Multi-unmanned aerial vehicle cooperative observation enhanced yaw angle planning method based on auction algorithm

Abstract

The invention belongs to the technical field of unmanned aerial vehicles, and discloses a multi-unmanned aerial vehicle cooperative observation enhanced yaw angle planning method based on an auction algorithm, and the method comprises the steps: S1, the problem scene description of the auction algorithm and the determination of related symbol marks; s2, establishing an optimization problem model of an auction algorithm; s3, establishing an auction algorithm utility calculation model; s4, circulating an auction process of the auction algorithm; s5, checking target observation execution conditions; and S6, calculating a yaw angle and angular velocity output. According to the multi-unmanned aerial vehicle yaw angle distribution and planning method, friend mutual observation and enemy target observation can be enhanced, observation target distribution is faster and more effective, and multi-unmanned aerial vehicle cluster positioning is more robust and accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles, and in particular relates to a multi-unmanned aerial vehicle collaborative observation enhanced yaw angle planning method based on an auction algorithm. Background Art

[0002] At present, multi-UAV collaboration technology has a wide range of applications in many fields such as viewing, industry, and security. For example, using multi-UAV formations to create aerial landscapes, multi-UAV collaboration to complete the transportation of objects, and multi-UAV collaboration to capture illegal intruders, suspicious vehicles, fugitives and other moving targets. In the process of implementing collaborative tasks, cluster UAVs need to collaborate in environmental perception, target perception, cluster positioning, target positioning, etc. Among them, visual observation plays an important role in positioning constraints. The cluster positioning and target positioning formed by this have an important impact on the accuracy of subsequent collaborative operations. In general, the yaw angle planning method that helps enhance cluster mutual observation has the effect of improving the utilization rate of idle yaw angle observation resources and improving the robustness of cluster positioning and target positioning for multi-UAVs to perform cluster collaborative tasks. It can help cluster UAVs obtain accurate and robust cluster UAV positioning and target positioning, and improve the accuracy and robustness of subsequent task planning and control. However, the current technical solution has the problems of insufficient utilization of idle yaw angle observation resources and lack of observation constraints between UAVs. Summary of the invention

[0003] The purpose of the present invention is to provide a multi-UAV collaborative observation enhanced yaw angle planning method based on an auction algorithm to solve the above-mentioned technical problems.

[0004] In order to solve the above technical problems, the specific technical solution of the multi-UAV collaborative observation enhanced yaw angle planning method based on the auction algorithm of the present invention is as follows:

[0005] A multi-UAV collaborative observation enhanced yaw angle planning method based on an auction algorithm includes multiple UAVs, each of which is equipped with a camera, an IMU, other sensors required for the task, an onboard computer, and a flight control and power suit required for takeoff. The method uses the camera, the IMU, and other sensors required for the task to transmit data to the onboard computer, and performs the following steps:

[0006] S1: Description of the auction algorithm problem scenario and establishment of the symbolic notations involved;

[0007] S2: Establish the optimization problem model of the bidding algorithm;

[0008] S3: Establishment of the auction algorithm utility calculation model;

[0009] S4: auction process loop of the auction algorithm;

[0010] S5: target observation execution condition test;

[0011] S6: Calculate the yaw angle and angular velocity output.

[0012] Furthermore, the S1 includes problem scenario modeling applicable to multi-UAV collaborative observation enhanced yaw angle planning and establishment of symbolic markings used in the problem;

[0013] Problem scenario description: The bidding algorithm simulates an auction to solve the observation relationship allocation problem in multi-UAV collaborative operations. There are N observers and M objects to be observed in multi-UAV collaborative operations. The problem is to find an allocation scheme S to pair observers with objects to be observed, so that under this allocation scheme, the net profit of all observers who auction the M objects to be observed to N observers is maximized. The number of observers must be less than or equal to the objects to be observed, that is, N≤M.

[0014] The symbols used in the question are:

[0015] Let I={i 1 ,…,i N} is the set of observer numbers, J = {j 1 ,…,j M} is the set of object numbers to be observed;

[0016] is the allocation scheme, whose set size is T;

[0017] W = {w ij |i∈I,j∈J} is the utility set estimated by observer i from observing the observed object j;

[0018] G = {g ij |i∈I,j∈J} is the net benefit obtained by observer i from observing the observed object j;

[0019] P = {p ij |i∈I,j∈J} is the bid made by observer i in this round to bid for the observed object j;

[0020] B = {b j |j∈J} is the final bid price of the observed object j after a round of bidding;

[0021] F={f ij |i∈I,j∈J} is a set of flag variables for whether observer i is assigned to the observed object j. If the assignment relationship holds, then f ij =1, otherwise f ij =0.

[0022] Furthermore, the S2 comprises the following steps:

[0023] When the allocation scheme is S, the net benefit calculation formula for all observers is:

[0024]

[0025] In addition, there are constraints:

[0026] 1) Each observer is assigned at most one object to be observed, then:

[0027]

[0028] 2) Each observer must eventually be assigned to an object to be observed, so:

[0029] T=N (3)

[0030] In order to maximize the net benefits of all observers, we combine equations (1)-(3) to establish an optimization problem:

[0031]

[0032] T=N.

[0033] Furthermore, S3 designs the following utility items based on the principle of establishing observation relationships in multi-UAV collaboration:

[0034] 1) Distance cost term, the distance d between observer i and object j to be observed ij The closer, the better the observation effect, so we use As the distance utility, where δ is a very small positive number that prevents the denominator from being zero;

[0035] 2) Target priority item: The number of times the object j to be observed has been observed in the last five allocations f j The fewer the number, the higher the priority in this allocation, so we use As a goal priority utility;

[0036] 3) Yaw angle cost term: the angle θ between the original yaw angle of observer i and the yaw angle required to observe the object j to be observed ij The smaller the value, the lower the cost, so we use w yaw =cos(θ ij ) as the yaw angle utility;

[0037] 4) Observer Idleness Term: The shortest distance d between observer i and the edge of the site or obstacles min.i The closer it is, the more the observer i is in a non-idle state, and the more difficult it is to allocate observation resources to the object to be observed, so we use As the observer's idleness utility, d crash is the predefined safety distance, d maxis the maximum distance that a drone can reach from the edge of the site. After pre-screening by observers, d min.i >d crash , So leisure ∈(0,1];

[0038] Combining the above utility items, we can list the utility calculation formula of observer i towards observed object j:

[0039]

[0040] where α 1 ,α 2 ,α 3 is the weight parameter of each utility item.

[0041] Further, the S4 comprises the following steps:

[0042] First, initialize the parameters, set the initial bids P of all observers to the observed objects and the initial bids B of all objects to be observed to 0; set the slack complementarity parameter ∈ = 0.01, set a bidder set X, and initialize X = I;

[0043] Next, enter the auction process loop and traverse each bidder i in the bidder set X:

[0044] 1) Calculate the utility set π of the observer i for each observed object j i ={w ij -p ij} j∈J ;

[0045] 2) Calculate the maximum benefit π in the utility set max,i =max{π i}、Maximum benefit to be observed object j * and the next largest return π max2,i =max{π i |j≠j *}、The second largest return to be observed object j ** ;

[0046] 3) Update the maximum benefit of the observer i to the observed object j * The quote is The purpose is to * The corresponding bid for the scarcity increase of observer i;

[0047] 4) If the observed object j * If you have received quotes from other observers before, put these observers into the set X;

[0048] 5) Complete the bid of observer i and take observer i out of set X.

[0049] Execute the above steps 1)-5) for each observer i in the set X. When there are no more unassigned observers in the set X, the condition T=N is satisfied and the loop ends. The allocation plan S at this time is the final plan.

[0050] Further, the S5 comprises the following steps:

[0051] Allocation plan based on the auction algorithm Check whether observer i meets the validity conditions:

[0052] 1) Safety time condition: Observer i needs to have enough time to change the yaw angle from the observed object j to the yaw angle of the observed environment before flying close to the edge of the field or obstacles, otherwise the observation will not be performed. Judgment conditions:

[0053] a max (t out -t turn )≥v now (6)

[0054] where a max is the maximum acceleration of observer i, v now is the current speed of observer i, t out The time remaining before observer i flies close to the edge of the field or a safe distance around obstacles. The time required for observer i to observe object j and then return to the original position;

[0055] If the condition shown in formula (6) is true, the test passes; otherwise, the test fails and the observation is not performed;

[0056] 2) Observation effective distance condition: The distance between observer i and object j cannot exceed the maximum observation distance of 3m, and there must be no obstacles between them;

[0057] Observation pairs {i, j} that do not meet any of the above conditions 1) and 2) are excluded from the allocation scheme S, and the subsequent steps are not executed, and are not counted in the subsequent observation frequency p j .

[0058] Further, the S6 comprises the following steps:

[0059] Calculate the expected yaw angle for each observer i in the allocation S And perform steering at a constant angular velocity ω, assuming that the current yaw angle of observer i is The positions of observer i and object j in the world coordinate system are posi ,pos j , then the expected yaw angle of observer i is for:

[0060]

[0061] in is the X-axis direction vector of the world coordinate system, which is the same as pos j The angle is the absolute yaw angle that observer i wants to align

[0062] Execute at constant angular velocity ω arrive The direction in which the angular velocity should rotate is determined by calculating the cross product:

[0063]

[0064]

[0065] in is the X-axis direction vector of the observer i's body coordinate system, through which j The cross product k 3 The positive or negative value can determine which direction of the angular velocity can reach the destination faster, and the final result is the angular velocity ω of the observer i during the turning process;

[0066] After the desired yaw angle and angular velocity are output, the yaw angle plan is output to the drone controller to combine the position plan and the yaw angle plan, and the pose is solved for subsequent execution.

[0067] The multi-UAV collaborative observation enhanced yaw angle planning method based on the auction algorithm of the present invention has the following advantages:

[0068] (1) The present invention provides an observation relationship allocation solution that maximizes the collective observation utility through the design of an auction algorithm optimization problem.

[0069] (2) The present invention quantifies the attractiveness of the observed target to each observer through utility terms such as distance cost, target priority, and yaw angle cost, thereby improving the utility differentiation of different observation allocations in actual cluster observations.

[0070] (3) The present invention quantifies the abundance of the observer's own idle observation resources through the observer's idleness utility term, thereby reducing the observation allocation rights of non-idle observers.

[0071] (4) The present invention ensures the safety and effectiveness of the observation behavior through conditional verification of target observation execution.

[0072] The present invention can realize a multi-UAV yaw angle allocation and planning method for enhancing friendly mutual observation and enemy target observation, making the observation target allocation more rapid and effective and the multi-UAV cluster positioning more robust and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0074] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of a multi-UAV collaborative observation enhanced yaw angle planning method based on an auction algorithm of the present invention in conjunction with the accompanying drawings.

[0075] Embodiment 1: Prepare several drones, each equipped with a camera, an IMU, sensors required for other tasks, an onboard computer, a flight control system required for takeoff, a power kit, etc.; start the drone, use the above sensors to transmit data to the onboard computer, and implement a multi-drone collaborative observation enhanced yaw angle planning method based on an auction algorithm, such as Figure 1 As shown, perform the following steps:

[0076] S1: Description of the auction algorithm problem scenario and the establishment of the symbolic notations involved, including the problem scenario modeling applicable to multi-UAV collaborative observation enhanced yaw angle planning and the establishment of the symbolic notations used in the problem.

[0077] Problem scenario description: The bidding algorithm simulates an auction to solve the observation relationship allocation problem in multi-UAV collaborative operations. There are N observers and M objects to be observed in multi-UAV collaborative operations. The problem is to find an allocation scheme S to pair observers with objects to be observed, so that the net profit of all observers who auction M objects to be observed to N observers under this allocation scheme is maximized. Since the observer drones only include our drones, and the object drones to be observed include not only our drones but also enemy drones, the number of observers must be less than or equal to the objects to be observed, that is, N≤M.

[0078] The symbols used in the question are:

[0079] Let I={i 1 ,…,i N} is the set of observer numbers, J = {j 1 ,…,j M} is the set of object numbers to be observed;

[0080] is the allocation scheme, whose set size is T;

[0081] W = {w ij |i∈I,j∈J} is the utility set estimated by observer i from observing the observed object j;

[0082] G = {g ij |i∈I,j∈J} is the net benefit obtained by observer i from observing the observed object j;

[0083] P = {p ij |i∈I,j∈J} is the bid made by observer i in this round to bid for the observed object j;

[0084] B = {b j |j∈J} is the final bid price of the observed object j after a round of bidding;

[0085] F={f ij |i∈I,j∈J} is a set of flag variables for whether observer i is assigned to the observed object j. If the assignment relationship holds, then f ij =1, otherwise f ij =0.

[0086] S2: Establish an optimization problem model for the auction algorithm.

[0087] When the allocation scheme is S, the net benefit calculation formula for all observers is:

[0088]

[0089] In addition, there are constraints:

[0090] 1) Each observer is assigned at most one object to be observed, then:

[0091]

[0092] 2) Each observer must eventually be assigned to an object to be observed, so:

[0093] T=N (3)

[0094] In order to maximize the net benefits of all observers, we combine equations (1)-(3) to establish an optimization problem:

[0095]

[0096] S3: Establishment of the auction algorithm utility calculation model. Based on the principle of establishing observation relationships in multi-UAV collaboration, the following utility items are designed:

[0097] 5) Distance cost term, the distance d between observer i and object j to be observed ij The closer, the better the observation effect, so we use As the distance utility, where δ is a very small positive number that prevents the denominator from being zero.

[0098] 6) Target priority item: The number of times the object j to be observed has been observed in the last five allocations f j The fewer the number, the higher the priority in this allocation, so we use As a goal priority utility.

[0099] 7) Yaw angle cost term: the angle θ between the original yaw angle of observer i and the yaw angle required to observe the object j to be observed ij The smaller the value, the lower the cost, so we use w yaw =cos(θ ij ) as the yaw angle utility.

[0100] 8) Observer Idleness Term: The shortest distance d between observer i and the edge of the site or obstacles min.i The closer it is, the more the observer i is in a non-idle state, and the more difficult it is to allocate observation resources to the object to be observed, so we use As the observer's idleness utility. crash is the predefined safety distance, d max The maximum distance that a drone can reach from the edge of the field. After pre-screening of observers, it can be guaranteed that d min.i >d crash , So leisure ∈(0,1].

[0101] Combining the above utility items, we can list the utility calculation formula of observer i towards observed object j:

[0102]

[0103] where α 1 ,α 2 ,α 3 is the weight parameter of each utility item.

[0104] S4: The bidding process loop of the bidding algorithm.

[0105] First, initialize the parameters, set the initial bids P of all observers to the observed objects and the initial bids B of all objects to be observed to 0; set the slack complementary parameter ∈ = 0.01. Set a bidder set X, and initialize X = I.

[0106] Next, enter the auction process loop and traverse each bidder i in the bidder set X:

[0107] 6) Calculate the utility set π of the observer i for each observed object j i ={w ij -p ij} j∈J ;

[0108] 7) Calculate the maximum benefit π in the utility set max,i =max{π i}、Maximum benefit to be observed object j * and the next largest return π max2,i =max{π i |j≠j *}、The second largest return to be observed object j ** ;

[0109] 8) Update the maximum benefit of the observer i to the observed object j * The quote is The purpose is to * The corresponding bid for the scarcity increase of observer i;

[0110] 9) If the observed object j * If you have received quotes from other observers before, put these observers into the set X;

[0111] 10) Complete the bid of observer i and take observer i out of set X. Perform the above steps 1)-5) for each observer i in set X. When there are no more unassigned observers in set X (satisfying the condition T=N), the loop ends and the allocation plan S at this time is the final plan.

[0112] S5: Target observation execution condition test. Allocation scheme given by the auction algorithm Check whether observer i meets the validity conditions:

[0113] 1) Safety time condition: Observer i needs to have enough time to turn the yaw angle from the observed object j back to the yaw angle of the observed environment before flying close to the edge of the field or obstacles, otherwise the observation will not be performed. Judgment conditions:

[0114] a max (t out -t turn )≥v now (6)

[0115] where a max is the maximum acceleration of observer i, v now is the current speed of observer i, t out The time remaining before observer i flies close to the edge of the field or a safe distance around obstacles. It is the time required for observer i to observe object j and then return to the original position.

[0116] If the condition shown in formula (6) is true, the test passes; otherwise, the test fails and the observation is not performed.

[0117] 2) Observation effective distance condition: The distance between observer i and the object to be observed j cannot exceed the maximum observation distance of 3m, and there must be no obstacles between the two.

[0118] Observation pairs {i, j} that do not meet any of the above conditions 1) and 2) are excluded from the allocation scheme S, and the subsequent steps are not executed, and are not counted in the subsequent observation frequency p j .

[0119] S6: Calculate the yaw angle and angular velocity output. Calculate the expected yaw angle of each observer i in the allocation scheme S And perform steering at a constant angular velocity ω. Assume that the current yaw angle of observer i is The positions of observer i and object j in the world coordinate system are pos i ,pos j . Then the expected yaw angle of observer i is for:

[0120]

[0121] in is the X-axis direction vector of the world coordinate system, which is the same as pos j The angle is the absolute yaw angle that observer i wants to align

[0122] Execute at constant angular velocity ω arrive The direction in which the angular velocity should rotate can be determined by calculating the cross product:

[0123]

[0124]

[0125] in is the X-axis direction vector of the observer i's body coordinate system, through which j The cross product k 3 The positive or negative value can determine which direction of the angular velocity can reach the destination faster, and the final result is the angular velocity ω of the observer i during the turning process.

[0126] After the desired yaw angle and angular velocity are output, the yaw angle plan is output to the drone controller to combine the position plan and the yaw angle plan, and the pose is solved for subsequent execution.

[0127] The present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the purpose of the present invention. For example, it can be used for multi-robot yaw angle planning based on any purpose to assist cluster positioning based on direction-sensitive ranging.

[0128] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A method for enhancing yaw angle planning by collaborative observation of multiple UAVs based on an auction algorithm, comprising multiple UAVs, each of which is equipped with a camera, an IMU, sensors required for other tasks, an onboard computer, and a flight control and power set required for takeoff, characterized in that: The method uses the camera, IMU, and other sensors required for the mission to transmit data to the onboard computer and perform the following steps: S1: Description of the auction algorithm problem scenario and establishment of the symbolic notations involved; S2: Establish the optimization problem model of the bidding algorithm; S3: Establishment of the auction algorithm utility calculation model; S4: auction process loop of the auction algorithm; S5: target observation execution condition test; S6: Calculate the yaw angle and angular velocity output.

2. The method for enhancing yaw angle planning by multi-UAV cooperative observation based on bidding algorithm according to claim 1 is characterized in that: S1 includes problem scenario modeling applicable to multi-UAV collaborative observation enhanced yaw angle planning and the establishment of symbolic markings used in the problem; Problem scenario description: The bidding algorithm simulates an auction to solve the observation relationship allocation problem in multi-UAV collaborative operations. There are N observers and M objects to be observed in multi-UAV collaborative operations. The problem is to find an allocation scheme S to pair observers with objects to be observed, so that under this allocation scheme, the net profit of all observers who auction M objects to be observed to N observers is maximized. The number of observers must be less than or equal to the number of objects to be observed. That is, N≤M. The symbols used in the question are: Let I={i1,...,i N } is the set of observer numbers, J = {j1, ..., j M } is the set of object numbers to be observed; is the allocation scheme, whose set size is T; W={w ij |i∈I, j∈J} is the utility set estimated by observer i from observing the observed object j; G = {g ij |i∈I, j∈J} is the net benefit obtained by observer i from observing the observed object j; P = {p ij |i∈I, j∈J} is the bid made by observer i in this round to bid for the observed object j; B = {b j |j∈J} is the final bid price of the observed object j after a round of bidding; F={f ij |i∈I, j∈J} is a set of flag variables for whether observer i is assigned to the observed object j. If the assignment relationship holds, then f ij =1, otherwise f ij =0.

3. The method for enhancing yaw angle planning by multi-UAV cooperative observation based on bidding algorithm according to claim 1 is characterized in that: The S2 comprises the following steps: When the allocation scheme is S, the net benefit calculation formula for all observers is: In addition, there are constraints: 1) Each observer is assigned at most one object to be observed, then: 2) Each observer must eventually be assigned to an object to be observed, so: T=N(3) In order to maximize the net benefits of all observers, we combine equations (1)-(3) to establish an optimization problem:

4. The method for enhancing yaw angle planning by multi-UAV collaborative observation based on bidding algorithm according to claim 1 is characterized in that: The S3 is based on the principle of establishing observation relationships in multi-UAV collaboration and designs the following utility items: 1) Distance cost term, the distance d between observer i and object j to be observed ij The closer, the better the observation effect, so we use As the distance utility, where δ is a very small positive number that prevents the denominator from being zero; 2) Target priority item: The number of times the object j to be observed has been observed in the last five allocations f j The fewer the number, the higher the priority in this allocation, so we use As a goal priority utility; 3) Yaw angle cost term: the angle θ between the original yaw angle of observer i and the yaw angle required to observe the object j to be observed ij The smaller the value, the lower the cost, so we use w yaw =cos(θ ij ) as the yaw angle utility; 4) Observer Idleness Term: The shortest distance d between observer i and the edge of the site or obstacles min.i The closer it is, the more the observer i is in a non-idle state, and the more difficult it is to allocate observation resources to the object to be observed, so we use As the observer's idleness utility, d crash is the predefined safety distance, d max The maximum distance that a drone can reach from the edge of the site. After pre-screening by observers, it can be guaranteed So leisure ∈(0,1]; Combining the above utility items, we can list the utility calculation formula of observer i towards observed object j: Among them, α1, α2, and α3 are the weight parameters of each utility item.

5. The method for enhancing yaw angle planning by multi-UAV cooperative observation based on bidding algorithm according to claim 1 is characterized in that: The S4 comprises the following steps: First, initialize the parameters, set the initial bids P of all observers to the observed objects and the initial bids B of all objects to be observed to 0; set the slack complementarity parameter ∈ = 0.01, set a bidder set X, and initialize X = I; Next, enter the auction process loop and traverse each bidder i in the bidder set X: 1) Calculate the utility set π of the observer i for each observed object j i ={w ij -p ij } j∈J ; 2) Calculate the maximum benefit π in the utility set max,i =max{π i }、Maximum benefit to be observed object j * and the next largest return π max2，i =max{π i |j≠j * }、The second largest return to be observed object j ** ; 3) Update the maximum benefit of the observer i to the observed object j * The quote is The purpose is to increase the corresponding bid according to the scarcity of the observed object j* for the observer i; 4) If the observed object j* has received quotations from other observers before, these observers are placed in the set X; 5) Complete the bid of observer i and take observer i out of set X. Execute the above steps 1)-5) for each observer i in the set X. When there are no more unassigned observers in the set X, the condition T=N is satisfied and the loop ends. The allocation plan S at this time is the final plan.

6. The method for enhancing yaw angle planning by multi-UAV cooperative observation based on bidding algorithm according to claim 1 is characterized in that: The S5 comprises the following steps: Allocation plan based on the auction algorithm Check whether observer i meets the validity conditions: 1) Safety time condition: Observer i needs to have enough time to change the yaw angle from the observed object j to the yaw angle of the observed environment before flying close to the edge of the field or obstacles, otherwise the observation will not be performed. Judgment conditions: a max (t out -t turn )≥v now (6) where a max is the maximum acceleration of observer i, v now is the current speed of observer i, t out The time remaining before observer i flies close to the edge of the field or a safe distance around obstacles. The time required for observer i to observe object j and then return to the original position; If the condition shown in formula (6) is true, the test passes; Otherwise, the test fails and the observation is not performed; 2) Observation effective distance condition: The distance between observer i and object j cannot exceed the maximum observation distance of 3m, and there must be no obstacles between them; Observation pairs {i, j} that do not meet any of the above conditions 1) and 2) are excluded from the allocation scheme S, and the subsequent steps are not executed, and are not counted in the subsequent observation frequency p j .

7. The method for enhancing yaw angle planning by multi-UAV cooperative observation based on bidding algorithm according to claim 1 is characterized in that: The S6 comprises the following steps: Calculate the expected yaw angle for each observer i in the allocation S And perform steering at a constant angular velocity ω, assuming that the current yaw angle of observer i is The positions of observer i and object j in the world coordinate system are pos i ,pos j , then the expected yaw angle of observer i is for: in is the X-axis direction vector of the world coordinate system, which is the same as pos j The angle is the absolute yaw angle that observer i wants to align Execute at constant angular velocity ω arrive The direction in which the angular velocity should rotate is determined by calculating the cross product: in is the X-axis direction vector of the observer i's body coordinate system, through which j The positive or negative value of k3 obtained by the cross product can determine which direction of the angular velocity can reach the destination faster. The final result is the angular velocity ω of the observer i during the turning process. After the desired yaw angle and angular velocity are output, the yaw angle plan is output to the drone controller to combine the position plan and the yaw angle plan, and the pose is solved for subsequent execution.

Images