment state according to the current state of the unmanned aerial vehicle, so that the unmanned aerial vehicle is prevented from being far away from the target.
2. The method for tracking the fast moving target by the fixed-wing unmanned aerial vehicle based on the pan-tilt-zoom camera according to claim 1, wherein in the tracking state, the fixed-wing unmanned aerial vehicle performs hovering tracking around the target, and the controller is represented by using an image-based visual servo control method:
Figure FDA0002655744490000014
wherein u isψYaw rate u at the pan/tilt head for the yaw rate of the dronepAnd pitch angle rate utIn the design formula of (2), λuAnd λvA positive constant, the goal is to force the object on the current image toward the center of the image.
3. The pan-tilt-camera-based method for tracking a fast moving object by a fixed-wing drone according to claim 2, characterized in thatIn the adjusted state, when the condition theta is1Satisfied and the drone is still flying in the current hover direction,
Figure FDA0002655744490000015
and
Figure FDA0002655744490000016
respectively representing the thresholds of the yaw angle and the pitch angle of the holder, and quickly saturating the posture of the holder, so that the target moves to the edge of the image until the target disappears; let unmanned aerial vehicle change direction of circling, on the other hand lets the cloud platform rotate to unmanned aerial vehicle's opposite side, defines this process as the adjustment attitude
Figure FDA0002655744490000017
Figure FDA0002655744490000018
4. The pan-tilt-camera-based method for tracking a fast moving object by a fixed-wing drone according to claim 3, wherein the drone is caused to have a maximum yaw rate
Figure FDA0002655744490000021
Adjustment is made while the pan/tilt head is also at a maximum yaw rate
Figure FDA0002655744490000022
And pitch rate
Figure FDA0002655744490000023
Adjusting; desired yaw and pitch angles for the pan-tilt to the other side are
Figure FDA0002655744490000024
And
Figure FDA0002655744490000025
these two angles need to be determined to ensure that the target can be detected again; considering that the larger the angle of the unmanned aerial vehicle deflecting to the other side is, the larger the area swept by the camera is, and thus the faster the target can be detected; if the current yaw angle of the holder is
Figure FDA0002655744490000026
Then
Figure FDA0002655744490000027
Maximum yaw angle gamma of hook holderpThe elapsed time is:
Figure FDA0002655744490000028
5. the pan-tilt-camera-based method for tracking a fast moving object by a fixed-wing drone according to claim 4, wherein the method is characterized in that
Figure FDA0002655744490000029
Is determined to depend on
Figure FDA00026557444900000210
And the position of the drone relative to the target; o is1Expression Condition Θ1Position of the drone just when satisfied, O2Representing the position of the drone after a time of Δ t; the corresponding optical axes of the cameras at the two positions are O1B1And O2B2And is less than G1O2B2Representing half the camera field angle.
6. The method for tracking the fast moving target by the fixed-wing unmanned aerial vehicle based on the pan-tilt camera according to any one of claims 1 to 5, wherein in the transition state, the mark whether the target is detected is marked as NtWhen the object is detected Nt1, otherwise Nt0; order:
Figure FDA00026557444900000211
7. the pan-tilt-camera-based method for tracking fast moving target by fixed-wing drone according to claim 6, wherein when Θ is2When the conditions are met, the unmanned aerial vehicle leaves
Figure FDA00026557444900000212
A state; o is1，O2And O3Indicating the position of the drone, O2H2And O3H3Normal, O, representing the unmanned aerial vehicle trajectory at that point2C1And O3C2Representing the camera optical axis; when the unmanned aerial vehicle is positioned at O1Then the target is detected for the first time, when it is located at the edge of the image; if the next state is
Figure FDA00026557444900000213
The cradle head needs to make the target tend to the center of the image by adjusting the posture, so that the posture of the cradle head continues to increase; when the target is at O2When the position reaches near the center of the image, the condition Θ1May be satisfied again, causing the drone to re-enter
Figure FDA00026557444900000214
A state; o is3The same applies to (1).
CN202010892049.8A 2020-08-28 2020-08-28 Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera Active CN111984036B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010892049.8A CN111984036B (en) 2020-08-28 2020-08-28 Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010892049.8A CN111984036B (en) 2020-08-28 2020-08-28 Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera

Publications (2)

Publication Number Publication Date
CN111984036A true CN111984036A (en) 2020-11-24
CN111984036B CN111984036B (en) 2022-09-16

Family

ID=73439740

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010892049.8A Active CN111984036B (en) 2020-08-28 2020-08-28 Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera

Country Status (1)

Country Link
CN (1) CN111984036B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090187299A1 (en) * 2008-01-23 2009-07-23 Fregene Kingsley Oc Method and System for Autonomous Tracking of a Mobile Target by an Unmanned Aerial Vehicle
CN103149939A (en) * 2013-02-26 2013-06-12 北京航空航天大学 Dynamic target tracking and positioning method of unmanned plane based on vision
US20130176423A1 (en) * 2012-01-05 2013-07-11 Parrot Method for piloting a rotary wing drone for taking an exposure through an onboard camera with minimization of the disturbing movements
US20140336848A1 (en) * 2013-05-10 2014-11-13 Palo Alto Research Center Incorporated System and method for detecting, tracking and estimating the speed of vehicles from a mobile platform
WO2018019290A1 (en) * 2016-07-28 2018-02-01 纳恩博（北京）科技有限公司 Method and device for controlling cradle head, and storage medium
CN108107920A (en) * 2017-12-19 2018-06-01 天津工业大学 A kind of microminiature twin shaft vision stablizes holder target detection tracing system
CN108399642A (en) * 2018-01-26 2018-08-14 上海深视信息科技有限公司 A kind of the general target follower method and system of fusion rotor wing unmanned aerial vehicle IMU data
CN108614580A (en) * 2018-06-22 2018-10-02 中国人民解放军国防科技大学 Layered obstacle avoidance control method in target tracking of unmanned aerial vehicle
CN110764537A (en) * 2019-12-25 2020-02-07 中航金城无人系统有限公司 Automatic tripod head locking system and method based on motion estimation and visual tracking

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090187299A1 (en) * 2008-01-23 2009-07-23 Fregene Kingsley Oc Method and System for Autonomous Tracking of a Mobile Target by an Unmanned Aerial Vehicle
US20130176423A1 (en) * 2012-01-05 2013-07-11 Parrot Method for piloting a rotary wing drone for taking an exposure through an onboard camera with minimization of the disturbing movements
CN103149939A (en) * 2013-02-26 2013-06-12 北京航空航天大学 Dynamic target tracking and positioning method of unmanned plane based on vision
US20140336848A1 (en) * 2013-05-10 2014-11-13 Palo Alto Research Center Incorporated System and method for detecting, tracking and estimating the speed of vehicles from a mobile platform
WO2018019290A1 (en) * 2016-07-28 2018-02-01 纳恩博（北京）科技有限公司 Method and device for controlling cradle head, and storage medium
CN108107920A (en) * 2017-12-19 2018-06-01 天津工业大学 A kind of microminiature twin shaft vision stablizes holder target detection tracing system
CN108399642A (en) * 2018-01-26 2018-08-14 上海深视信息科技有限公司 A kind of the general target follower method and system of fusion rotor wing unmanned aerial vehicle IMU data
CN108614580A (en) * 2018-06-22 2018-10-02 中国人民解放军国防科技大学 Layered obstacle avoidance control method in target tracking of unmanned aerial vehicle
CN110764537A (en) * 2019-12-25 2020-02-07 中航金城无人系统有限公司 Automatic tripod head locking system and method based on motion estimation and visual tracking

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
CHENG XU: "Target location of unmanned aerial vehicles based on the electro-optical stabilization and tracking platform", 《MEASUREMENT》 *
PENG YAO: "Real-time path planning of unmanned aerial vehicle for target tracking and obstacle avoidance in complex dynamic environment", 《AEROSPACE SCIENCE AND TECHNOLOGY》 *
杨凌杰: "An Optimized Image-Based Visual Servo Control for Fixed-Wing Unmanned Aerial Vehicle Target Tracking With Fixed Camera", 《IEEE ACCESS》 *
车飞: "无人机保距跟踪中的视觉跟踪算法研究", 《无人系统技术》 *

Also Published As

Publication number Publication date
CN111984036B (en) 2022-09-16

Similar Documents

Publication Publication Date Title
CN111596693B (en) Ground target tracking control method and system for unmanned aerial vehicle based on pan-tilt camera
Roelofsen et al. Reciprocal collision avoidance for quadrotors using on-board visual detection
CN108594848B (en) Unmanned aerial vehicle staged autonomous landing method based on visual information fusion
Monajjemi et al. UAV, come to me: End-to-end, multi-scale situated HRI with an uninstrumented human and a distant UAV
WO2018107419A1 (en) Control method, device and apparatus, and movable platform
CN110692027A (en) System and method for providing easy-to-use release and automatic positioning of drone applications
KR101740312B1 (en) Induction control method using camera control information of unmanned air vehicle
CN105468014A (en) Single autopilot integrated aircraft system and two-dimensional holder control method thereof
CN110825108B (en) Cooperative anti-collision method for multiple tracking unmanned aerial vehicles in same airspace
JP2009173263A (en) Method and system for autonomous tracking of mobile target by unmanned aerial vehicle (uav)
CN108733066B (en) Target tracking control method based on pod attitude feedback
CN109976386A (en) A kind of method and system of multiple no-manned plane collaboration tracking target
Li et al. UAV autonomous landing technology based on AprilTags vision positioning algorithm
Mills et al. Vision based control for fixed wing UAVs inspecting locally linear infrastructure using skid-to-turn maneuvers
Kassab et al. UAV target tracking by detection via deep neural networks
NO339419B1 (en) Path-Based Flight Maneuvering System
Yang et al. An optimized image-based visual servo control for fixed-wing unmanned aerial vehicle target tracking with fixed camera
Poiesi et al. Distributed vision-based flying cameras to film a moving target
CN111414007B (en) Multi-machine formation control description, transformation control and target tracking method
CN112783203B (en) Multi-sensor-based control system and method for unmanned aerial vehicle formation maintenance
CN111984036B (en) Method for tracking fast moving target by fixed-wing unmanned aerial vehicle based on pan-tilt camera
CN112947569A (en) Visual servo target tracking control method for quad-rotor unmanned aerial vehicle based on preset performance
CN110291013B (en) Control method of cradle head, cradle head and unmanned aerial vehicle
CN106919168A (en) It is indoor to test alignment system from model plane maneuverability
Liu et al. Visual-based online control of gimbal on the uav for target tracking

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant