Target assignment and route planning method for multi-agent unmanned surface vehicle

Abstract

A target assignment and route planning method for a multi-agent unmanned surface vehicle. The method comprises: sending task information to a task assigner (S1); performing screening to obtain targets to be assigned and unmanned surface vehicles to be subjected to assignment, and assigning one target to each of the unmanned surface vehicles on the basis of the comprehensive distance between the target and the unmanned surface vehicle, which comprehensive distance is obtained by taking a navigation distance and a steering penalty into consideration (S2); a route planner generating as a navigation route a smooth waypoint sequence for the matched unmanned surface vehicle and target (S3); and during navigation, a navigation controller constructing a speed optimization model on the basis of cooperative obstacle avoidance and sea area constraint requirements, and controlling, in real time, the unmanned surface vehicle to navigate along the waypoint sequence (S4).

Description

A target allocation and route planning method for multi-agent unmanned boats Technical Field

[0001] The present invention relates to the technical field of unmanned boat target allocation, and in particular to a target allocation and route planning method for a multi-agent unmanned boat. Background Art

[0002] In recent years, with the rapid development of artificial intelligence and robotics, unmanned vessels have been widely used in the field of marine operations. In particular, when performing complex tasks, the task collaboration of multi-agent unmanned vessels has become a research focus.

[0003] Target allocation algorithms are key technologies for unmanned platform collaboration. To effectively allocate tasks and plan routes, many researchers have proposed different algorithms and methods. For example, some studies have employed particle swarm optimization and genetic algorithms to address multi-task allocation and route planning, and these methods have proven successful in the field of unmanned aerial vehicles.

[0004] However, current research has not fully considered the unique maneuverability characteristics of unmanned vessels compared to other unmanned platforms, such as weak steering capabilities and susceptibility to hydrodynamic disturbances, as well as the constraints inherent in maritime environments, such as restricted navigation zones and mission boundaries. Furthermore, the need for collaborative obstacle avoidance between multi-agent unmanned vessels and targets requires further consideration. Therefore, it is essential to develop a task allocation and route planning system that can account for the unmanned vessel's own posture and orientation, as well as plan constrained maritime routes.

[0005] Summary of the Invention

[0006] In view of this, the present invention provides a target allocation and route planning method for a multi-agent unmanned vessel, aiming to improve the collaboration efficiency and mission success rate of the unmanned vessel.

[0007] The present invention achieves the above technical objectives through the following technical means.

[0008] A target allocation and route planning method for multi-agent unmanned boats:

[0009] (1) The task allocator receives task information, wherein the task information includes a task area, a prohibited navigation area, parameters and state quantities of the target, and parameters and state quantities of the unmanned boat;

[0010] The unassigned target set T is selected by using the parameters and state quantities of the target and unmanned boat. un and the unmanned boat set U un ;

[0011] (2) Assigning a corresponding target to each unmanned boat, including: calculating the adjusted distance d′ from each target j to the unmanned boat i ij :

[0012] d′ ij =d ij +Δd ij +Ωd ij

[0013] Among them, d ij is the Euclidean distance from the i-th unmanned boat to the j-th target, Δd ij To consider the adjustment item of the unmanned boat performance parameters to the distance, Ωd ij The penalty term for each target j to bypass the restricted area of ​​the unmanned boat i;

[0014] By the d′ ij Construct the adjustment distance matrix D between the unmanned boat and the target m×n , define the submatrix D′, D′∈D m×n , D′ contains the unassigned unmanned boat u i and target t j The corresponding row and column are based on the minimum element in D′ and the unassigned unmanned boat u i and target t j Matching; among them, unmanned boat u i Belongs to the unmanned boat set U that has no assigned tasks un , target t j Belongs to the unassigned target set T un ;

[0015] (3) For the matched unmanned boat and the corresponding target, the unmanned boat u is calculated. i To target t j Waypoint sequence P to bypass the restricted area ij , for P ij Perform interpolation to generate a new waypoint sequence P′ ij ; With sequence P′ ij As an unmanned boat i To target t j The navigation waypoint sequence is sent and stored in the navigation system of the corresponding unmanned boat;

[0016] (4) Unmanned boat u i Parameters and state quantities, obstacle set Obs i 、No-fly zone F m , mission area L and unmanned boat u i To its next waypoint P i ∈P′ ij The desired velocity vector v des Input to the optimizer to solve and obtain the speed control quantity;

[0017] Use speed control quantity to drive the unmanned boat u i Towards the target waypoint P i Drive, and when you reach the waypoint, press P i From the waypoint sequence P′ ij Delete until the waypoint sequence P′ is reached ij The last waypoint in the j Mission area, after completing the mission, the unmanned boat will be i Join U un , restart execution (2) until all tasks are completed.

[0018] A further technical solution is that the process of obtaining the penalty item for each target j to the unmanned boat i bypassing the restricted navigation zone is as follows:

[0019] 1) Divide the straight line path from the unmanned boat i to the target j into N segments, each segment is l long, and the endpoints of each segment are recorded as sampling points P0…P k …P N ;

[0020] 2) Determine the connection P k P k+1 and No-fly Zone F m Do the boundaries intersect? If so, record the first and last intersection points as a k 、b k , and delete a k 、b k Sampling points between

[0021] 3) In F m Find the boundary with a k The nearest point n k ;

[0022] 4) Calculate n k Distance F along the normal vector direction m Center point c m Point a after extending outward by a distance d n,k ;

[0023] 5) From a n,k Start by traversing F clockwise and counterclockwise respectively. m Point a is the point extending outwards from the boundary point by a distance d. n,k+i , until a n,k+i with b k The connection line with F m Do not intersect, record the path length L for clockwise and counterclockwise respectively cw , L ccw ;

[0024] 6) Select Lcw and L ccw The smaller value is denoted as L k ;

[0025] 7) Will bypass F m The path point, namely L k Add the points in the sample point a k 、b k Repeat 2) to 6) to process the subsequent prohibited flight zones F m+i ;

[0026] 8) The final sampling point sequence P is obtained to bypass all F m The path has a length of Ωd′ ij ;

[0027] 9) For the sampling point sequence P, calculate each adjacent sampling line segment P k P k+1 to P k+1 P k+2 The steering angle θ k ;

[0028] 10) Calculate the path turning penalty: ω k is the steering penalty parameter;

[0029] 11) The penalty term Ωd for each target j to the unmanned boat i to bypass the restricted area ij =Ωd ij =Ωd ij ′+Ωω.

[0030] In a further technical solution, the adjustment term considering the performance parameters of the unmanned boat for the distance is calculated according to the following formula:

[0031] Among them, v i is the velocity vector of the i-th unmanned boat, θ i is the direction of the velocity vector, V i is the maximum speed, ω i is the maximum turning speed, θ ij is the azimuth angle of the i-th unmanned boat pointing to the j-th target, and ω1 and ω2 are weight coefficients.

[0032] Further technical solution, the P ij Perform interpolation to generate a new waypoint sequence P′ ij , specifically using the following methods:

[0033] 1)P ij Construct a cubic B-spline curve C(u) as a sequence of control points;

[0034] 2) Define the steering energy function Indicates steering energy;

[0035] 3) Define the route length function Indicates the length of the route;

[0036] 4) Construct the objective function E = w1E b +w2E l , where w1 and w2 are weight coefficients;

[0037] 5) By adjusting the values ​​of w1 and w2 to minimize the objective function, a smooth curve C is obtained. * (u);

[0038] 6) C * (u) curve is sampled to generate a new waypoint sequence P tij ;

[0039] 7) Delete the new waypoint sequence P tij The points in the restricted area or in the mission area are obtained as the sequence P′ ij .

[0040] In a further technical solution, the objective function of the optimizer is:

[0041] J(v)=|v des -v| 2 +c penalty (v)

[0042] Among them, v is the optimization variable, c penalty For penalty items.

[0043] In a further technical solution, the constraints constructed by the optimizer are:

[0044] c1:|v|≤V i , ensure that the speed is less than the maximum speed V i ;

[0045] c2: Ensure that the future location is not in the prohibited flight zone F m Inside;

[0046] c3:p next ∈L, ensuring that the future position is within the task area L;

[0047] Among them, p next =P i +v·dt is the position at the next moment estimated based on the speed control value v.

[0048] A further technical solution is to solve the speed control variable v that minimizes the objective function J(v) * :

[0049] The solution v * As the speed control quantity.

[0050] A further technical solution is to solve the problem of each unmanned boat u in each time step dt solved by the optimizer. i , check if there are other unmanned boats u within the search radius j or target t k If so, the other unmanned boats u j or target t k Add obstacle set Obs i .

[0051] According to a further technical solution, the parameters and status of the unmanned boat include a unique identifier, maximum speed, turning speed, search radius, whether a task has been assigned, current position and current velocity vector; the parameters and status of the target include a unique identifier, whether it has been assigned, operating area radius and current position.

[0052] According to a further technical solution, the mission area is a convex polygon defined by a number of two-dimensional coordinates that meet the WSG84 standard; and the no-fly zone is a convex polygon defined by a number of two-dimensional coordinates that meet the WSG84 standard.

[0053] The beneficial effects of the present invention are:

[0054] (1) Efficient task allocation strategy: The present invention adopts a task allocation algorithm based on the adjusted distance matrix, which not only considers the straight-line distance, but also adjusts the distance in combination with the performance parameters of the unmanned boat and the penalty item for bypassing the restricted area.

[0055] (2) Detailed route planning: By considering details such as the straight path between the UAV and the target, detours in restricted areas, and smoothing of the route, the present invention ensures that the generated waypoints meet the actual navigation needs while avoiding unnecessary detours and turns.

[0056] (3) Dynamic collaborative obstacle avoidance: During navigation, the navigation controller of the present invention can detect and adjust the navigation posture of the unmanned boat in real time to meet the requirements of collaborative obstacle avoidance and sea area constraints, ensuring that there will be no conflict between multiple unmanned boats, while also ensuring a safe distance from other targets.

[0057] (4) Flexibility and scalability: The system design of the present invention takes into account the collaborative work of unmanned boats of different numbers and performances. It has good flexibility and scalability and can adjust parameters and strategies according to actual mission requirements.

[0058] (5) Adaptability: Parameters such as the speed and turning speed of the unmanned boat, as well as factors such as the number and position of targets, may change in actual applications. The system of the present invention can adjust task allocation and route planning in real time according to these parameters, ensuring the real-time and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order that the present application may be better understood, its various forms given by way of example will now be described with reference to the accompanying drawings, in which:

[0060] FIG1 is a flow chart of target allocation and route planning for a multi-agent unmanned vehicle according to the present invention;

[0061] FIG2 is a schematic diagram of obtaining the penalty item for the target j to the unmanned boat i bypassing the restricted navigation zone according to the present invention;

[0062] FIG3 is a diagram of an actual machine operation interface of the method of the present invention.

[0063] The drawings are for illustration purposes only and are not intended to limit the scope of protection of the present application in any manner or form. DETAILED DESCRIPTION

[0064] The present invention will be described in detail below with reference to the various embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0065] As shown in FIG1 , the present invention provides a target allocation and route planning method for a multi-agent unmanned vehicle, which comprises the following steps:

[0066] Step (1): The upper-level command controller sends the mission information to the mission dispatcher, where the mission information includes the mission area, the restricted area, the relevant parameters of the target and the relevant parameters of the unmanned boat; the mission dispatcher processes the received mission information as follows:

[0067] S11, set an irregular sea area L (i.e., mission area), which is composed of n a A convex polygon defined by two-dimensional coordinates that meet the WSG84 standard;

[0068] S12, set n f irregular no-fly zones Each restricted area is A convex polygon defined by two-dimensional coordinates that meet the WSG84 standard;

[0069] S13, set n u The parameters and status of an unmanned boat, including the unique identifier id i , Maximum speed V i, steering speed ω i , search radius r i , whether task b has been assigned assign,i 、Current location p i 、Current velocity vector v i , as a structure and stored in the task dispatcher, corresponding one-to-one with the connected unmanned boat real machine;

[0070] S14, set m t The parameters and status of the target to be completed, including the unique identifier id j , has it been assigned b target,j , radius of the operating area r j 、Current locationq j , as a structure and stored in the task dispatcher, corresponding one to one with the target; targets can be added or deleted dynamically;

[0071] S15, filter out the unassigned targets T through the state of the targets and the unmanned boats un and unmanned boats without assigned tasks un , as the initial input for each assigned task.

[0072] Step (2): The task allocator assigns corresponding targets to each unmanned boat, specifically including:

[0073] S21, calculate the adjusted distance d′ from each target j to the unmanned boat i ij , is given by the following formula:

[0074] d′ ij =d ij +Δd ij +Ωd ij

[0075] Among them, d ij is the Euclidean distance from the i-th unmanned boat to the j-th target; Δd ij In order to consider the adjustment item of the performance parameters of the unmanned boat to the distance, the following formula is given:

[0076] Among them, v i is the velocity vector of the i-th unmanned boat, θ i is the direction of the velocity vector, V i is the maximum speed, ω i is the maximum turning speed, θ ij is the azimuth angle of the i-th unmanned boat pointing to the j-th target, ω1 and ω2 are weight coefficients, and the specific values ​​are set according to different unmanned boat models. In this embodiment, they are set to 0.7 and 0.3 respectively; see Figure 2, the penalty term Ωd for each target j to unmanned boat i bypassing the restricted area ijThe specific steps for calculating are as follows:

[0077] 1) Divide the straight line path from the unmanned boat i to the target j into N segments, each segment is l long, and the endpoints of each segment are recorded as sampling points P0…P k …P N , the unmanned boat is used as sampling point P0;

[0078] 2) For values ​​of k ranging from 0 to N-1, determine the connection P k P k+1 and No-fly Zone F m Do the boundaries intersect? If so, record the first and last intersection points as a k 、b k , and delete a k 、b k Sampling points between

[0079] 3) In F m Find the boundary point with a k The nearest point n k ;

[0080] 4) Calculate n k Distance F along the normal vector direction m Center point c m Point a after extending outward by a distance d (set according to the size of the no-fly zone) n,k ;

[0081] 5) From a n,k Start by traversing F clockwise and counterclockwise m Point a is the point extending outwards from the boundary point by a distance d. n,k+i , until a n,k+i with b k The connection line with F m Do not intersect, record the length L of the clockwise and counterclockwise paths respectively cw , L ccw ;

[0082] 6) Select L cw and L ccw The smaller value is denoted as L k ;

[0083] 7) Will bypass F m The path point, namely L k Add the points in the sample point a k 、b k Repeat 2) to 6) to process the subsequent prohibited flight zones F m+i ;

[0084] 8) The final sampling point sequence P is obtained to bypass all F m The path, that is, the sampling point sequence P consists of Np waypoints, whose length is Ωd′ ij ;

[0085] 9) For the sampling point sequence P, calculate each adjacent sampling line segment P k P k+1 to P k+1 P k+2 The steering angle θ k ;

[0086] 10) Calculate the path turning penalty: ω k is the steering penalty parameter;

[0087] 11) Finally, the penalty term Ωd for each target j to the unmanned boat i to bypass the restricted area ij is the sum of the sailing distance and the turning penalty: Ωd ij =Ωd ij ′+Ωω.

[0088] S22, the task allocator assigns each unmanned boat u i Allocation target t j , unmanned boat u i Unmanned boat U un , target t j Belongs to the target T that has not been assigned un ; Specifically include:

[0089] 1) Define the unmanned boat set U un ={u1,u2,...,u i ,...,u m}, target set T un ={t1,t2,...,t j ,...t n}, where m and n are the number of unmanned boats to be assigned and the number of targets to be assigned, respectively, which serve as the initial input for each assignment task;

[0090] 2) By d′ ij Construct the adjustment distance matrix D between the unmanned boat and the target m×n ;

[0091] 3) Initialize each unmanned boat u i The result of the assignment is null, that is, match(u i )=null;

[0092] 4) Define the submatrix D′, D′∈D m×n , D′ contains the rows and columns corresponding to the unassigned UAVs and targets;

[0093] 5) Find the minimum element d′ in the submatrix D′ ij , determined by the row and column of the minimum element to give the unmanned boat u i Assigned target t j ;

[0094] 6) If the target t j If it has not been assigned yet, u i Assigned to t j , set match(u i )=t j ;

[0095] 7) Delete the i-th row and j-th column from D′, indicating that the corresponding unmanned boat and target have been matched;

[0096] 8) Repeat steps 5) to 7) until all unmanned boats to be assigned and targets are matched.

[0097] S23, the matching result match(u i ) are marked as assigned, specifically:

[0098] 1) For the matching result match(u i ) Each unmanned boat u i , and b assign,i Mark as True;

[0099] 2) For the matching result match(u i ) for each target t j , and b target,j Mark as True;

[0100] Mark the unmanned vessel as assigned from U un Remove the target marked as allocated from T un Removed.

[0101] Step (3): The route planner (installed on the unmanned boat) sets the waypoints from itself to the target for each unmanned boat, specifically including:

[0102] S31, for the matched unmanned boat and the corresponding target, refer to the unmanned boat u calculated in S21 i To target t j Waypoint sequence P to bypass the restricted area ij ;

[0103] S32, using the cubic B-spline interpolation algorithm to calculate P ij Perform interpolation to generate a new waypoint sequence P′ ij , the specific steps are as follows:

[0104] 1)P ijConstruct a cubic B-spline curve C(u) as a sequence of control points;

[0105] 2) Define the steering energy function Indicates steering energy;

[0106] 3) Define the route length function Indicates the length of the route;

[0107] 4) Construct the objective function E = w1E b +w2E l , where w1 and w2 are weight coefficients;

[0108] 5) By adjusting the values ​​of w1 and w2 to minimize the objective function, a smooth curve C is obtained. * (u);

[0109] 6) C * (u) curve is sampled, the distance between sampling points is l′, and a new waypoint sequence P is generated. tij ;

[0110] 7) Delete the new waypoint sequence P tij The points in the restricted area or in the mission area are obtained as the sequence P′ ij .

[0111] S33, with sequence P′ ij As an unmanned boat i To target t j The navigation waypoint sequence is sent and stored in the navigation system of the corresponding unmanned boat.

[0112] Step (4): The navigation controller on the UAV continuously adjusts the navigation posture of the UAV during navigation to meet the requirements of coordinated obstacle avoidance and sea area constraints, specifically including:

[0113] S41, in each time step dt solved by the optimizer, for each unmanned boat u i , check the search radius r i Are there other unmanned boats in the j (i≠j) or target t k If so, the other unmanned boats u j (i≠j) or target t k Add obstacle set Obs i ;

[0114] Calculate the unmanned boat u i To its next waypoint P i ∈P′ ij The desired velocity vector v des , the direction is from u i to P iThe unit vector of the maximum speed V i ;

[0115] S42, unmanned boat u i Parameters and state quantities, obstacle set Obs i 、No-fly zone F m , mission area L and desired velocity vector v des Input to the optimizer for solution operation. The specific steps are as follows:

[0116] 1) The optimizer builds the objective function:

[0117] J(v)=|v des -v| 2 +c penalty (v)

[0118] Among them, v is the optimization variable, which represents the speed control vector of the unmanned boat; c penalty is a penalty term to prevent the unmanned boat from falling into a local optimum and then running still. When v is less than max(0.8*v des , 0.8*V i ), c penalty is 100 if yes, otherwise it is 0.

[0119] 2) Optimizer construction constraints:

[0120] c1:|v|≤V i , ensure that the speed is less than the maximum speed;

[0121] c2: Ensure that the future location is not within a restricted area;

[0122] c3:p next ∈L, ensuring that the future position is within the task area.

[0123] Among them, p next =P i +v·dt is the position at the next moment estimated based on the speed control value v.

[0124] 3) Use the SLSQP algorithm to solve the speed control variable v that minimizes the objective function J(v) * :

[0125] The solution v * Output as speed control quantity.

[0126] S43, use speed control value v * Driving unmanned boat i Towards the target waypoint P i Drive, and when you reach the waypoint, press P iFrom the waypoint sequence P′ ij Delete until the waypoint sequence P′ is reached ij The last waypoint in the j Mission area, after completing the mission, the unmanned boat will be i Join U un , restart step 2 until all tasks are completed.

[0127] Figure 3 is a diagram showing the actual operation interface of a method for target allocation and route planning for multi-agent unmanned boat collaboration described in this application. First of all, it should be clear that Figure 3 exists only as an example diagram and does not limit the scope of this application.

[0128] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for target allocation and route planning of a multi-agent unmanned boat, characterized in that: (1) The task allocator receives task information, wherein the task information includes a task area, a prohibited navigation area, parameters and state quantities of a target, and parameters and state quantities of an unmanned boat; The unassigned target set T is selected through the parameters and state quantities of the target and unmanned boat. un and the unmanned boat set U un ; (2) Assigning a corresponding target to each unmanned boat, including: calculating the adjusted distance d′ from each target j to the unmanned boat i ij : d′ ij =d ij +Δd ij +Ωd ij Among them, d ij is the Euclidean distance from the i-th unmanned boat to the j-th target, Δd ij To consider the adjustment item of the unmanned boat performance parameters to the distance, Ωd ij For each target j to unmanned boat i bypass the restricted area of ​​the penalty term; According to the d′ ij Construct the adjustment distance matrix D between the unmanned boat and the target m×n , define the submatrix D′, D′∈D m×n , D′ contains the unassigned unmanned boat u i and target t j The corresponding rows and columns are based on the smallest element in D′ and the unassigned unmanned boat u i and target t j Matching; among them, unmanned boat u i Belongs to the unmanned boat set U that has no assigned tasks un , target t j Belongs to the unassigned target set T un ; The process of obtaining the penalty item for each target j to the unmanned boat i bypassing the restricted navigation zone is as follows: 1) Divide the straight line path from unmanned boat i to target j into N segments, each segment is l long, and the endpoints of each segment are recorded as sampling points P0…P k …P N ; 2) Determine the connection P k P k+1 With the restricted area F m Do the boundaries intersect? If so, record the first and last intersection points as a k 、b k , and delete a k 、b k The sampling points between 3) In F m Find the boundary with a k The nearest point n k ; 4) Calculate n k Distance F along the normal vector direction m Center point c m Extending outwards a point a after a distance d n,k ; 5) From a n,k Start by traversing F clockwise and counterclockwise respectively. m The boundary point extends outwards to point a at a distance d n,k+i , until a n,k+i With b k The connection line with F m Do not intersect, record the path length L for clockwise and counterclockwise respectively cw , L ccw ; 6) Select L cw and L ccw The smaller value of k ; 7) Will bypass F m The path point, i.e. L k Add the points in to the sampling point a k 、b k Repeat 2) to 6) to process the subsequent prohibited flight zone F m+i ; 8) The final sampling point sequence P is obtained to bypass all F m The path has a length of Ωd′ ij ; 9) For the sampling point sequence P, calculate each adjacent sampling line segment P k P k+1 To P k+1 P k+2 The steering angle θ k ; 10) Calculate the path turning penalty: ω k is the steering penalty parameter; 11) The penalty term Ωd for each target j to unmanned boat i bypassing the restricted navigation zone ij =Ωd ij =Ωd ij ′+Ωω; (3) For the matched unmanned boat and the corresponding target, the unmanned boat u is calculated. i To target t j Waypoint sequence P to bypass the restricted area ij , for P ij Interpolate and generate a new waypoint sequence P′ ij ; With sequence P′ ij As an unmanned boat i To target t j The navigation waypoint sequence is sent and stored in the navigation system of the corresponding unmanned boat; (4) Unmanned boat u i Parameters and state quantities, obstacle set Obs i 、No-fly zone F m , mission area L and unmanned boat u i To its next waypoint P i ∈P′ ij The expected velocity vector v des Input to the optimizer to solve and obtain the speed control quantity; Use speed control quantity to drive the unmanned boat u i Towards the target waypoint P i Drive, and when you reach the waypoint, press P i From the waypoint sequence P′ ij until the waypoint sequence P′ is reached ij The last waypoint in the j Mission area, after completing the mission, the unmanned boat will be i Join U un , and restart execution (2) until all tasks are completed.

2. The target allocation and route planning method of a multi-agent unmanned boat according to claim 1 is characterized in that: The adjustment item considering the unmanned boat performance parameters for the distance is calculated according to the following formula: Among them, v i is the velocity vector of the i-th unmanned boat, θ i is the velocity vector direction, V i is the maximum speed, ω i is the maximum turning speed, θ ij is the azimuth of the i-th unmanned boat pointing to the j-th target, and ω1 and ω2 are weight coefficients.

3. The target allocation and route planning method of a multi-agent unmanned boat according to claim 1 is characterized in that: The pair P ij Interpolate and generate a new waypoint sequence P′ ij , the specific method is as follows: 1) P ij Construct a cubic B-spline curve C(u) as a sequence of control points; 2) Define the steering energy function Indicates the turning energy; 3) Define the route length function Indicates the length of the route; 4) Construct the objective function E = w1E b +w2E l , where w1 and w2 are weight coefficients; 5) By adjusting the values ​​of w1 and w2 to minimize the objective function, we can get the smooth curve C * (u); 6) C * (u) curve is sampled to generate a new waypoint sequence P tij ; 7) Delete the new waypoint sequence P tij The points in the restricted area or in the mission area are obtained as the sequence P′ ij .

4. The target allocation and route planning method of a multi-agent unmanned boat according to claim 1 is characterized in that: The objective function of the optimizer is: J(v) = |v des -v| 2 +c penalty (v) Among them, v is the optimization variable, c penalty For penalty items.

5. The target allocation and route planning method of a multi-agent unmanned boat according to claim 4 is characterized in that: The constraints constructed by the optimizer are: c1:|v|≤V i , ensure that the speed is less than the maximum speed V i ; c2: Ensure that the future location is not in the prohibited flight zone F m Inside; c3:p next ∈L, ensuring that the future position is within the task area L; Among them, p next =P i +v·dt is the position at the next moment estimated based on the speed control amount v.

6. The target allocation and route planning method of a multi-agent unmanned boat according to claim 5 is characterized in that: Find the speed control variable v that minimizes the objective function J(v) * : The solution v * As speed control quantity.

7. The target allocation and route planning method of a multi-agent unmanned boat according to claim 6 is characterized in that: In each time step dt solved by the optimizer, for each unmanned boat u i , check if there are other unmanned boats u within the search radius j or target t k If yes, then other unmanned boats u j or target t k Add obstacle set Obs i .

8. The target allocation and route planning method of a multi-agent unmanned boat according to claim 1 is characterized in that: The parameters and state quantities of the unmanned boat include a unique identifier, a maximum speed, a turning speed, a search radius, whether a task has been assigned, a current position and a current velocity vector; the parameters and state quantities of the target include a unique identifier, whether it has been assigned, a radius of an operating area and a current position.

9. The target allocation and route planning method of a multi-agent unmanned boat according to claim 1, characterized in that: The mission area is a convex polygon defined by a number of two-dimensional coordinates that meet the WSG84 standard; the no-fly zone is a convex polygon defined by a number of two-dimensional coordinates that meet the WSG84 standard.

Images