Multi-robot target tracking method in offshore multi-reef environment

Through the multi-robot leadership collaborative control system, the positive switching multi-integrity model and matrix decomposition technology are constructed, which solves the navigation problems that single robots cannot cope with in complex offshore multi-reef environments, and achieves efficient and robust target tracking and obstacle avoidance capabilities. Improves the adaptability and stability of the system.

CN118897545APending Publication Date: 2024-11-05HAINAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410933222.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

In a complex and changeable offshore multi-reef environment, a single robot is difficult to deal with complex terrain and obstacles. Traditional navigation algorithms are not enough to meet the requirements of high accuracy and robustness, especially in the reef waters and objects unknown. under conditions.

Method used

Using a multi-robot leadership collaborative control system, a positive multi-article model and leadership collaborative control technology of offshore multi-reef robot control system is designed by constructing a positive multi-article model and matrix decomposition technology, collaborative Liyapunov function and linear planning. , realize real-time monitoring and management, use multiple sensor information sharing and complementarity, and dynamically adjust tracking strategies and paths.

Benefits of technology

It improves the adaptability and flexibility of the robot system, enhances the perception and response capabilities of complex environments, achieves more efficient target tracking and obstacle avoidance capabilities, and improves the stability and robustness of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897545A_ABST
    Figure CN118897545A_ABST
Patent Text Reader

Abstract

The invention relates to the field of automation and modern control engineering and application, and discloses an offshore multi-reef environment multi-robot target tracking method, which comprises the following steps of: firstly, constructing a positive switching multi-agent system state space model by utilizing partial agent output based on a leader cooperation protocol; secondly, an exogenous interference signal model of the offshore multi-reef robot control system is constructed, unpredictable external interference can be simulated by designing the exogenous interference signal model, and the stability and reliability of the offshore multi-reef robot control system in a complex environment are improved; and finally, analyzing robot leader cooperative control by using a matrix decomposition technology, a Lyapunov function and a linear programming score, and dynamically adjusting a tracking strategy and path of each robot according to the motion trail and speed of the target, thereby realizing optimization of task allocation and improving tracking efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of automation and modern control engineering and applications, and specifically relates to a positive switching multi-agent modeling and target tracking method for an offshore multi-reef robot control system. Background Art

[0002] In the past few decades, communication technology and control technology have ushered in rapid development, making robot technology play a huge role in various work fields. Robot technology has also been applied in various fields such as map exploration, security patrol, remote monitoring, life search and special rescue, object transportation, etc. Among them, the multi-robot system has a performance far superior to that of an independent robot system due to its distribution characteristics of time, space, function, information and resources, and is an important development direction in the field of robotics. Compared with an independent robot system, a multi-robot system can complete relatively complex functional operation requirements through the cooperation of individuals with relatively simple functional structures, complete extremely efficient operations with extremely low-cost design, and has the characteristics of greater load capacity, larger operating space, and better system redundancy.

[0003] In offshore areas, reefs are unevenly distributed and have different shapes. They are often accompanied by seaweed, shellfish and other attachments, which hinder the movement of robots. Traditional single robots cannot adapt to changes in complex environments. At the same time, the reef area has the characteristics of shallow water depth and unknown underwater objects, which also puts higher requirements on the navigation ability of robots. Robots need to have accurate terrain perception and obstacle avoidance capabilities to avoid collisions with reefs. In this environment, traditional navigation algorithms may need to be optimized or improved to adapt to complex and changeable terrain features. In contrast, when multiple robots work together, they can each carry different types of sensors to improve their comprehensive perception of complex environments through information sharing and complementarity. For example, some robots may be good at optical image recognition, while others are good at acoustic detection. The combination of the two can more accurately perceive the terrain, obstacles and other underwater targets in the reef area. In the complex and changeable offshore multi-reef environment, various emergencies may be encountered (such as changes in ocean currents, sudden appearance of obstacles, etc.). Multi-robot collaborative control can jointly respond to these emergencies through collaborative decision-making and rapid response, thereby improving the robustness and response capabilities of the system.

[0004] In a complex and ever-changing environment, a single robot may find it difficult to cope with various challenges. Multi-robot leadership and collaboration can jointly cope with complex environments through decentralized deployment and collaborative operations, thereby improving the adaptability and flexibility of the system. For example, different robots can perform collaborative navigation and obstacle avoidance in reef areas according to their own characteristics and task requirements to achieve efficient tracking of targets. Multi-robot leadership and collaboration can assign different tasks to the most suitable robots to execute based on the strengths and capabilities of each robot. In the process of tracking the target, the tracking strategy and path of each robot can be dynamically adjusted according to the target's motion trajectory and speed to optimize task allocation and improve tracking efficiency. The use of multi-robot leadership and collaborative control has significant advantages in tracking targets. Summary of the invention

[0005] The purpose of the present invention is to study the offshore multi-reef robot system using the actual consensus protocol, and to establish a positive multi-agent model and target tracking technology for the offshore multi-reef robot control system based on the leadership and collaborative control problem in the offshore multi-reef robot control system using modern control theory and technology. First, based on the theory of leadership and collaboration, a state space model of a positive switching multi-agent system is constructed using partial agent outputs, and the corresponding gain matrix is ​​designed to achieve real-time monitoring and management of the robot system. Finally, matrix decomposition technology, collaborative Lyapunov functions and linear programming are used to design leadership and collaborative consistency for multiple robots. Designing a positive multi-agent model and leadership and collaborative control technology for an offshore multi-reef robot control system has important scientific research significance and practical application significance. Its specific technical scheme is as follows:

[0006] A method for tracking a target by multiple robots in an offshore multi-reef environment comprises the following steps:

[0007] Step 1: Establish a state space model of the positive switching multi-agent system of the offshore multi-reef robot control system;

[0008] Step 2: Establish a dynamic model of external interference signals of the offshore multi-reef robot control system;

[0009] Step 3, establish the dynamic model of the leader of the offshore multi-reef robot control system;

[0010] Step 4: construct the actual consensus protocol of the offshore multi-reef robot control system;

[0011] Step 5, design the offshore multi-reef robot control system based on the dual observer exchange protocol;

[0012] Step 6: Establish a closed-loop matrix of the offshore multi-reef robot control system;

[0013] Step 7: Design the positivity conditions for the offshore multi-reef robot control system under the positive switching multi-agent system;

[0014] Step 8: Design the stability conditions of the offshore multi-reef robot control system under the positive switching multi-agent system;

[0015] Step 9, verify the positivity of the offshore multi-reef robot control system under the positive multi-agent system;

[0016] Step 10: Verify the stability of the offshore multi-reef robot control system under the positive multi-agent system.

[0017] Furthermore, the state space model of the positive switching multi-agent system of the offshore multi-reef robot control system in step 1 is constructed in the following form:

[0018]

[0019] in, represents the system state of the i-th robot, including all the state information of the robot at time t;

[0020] represents the control input of the i-th robot, which contains all external commands or signals received by the agent at time t; represents the output of the i-th robot system, which contains all the signals or information sent by the agent at time t; Represents the interference signal from the external environment; σ i (t) is The switching law of the ith robot, and in the finite set S = {1,2,...,N}, matrix and are the known n×n, n×r, n×s, o×n, o×s dimensional matrices respectively; and σ i (t)∈S, assuming that for the matrix It's Metzler's. and are all greater than 0, i represents the i-th robot.

[0021] Furthermore, the dynamic model of the external interference signal of the offshore multi-reef robot control system in step 2 is constructed as follows:

[0022]

[0023] Among them, ξ i (t) is the additional signal for each robot, is a Metzler matrix of s×s dimensions, is a matrix of s×s dimensions greater than 0; in step 3, a dynamic model of the leader of the offshore multi-reef robot control system is established, and its construction form is as follows:

[0024]

[0025] Among them, η(t) is the state of the leader, is a Metzler matrix of dimension n×n.

[0026] Furthermore, the actual consensus protocol for constructing the offshore multi-reef robot control system in step 4 is constructed as follows:

[0027]

[0028] where h, represents a constant greater than 0, x * represents the position that the robot wants to reach, and η * Represents the position that the leader robot wants to achieve.

[0029] Furthermore, in step 5, the offshore multi-reef robot control system is designed based on the exchange protocol of two observers, and its construction form is as follows:

[0030]

[0031] in, and are the estimated values ​​of the state and exogenous disturbance, and is the controller gain matrix of r×n, r×s, r×o, r×n, r×n dimensions;

[0032] Design a state observer in the following form:

[0033]

[0034] in, is a Metzler matrix of dimension n×n, and Both are n×s-dimensional observer gain matrices;

[0035] The disturbance observer is constructed as:

[0036]

[0037] in, is the state of the observer, is a Metzler matrix of dimension s×n, and are the observer gain matrices of s×n and s×o.

[0038] Furthermore, the closed-loop matrix of the offshore multi-reef robot control system in step 6 is constructed as follows:

[0039] Define the error variables of the offshore multi-reef robot control system, and ε i (t) = x i (t)-η(t), therefore, the closed-loop system is:

[0040]

[0041] Furthermore, in step 7, the positivity conditions for ensuring the control system of offshore multi-reef robots under the positive switching multi-agent system are designed, and the construction conditions are as follows:

[0042] 7.1 The prerequisites for ensuring the operation of the observer are as follows:

[0043] Design constant m>0,α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality:

[0044]

[0045]

[0046] 7.2 Ensure the positive conditions of the offshore multi-reef robot control system as follows:

[0047]

[0048] for and i=1,2,…,n, hold; where, is the Kronecker product operator, is an r-dimensional vector whose ιth element is 1. Similarly, and is an n-dimensional and s-dimensional vector whose ιth element is 1; I n and I s represents n-dimensional and s-dimensional unit vectors;

[0049] 7.3 Under the control protocol and dual observers, the system is positive and reaches regional consensus. Define and The gain matrix is ​​as follows:

[0050]

[0051] And the average residence time satisfies the following inequality:

[0052]

[0053] Furthermore, in step 8, the stability conditions for ensuring the control system of the offshore multi-reef robot under the positive switching multi-agent system are designed, and the construction conditions are as follows:

[0054] 8.1 Design constant m>0, α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality:

[0055]

[0056] (μ 1 -α 1 )γ 1 -mδ 1 ≤0, (2)

[0057]

[0058] (μ 1 -α 2 )γ 2 -mδ 3 ≤0, (4)

[0059]

[0060] for and i=1,2,…,n, holds;

[0061] 8.2 Under the control protocol and dual observers, the system is positive and reaches regional consensus. Define and The gain matrix is ​​as follows:

[0062]

[0063] And the average residence time satisfies the following inequality:

[0064]

[0065] Furthermore, in step 9, the positivity of the offshore multi-reef robot control system is verified under the positive multi-agent system, as follows:

[0066] 9.1 Based on the 8th and 9th conditions designed in step 7.2, the gain matrix design can be rewritten as the following inequality

[0067]

[0068] According to the 1st to 6th conditions in step 7.1 and the above inequality,

[0069]

[0070] Therefore, it is concluded and

[0071] 9.2 Through the first condition of step 7.2, we get

[0072]

[0073] matrix is Metzler's; combined and

[0074] and Easy to get According to the second condition in step 7.2, then

[0075]

[0076] Then we can get

[0077]

[0078] Same reason According to the third condition in step 7.2, we can deduce is Metzler's; from the fourth condition of step 7.2, we know Then we can get

[0079]

[0080] According to the conditions and graph theory knowledge, and introduce

[0081]

[0082] So

[0083]

[0084] And then launched

[0085]

[0086] 9.3 Binding conditions and have

[0087]

[0088] From the sixth and seventh conditions in step 7.2, the following inequality can be obtained:

[0089]

[0090] therefore, is Metzler's; and

[0091]

[0092] According to the above analysis, It's Metzler.

[0093] Furthermore, in step 10, the stability of the offshore multi-reef robot control system is verified under the positive multi-agent system, and the specific contents are as follows:

[0094] 10.1 Construct the Yu Zheng Lyapunov function as follows:

[0095]

[0096] in, and

[0097] Knowable

[0098]

[0099] 10.2 Through the design of the first condition, the second condition and the gain matrix in step 8, then

[0100]

[0101] but

[0102]

[0103] According to the design of the 3rd, 4th, 5th, and 6th conditions and the gain matrix in step 8, then

[0104]

[0105] And then launched

[0106]

[0107] Through the above inequality conditions, we can further obtain

[0108]

[0109] Integrating both sides of the above equation, we can get

[0110]

[0111] So

[0112]

[0113] Combine conditions 1 to 6 of step 7.1 with condition λ 1 >1, can be converted into

[0114]

[0115] It is easy to conclude that:

[0116]

[0117] in and Ensure system stability.

[0118] The advantages and beneficial effects of the present invention are that multi-robot leadership and collaboration can jointly cope with complex environments through decentralized deployment and collaborative operations, thereby improving the adaptability and flexibility of the system, thereby achieving more efficient applications and having broad market potential and application value. Multi-robot leadership and collaboration can assign different tasks to the most suitable robots to execute according to the strengths and capabilities of each robot. In the process of tracking the target, the tracking strategy and path of each robot can be dynamically adjusted according to the target's motion trajectory and speed, to optimize task allocation and improve tracking efficiency. The use of multi-robot leadership and collaborative control has significant advantages in tracking targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Figure 1 is a schematic diagram of multi-robot target tracking;

[0120] Figure 2 This is the control structure diagram of multiple robots based on dual observers in a near-shore multi-reef environment. DETAILED DESCRIPTION

[0121] The present invention is further described below in conjunction with specific embodiments, but the present invention is not limited to these specific embodiments. Those skilled in the art should recognize that the present invention covers all possible alternatives, improvements and equivalents within the scope of the claims.

[0122] like Figure 1 and 2 As shown, the method for tracking targets by multiple robots in a near-shore multi-reef environment of the present invention comprises the following specific steps:

[0123] Step 1: Establish a state space model of the positive switching multi-agent system of the offshore multi-reef robot control system. The specific method is:

[0124] Construct a positive switching multi-agent system with multiple equilibrium points composed of n robots, where the dynamic system of the i-th agent is as follows:

[0125]

[0126] Considering the dynamics of the robot's activities and tracking targets in the offshore environment, it is ensured that the model can reflect the robot's state changes in real time. Represents the system state of the i-th robot, which contains all the state information of the robot at time t, such as position, speed, etc. represents the control input of the i-th robot, which includes all external commands or signals received by the agent at time t. represents the output of the i-th robot system, which contains all the signals or information sent by the agent at time t. Indicates interference signals from the external environment, such as the influence of tides, wind and other factors, which cause changes in water flow speed and direction. i (t) is The switching law of the ith robot, and in the finite set S = {1,2,…,N}, matrix and are the known n×n, n×r, n×s, o×n, o×s dimensional matrices respectively. And, we assume that for σ i (t)∈S matrix It's Metzler's. and They are all greater than 0, and i represents the i-th robot.

[0127] Step 2: construct a dynamic model of external interference signals of the offshore multi-reef robot control system, and its construction form is as follows:

[0128]

[0129] The offshore reef area has a complex environment with many unpredictable external interferences, such as waves, tides, and reef distribution. Designing an external interference signal model can simulate these interferences, so that the robot can take these factors into account during the design phase, thereby improving its stability and reliability in complex environments.i (t) is the additional signal for each robot, is a Metzler matrix of s×s dimensions, It is an s×s dimensional matrix greater than 0, where i represents the i-th robot.

[0130] Step 3: Establish the dynamic model of the leader of the offshore multi-reef robot control system, and its construction form is as follows:

[0131]

[0132] Among them, η(t) is the state of the leader, is a Metzler matrix of dimension n×n.

[0133] Step 4: Establish a practical consensus protocol for the offshore multi-reef robot control system, which is constructed as follows:

[0134]

[0135] In a multi-robot system, the consensus protocol can ensure that each robot clearly defines its own tasks and completes the tasks together through collaborative work. This collaborative work avoids conflicts and redundancies between robots, improves work efficiency, and helps robots make more reasonable decisions. represents a constant greater than 0, x * represents the position that the robot wants to reach, and η * Represents the position that the leader robot wants to achieve.

[0136] Step 5: Design a dual-observer exchange protocol for each robot. The design is as follows:

[0137]

[0138] In offshore environments, robots may face various external interferences and uncertainties. The dual observer exchange protocol improves the system's resistance to external interference by having two or more observers monitor the system's status simultaneously, thereby enhancing the system's robustness. When one of the observers fails or is disturbed, the other observer can detect and report the problem in a timely manner, helping the system to quickly detect and isolate the fault, ensuring the overall stability and reliability of the system. and are the estimated values ​​of the state and exogenous disturbance, and is the controller gain matrix of r×n, r×s, r×o, r×n, r×n dimensions.

[0139] Design a state observer in the following form:

[0140]

[0141] in, is a Metzler matrix of dimension n×n, and are all n×s dimensional observer gain matrices. In order to measure the disturbance of the external source signal in step 2, the disturbance observer is constructed as:

[0142]

[0143] in, is the state of the observer, is a Metzler matrix of dimension s×n, and are the observer gain matrices of s×n and s×o.

[0144] Step 6: Define the error variables of the offshore multi-reef robot control system. and ε i (t) = x i (t)-η(t), therefore, the closed-loop system is:

[0145]

[0146] Dual observers can observe the system status from different angles, providing more comprehensive and accurate information. This information helps the control system to more accurately evaluate the current environment, thereby optimizing the control strategy and improving the robot's motion performance and operating efficiency. In the complex terrain with many reefs in the offshore area, the robot needs to frequently adjust its posture and path to adapt to environmental changes. The dual observer exchange protocol can perceive environmental changes more quickly, allowing the robot to respond more quickly and enhance its ability to adapt to complex environments.

[0147] Step 7: Design the positivity conditions for the offshore multi-reef robot control system under the positive switching multi-agent system as follows:

[0148] 7.1 Accurate observation data is the basis for effective target tracking. Ensuring the normal operation of the observer can provide accurate measurement results, thereby improving the accuracy of target positioning and tracking. The following are the prerequisites for ensuring the operation of the observer, in the following form:

[0149] Design constant m>0,α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality:

[0151]

[0152] 7.2 The positivity of the system state is a necessary condition for actual operation. When tracking a target, maintaining the positivity of the system helps to ensure the upper bound of the tracking error, thereby improving the tracking performance. The following are the conditions to ensure the positivity of the offshore multi-reef robot control system:

[0153]

[0154]

[0155] for and i=1,2,…,n, holds. Among them, is a Kronecker product operator, is an r-dimensional vector whose ιth element is 1. Similarly, and is an n-dimensional and s-dimensional vector with the ι-th element being 1. n and I s Represents n-dimensional and s-dimensional unit vectors.

[0156] 7.3 Under the control protocol and dual observers, the system is positive and reaches regional consensus. Define and The gain matrix is ​​as follows:

[0157]

[0158] And the average residence time satisfies the following inequality:

[0159]

[0160] Step 8: Design the following conditions to ensure the stability of the offshore multi-reef robot control system under the positive switching multi-agent system:

[0161] 8.1 Design constant m>0, α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality:

[0162]

[0163] (μ 1 -α 1 )γ 1 -mδ 1 ≤0, (2)

[0164]

[0165] (μ 1 -α 2 )γ 2 -mδ 3 ≤0, (4)

[0166]

[0167] for and i=1,2,…,n, holds. Among them, is a Kronecker product operator, is an r-dimensional vector whose ιth element is 1. Similarly, and is an n-dimensional and s-dimensional vector with the ι-th element being 1. n and I s Represents n-dimensional and s-dimensional unit vectors.

[0168] Step 9: The positive verification process of the offshore multi-reef robot control system under the positive multi-agent system is as follows:

[0169] 9.1 Based on the 8th and 9th conditions designed in step 7.2, the gain matrix design can be rewritten as the following inequality

[0170]

[0171]

[0172] According to the first to sixth conditions in step 7.1 and the above inequality, then

[0173]

[0174] Therefore, it is concluded and

[0175] 9.2 Through the first condition of step 7.2, we get

[0176]

[0177] Then we know that the matrix It is Metzler. Combined and Easy to get According to the second condition in step 7.2, then

[0178]

[0179] Then we can get

[0180]

[0181] Same reason According to the third condition in step 7.2, we can deduce is Metzler's. From the fourth condition of step 7.2, we know that Then we can get

[0182]

[0183] According to the conditions and graph theory knowledge, and introduce

[0184]

[0185] So, there is

[0186]

[0187] And then launched

[0188]

[0189] 9.3 Binding conditions and have

[0190]

[0191] Next, from the sixth and seventh conditions in step 7.2, the following inequality can be obtained:

[0192]

[0193] therefore, Metzler's. And

[0194]

[0195] According to the above analysis, it is not difficult to conclude It is Metzler's, and the positivity of the system is guaranteed.

[0196] Step 10: The stability verification process of the offshore multi-reef robot control system under the positive multi-agent system is as follows:

[0197] 10.1 Construct a co-positive Lyapunov function as follows:

[0198]

[0199] in, and

[0200] According to the above analysis, we can know

[0201]

[0202] 10.2 Through the design of the first condition, the second condition and the gain matrix in step 8, then

[0203]

[0204] This means

[0205]

[0206] According to the design of the 3rd, 4th, 5th, and 6th conditions and the gain matrix in step 8, then

[0207]

[0208] And then launched

[0209]

[0210] Through the above inequality conditions, we can further obtain

[0211]

[0212] Integrating both sides of the above equation, we can get

[0213]

[0214] So, there is

[0215]

[0216] Combine conditions 1 to 6 of step 7.1 with condition λ 1 >1, can be converted into

[0217]

[0218] It is easy to conclude that:

[0219]

[0220] in and Therefore, system stability can be guaranteed.

Claims

1. A multi-robot target tracking method in a near-shore multi-reef environment. Features The steps include: Step 1: Establish a state space model of the positive switching multi-agent system of the offshore multi-reef robot control system; Step 2: Establish a dynamic model of external interference signals of the offshore multi-reef robot control system; Step 3, establish the dynamic model of the leader of the offshore multi-reef robot control system; Step 4: construct the actual consensus protocol of the offshore multi-reef robot control system; Step 5, design the offshore multi-reef robot control system based on the dual observer exchange protocol; Step 6: Establish a closed-loop matrix of the offshore multi-reef robot control system; Step 7: Design the positivity conditions for the offshore multi-reef robot control system under the positive switching multi-agent system; Step 8: Design the stability conditions of the offshore multi-reef robot control system under the positive switching multi-agent system; Step 9, verify the positivity of the offshore multi-reef robot control system under the positive multi-agent system; Step 10: Verify the stability of the offshore multi-reef robot control system under the positive multi-agent system.

2. A method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 1, Features: The construction of the state space model of the positive switching multi-agent system of the offshore multi-reef robot control system in step 1 is as follows: in, represents the system state of the i-th robot, including all the state information of the robot at time t; represents the control input of the i-th robot, which contains all external commands or signals received by the agent at time t; represents the output of the i-th robot system, which contains all the signals or information sent by the agent at time t; Represents the interference signal from the external environment; σ i (t) is The switching law of the ith robot, and in the finite set S = {1,2,...,N}, matrix and are the known n×n, n×r, n×s, o×n, o×s dimensional matrices respectively; and σ i (t)∈S, assuming that for the matrix It's Metzler's. and are all greater than 0, i represents the i-th robot.

3. A method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 2, Features: The dynamic model of the external interference signal of the offshore multi-reef robot control system in step 2 is constructed as follows: Among them, ξ i (t) is the additional signal for each robot, is a Metzler matrix of s×s dimensions, is a matrix of s×s dimensions greater than 0; in step 3, a dynamic model of the leader of the offshore multi-reef robot control system is established, and its construction form is as follows: Among them, η(t) is the state of the leader, is a Metzler matrix of dimension n×n.

4. A method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 3, Features: The actual consensus protocol for constructing the offshore multi-reef robot control system in step 4 is constructed as follows: where h, represents a constant greater than 0, x * represents the position that the robot wants to reach, and η * Represents the position that the leader robot wants to achieve.

5. The method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 4, Features: In step 5, the offshore multi-reef robot control system is designed based on the exchange protocol of two observers, and its construction form is as follows: in, and are the estimated values ​​of the state and exogenous disturbance, and is the controller gain matrix of r×n, r×s, r×o, r×n, r×n dimensions; Design a state observer in the following form: in, is a Metzler matrix of dimension n×n, and Both are n×s-dimensional observer gain matrices; The disturbance observer is constructed as: in, is the state of the observer, is a Metzler matrix of dimension s×n, and are the observer gain matrices of s×n and s×o.

6. A method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 5, Features: The closed-loop matrix of the offshore multi-reef robot control system in step 6 is constructed as follows: Define the error variables of the offshore multi-reef robot control system, and ε i (t) = x i (t)-η(t), therefore, the closed-loop system is:

7. A method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 6, characterized in that: in step 7, the positivity conditions for ensuring the offshore multi-reef robot control system under the positive switching multi-agent system are designed, and the construction conditions are as follows: 7.1 The prerequisites for ensuring the operation of the observer are as follows: Design constant m>0,α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality: 7.2 The positive conditions for ensuring the control system of the offshore multi-reef robot are as follows: for and i=1,2,…,n, hold; where, is the Kronecker product operator, is an r-dimensional vector whose ιth element is 1. Similarly, and is an n-dimensional and s-dimensional vector whose ιth element is 1; I n and I s represents n-dimensional and s-dimensional unit vectors; 7.3 Under the control protocol and dual observers, the system is positive and reaches regional consensus. Define and The gain matrix is ​​as follows: And the average residence time satisfies the following inequality:

8. The method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 7, Features: In step 8, the stability conditions of the offshore multi-reef robot control system are designed under the positive switching multi-agent system. The construction conditions are as follows: 8.1 Design constant m>0, α 1 <0, α 2 <0,λ 1 >1,μ 1 >0,δ 2 >0,δ 3 >0,δ 4 >0,δ 5 >0,n-dimensional vector o-dimensional vector s-dimensional vector This makes the following inequality: (m 1 -a 1 ) γ 1 -mδ 1 ≤0, (2) (m 1 -a 2 ) γ 2 -mδ 3 ≤0, (4) for and i=1,2,…,n, holds; 8.2 Under the control protocol and dual observers, the system is positive and reaches regional consensus. Define and The gain matrix is ​​as follows: And the average residence time satisfies the following inequality:

9. The method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 8, Features: In step 9, the positivity of the offshore multi-reef robot control system is verified under the positive multi-agent system, as follows: 9.1 Based on the 8th and 9th conditions designed in step 7.2, the gain matrix design can be rewritten as the following inequality According to the 1st to 6th conditions in step 7.1 and the above inequality, Therefore, it is concluded and 9.2 Through the first condition of step 7.2, we get matrix is Metzler's; combined and Easy to get According to the second condition in step 7.2, then Then we can get Same reason According to the third condition in step 7.2, we can deduce is Metzler's; from the fourth condition of step 7.2, we know Then we can get According to the conditions and graph theory knowledge, and introduce So And then launched 9.3 Binding conditions and have From the sixth and seventh conditions in step 7.2, the following inequality can be obtained: therefore, is Metzler's; and According to the above analysis, It's Metzler.

10. The method for tracking targets by multiple robots in an offshore multi-reef environment according to claim 9, Features: In step 10, the stability of the offshore multi-reef robot control system is verified under the multi-agent system. The specific contents are as follows: 10.1 Construct the Yu Zheng Lyapunov function as follows: in, and Knowable 10.2 Through the design of the first condition, the second condition and the gain matrix in step 8, then but According to the design of the 3rd, 4th, 5th, and 6th conditions and the gain matrix in step 8, then And then launched Through the above inequality conditions, we can further obtain Integrating both sides of the above equation, we can get So Combine conditions 1 to 6 of step 7.1 with condition λ 1 >1, can be converted into It is easy to conclude that: in and Ensure system stability.