Anti-delay multi-mobile robot formation control method based on constraint following

Through the anti-delay multi-mobile robot formation control method based on constraint follow-up, the problem of the performance of multi-mobile robot system in complex trajectories and uncertain environments is solved, the consistency and stability of the system are achieved, and the scope of application is broadened.

CN119024840APending Publication Date: 2024-11-26ANHUI UNIV
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411122493.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-26

Smart Images

  • Figure CN119024840A_ABST
    Figure CN119024840A_ABST
Patent Text Reader

Abstract

The invention discloses an anti-delay multi-mobile robot formation control method based on constraint following, and relates to the technical field of robot automation and intelligent driving, and the method comprises the following steps: S1, building a two-dimensional dynamic model of a mobile robot group with time-varying bounded uncertainty; s2, converting a formation control problem into a constraint following problem; s3, designing an adaptive robust control strategy of approximate constraint following; s4, performing consistency analysis on the multi-mobile robot system with the time delay; the formation control problem of a multi-robot cooperative system is expanded from a traditional linear problem to a generalized planar problem, on the basis, the two-dimensional formation control problem is researched, longitudinal movement and transverse movement are considered at the same time, and the application of the autonomous multi-robot group system is no longer limited to robots in a road; meanwhile, the track is not limited to a preset road, possibly any route determined by the working scene, and possibly any route determined by the working scene.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of robot automation and intelligent driving technology, and in particular to a time-delay-resistant multi-mobile robot formation control method based on constraint following. Background Art

[0002] With the development of technology, various autonomous functional robots have emerged, such as unmanned road rollers (engineering field), multi-UAV formation target strike (military field), multi-AGV storage and handling system (industrial field), etc. With the development of Internet of Things technology, these autonomous functional robots can also be connected into an autonomous multi-robot swarm system to perform tasks in a coordinated manner. In this sense, the application of autonomous multi-mobile robot swarm systems is no longer limited to specific scenarios; at the same time, the trajectory is no longer a preset road, but may be an arbitrary route determined by the work scenario.

[0003] A multi-mobile robot system refers to a collection of robots with multiple autonomous functions. The goal is to enable several simple intelligent systems that are easy to manage and control to achieve complex intelligence through mutual cooperation. Compared with single autonomous function robots, multi-mobile robot systems have the advantages of coping with complex and uncertain situations, and having stronger autonomy and adaptability to location and environment. In a multi-mobile robot system, delay is one of the key factors affecting the collaborative performance of multiple robots. Due to various network environmental factors such as sensor fading, packet loss, channel noise and random link failures, delay is inherent in the communication interaction between robots. Therefore, a delay-resistant multi-mobile robot system plays a vital role in practical applications.

[0004] There is an approximate constraint following problem in multi-mobile robot systems. The Udwadia-Kalaba theory provides the basic equations of constrained dynamics, which combines the control problem with the solution of constraint forces and is used to deal with complete and incomplete constraints in the dynamic equations. Previous studies have focused on the linear motion of multi-robot cooperative control systems, which is not enough. In actual needs, the robot trajectories are complex and diverse and have uncertainty, so more exploration is needed on planar motion, including longitudinal and lateral motion. In recent years, by introducing state transformation in control design, the proposed control method can make the system present collision avoidance and stability without being affected by system uncertainty. However, system uncertainty is an inevitable problem in multi-robot cooperative systems and has an important impact on system performance. Therefore, uncertainty processing is of great significance in precise formation control. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides the following technical solution: a time-delay-resistant multi-mobile robot formation control method based on constraint following, comprising the following steps:

[0006] S1. Establish a two-dimensional dynamic model of mobile robot swarm with time-varying bounded uncertainty, introduce the constraint following idea, take the ideal group performance as the expected constraint, and construct the dynamic model of multi-mobile robot system;

[0007] S2. Transform the formation control problem into a constrained following problem, and perform a differential homeomorphic transformation on the spatial measurement between robots so that the spatial measurement is subject to positive constraints;

[0008] S3. Design an adaptive robust control strategy for approximate constraint following, define the performance index β as the control object for evaluating the constraint following error, optimize the control parameters in the adaptive robust constraint following controller, and substitute the obtained optimal parameters into the controller to make the multi-mobile robot system uniformly bounded and uniformly ultimately bounded;

[0009] S4. Conduct consistency analysis on multi-mobile robot systems with time delays, and propose necessary and sufficient conditions and control algorithms for multi-mobile robot systems with time delays to achieve consistency under the network topology of directed spanning trees.

[0010] S5. Finally, analyze the overall control effect of the system and draw conclusions.

[0011] As an improvement of the above technical solution, in step S1, the ideal group performance is used as the expected constraint to construct a dynamic model of the robot group system, in which the leader is regarded as a virtual robot, and its dynamic equation is described as:

[0012]

[0013] in, is acceleration, t∈R represents time, M0∈R 2 , M0>0 represents the inertia matrix;

[0014] Let X0 = [x0 y0] T , X i :=[x i yi ] T Represent the positions of the leader and follower, respectively. Denote the control inputs of the leader and the follower respectively, then the dynamic equation of the multi-robot collaborative system is described as:

[0015]

[0016] in, Indicates speed, represents acceleration, is a time-varying and bounded uncertain parameter, is bounded and uncertain, and is used to represent the system uncertainty parameter σ i The boundary of M i ∈R 2 , M i >0 is the inertia matrix of the system, represents the aerodynamic resistance, Represents the combined rolling resistance, acceleration resistance and other external disturbances. In addition, M i (·),C i (·) and F i (·) are all continuous functions;

[0017] The actual horizontal distance between robot i and the preceding robot i-1 is defined as:

[0018]

[0019] The actual vertical distance is:

[0020]

[0021] Number the robots naturally, that is, assume that robot i comes after robot i-1, and get x i-1 -x i >0, where |x i-1 (t)-x i (t)| represents the horizontal distance between two adjacent robots with their center of gravity as the reference point, |y i-1 (t)-y i (t)| represents the vertical distance between two adjacent robots with their center of gravity as the reference point, represents the robot radius of robot i-1 with its center of gravity, represents the robot radius with robot i as the center of gravity.

[0022] As an improvement of the above technical solution, in step S1, the horizontal space metric is introduced as follows:

[0023]

[0024] The vertical space measurement is introduced as:

[0025]

[0026] As an indirect measurement of the space between the two robots, the spacing error can be expressed as:

[0027]

[0028]

[0029] in, represents the expected horizontal distance of robot i, represents the expected vertical distance of robot i;

[0030] Therefore, the spacing error obtained based on the horizontal or vertical space measurement can be expressed as:

[0031]

[0032]

[0033] in, is the horizontal spacing error, is the vertical spacing error.

[0034] As an improvement of the above technical solution, in step S2, an indirect control method for horizontal space measurement and vertical space measurement is included, that is, collision avoidance state transformation, which can be expressed as:

[0035]

[0036] in, For additional variables, and

[0037] After inverse transformation of the spacing error, we get:

[0038]

[0039] Substituting the horizontal and vertical spatial metrics into the inverse transform expression for the spacing error yields:

[0040]

[0041] Take x i and i The second-order derivative of t is:

[0042]

[0043]

[0044] in, Defined as Defined as and and All are positive scalars;

[0045] make:

[0046]

[0047] get:

[0048]

[0049] set up q i ∈R 2 , so there is a first-order derivative And the second-order derivative

[0050] in,

[0051]

[0052] Therefore, (17) can be converted into matrix form as follows:

[0053]

[0054] Combining the system dynamics equation, we can get:

[0055]

[0056] The dynamic equation of the multi-mobile robot swarm system is rewritten as:

[0057]

[0058] Among them, M i ∈R 2 , M i >0 is the inertia matrix of the system, represents the aerodynamic resistance, Represents the combined rolling resistance, acceleration resistance and other external disturbances. In addition, M i (·),C i (·) and F i (·) are all continuous functions.

[0059] As an improvement of the above technical solution, in step S3, the ideal motion performance is used as the expected constraint, and a performance index β is designed to evaluate the constraint following error:

[0060] The ideal motion performance is taken as the first-order constraint on the robot:

[0061]

[0062] Among them, q i ∈R n is the state vector, G ij (·):R n ×R n →R represents the influence of agent j on agent i;

[0063] make:

[0064]

[0065] in, q∈R 2N ;

[0066] Take the ideal group performance as the desired constraint:

[0067]

[0068] Taking its first-order derivative with respect to time t, the second-order constraint is:

[0069]

[0070] in, Represents G i (q) the first derivative with respect to t;

[0071] The constrained following error is defined as:

[0072]

[0073] in, It represents the degree of expected constraint following. For complete expected following, β i ≡0;

[0074] Take the first derivative of β with respect to t:

[0075]

[0076] The β dynamics equation is described as:

[0077]

[0078] As an improvement of the above technical solution, in step S3, for uncertainty processing, M i , C i and F i The decomposition is as follows:

[0079]

[0080]

[0081]

[0082] Among them, M i ∈R 2 is the inertia matrix, is the aerodynamic drag, To combine rolling resistance, acceleration resistance and other external disturbances, and is the nominal part, ΔM i ,ΔC i and ΔF i is the uncertainty part;

[0083] when Symmetrical and positive timing, ΔM i (·), ΔC i (·), and ΔF i (·) are all continuous functions, let:

[0084]

[0085]

[0086]

[0087]

[0088] therefore,

[0089]

[0090] The decomposition of nominal control items is expressed as:

[0091]

[0092] For uncertain system control, the β dynamics equation can be rewritten as:

[0093]

[0094] Among them, M i ∈R 2 is the inertia matrix, is the aerodynamic drag, To combine rolling resistance, acceleration resistance and other external disturbances, and is the nominal part, ΔM i ,ΔC i and ΔF i The uncertainty part.

[0095] As an improvement of the above technical solution, in step S3, an adaptive robust control strategy of approximate constraint following is designed:

[0096]

[0097] in,

[0098]

[0099] Among them, λ i >0 is a constant design parameter, function Parameters in The leaky adaptive law is as follows:

[0100]

[0101] Among them, the scalar is the design parameter, α i represents the comprehensive impact of uncertainty on system performance, Represents the adaptive parameter α i Adaptive robust action.

[0102] As an improvement of the above technical solution, in step S3, the control parameters in the adaptive robust constraint following controller are optimized and selected so that the multi-robot collaborative system is uniformly bounded and uniformly ultimately bounded.

[0103] As an improvement of the above technical solution, in step S4, a consistency analysis is performed on the multi-mobile robot system with time delay:

[0104] The time-delay system is expressed as:

[0105]

[0106] where A1∈R n×n and A2∈R n×n is a constant matrix, x∈R n Represents state variables;

[0107] After the neutral model transformation, the time-delay system is obtained:

[0108]

[0109] For any s∈C+, when When , the stability of system (50) is the same as that of system (49);

[0110] Among them, A1∈R n×n and A2∈R n×n is a constant matrix, x∈R n represents the state variable, λ i (A) represents the i-th eigenvalue of matrix A, C+ represents the set of complex numbers with positive real part;

[0111] For any initial condition, the state of each robot in the system satisfies the consistency constraint:

[0112]

[0113] Among them, qi (t)-q j (t) represents the displacement of each mobile robot, represents the speed of each mobile robot;

[0114] When there is time delay in the mobile robot system, the robot consistency constraint in the system is expressed as:

[0115]

[0116] It can be further expressed as a vector:

[0117]

[0118] Where D∈R N×N , A∈R N×N They represent the in-degree matrix and the adjacent-order matrix of the directed graph G respectively, represents the Kronecker product, I n represents the n×n identity matrix, and τ represents the time lag.

[0119] As an improvement of the above technical solution, in step S4, under the network topology of a directed spanning tree, the necessary and sufficient conditions and control algorithm for the delay-resistant multi-mobile robot system to reach consistency are proposed:

[0120] In order to eliminate the constraint error in the control algorithm, The values ​​of k1 and k2 are appropriately selected so that the vector is globally asymptotically stable;

[0121] Taking k1=α, k2=β, we can get the second-order formal constraints of the mobile robot system:

[0122]

[0123] Among them, α and β are positive constants representing the coupling strength;

[0124] The control algorithm for achieving consistency of the multi-mobile robot system with time delay is obtained from the second-order constraints and the system dynamics equations as follows:

[0125]

[0126] Where M(q,t)∈R n×n is the inertia matrix, q∈R n are generalized coordinates, is a known force, N∈R n×n is a given positive definite matrix;

[0127] Let N = M -2 get:

[0128]

[0129] The i-th component can be expressed as:

[0130]

[0131] Substitute the control algorithm into the system dynamics equation and express it in matrix form as:

[0132]

[0133] Where D∈R N×N , A∈R N×N They represent the in-degree matrix and the adjacent-order matrix of the directed graph G respectively, represents the Kronecker product, I n represents the n×n identity matrix, τ represents the time delay, and α and β represent the coupling strength.

[0134] Beneficial effects of the present invention:

[0135] 1. The present invention extends the formation control problem of a multi-robot collaborative system from a traditional linear problem to a generalized planar problem, and on this basis, studies the two-dimensional formation control problem, while taking into account longitudinal and lateral motion, so that the application of autonomous multi-robot swarm systems is no longer limited to robots on the road; at the same time, the trajectory is no longer limited to preset roads, but may be any route determined by the work scenario, broadening the actual use scope of multi-robot collaborative operations and making the system more autonomous and adaptable to location and environment.

[0136] 2. The present invention proposes to use the ideal group performance of a multi-robot collaborative system as the desired constraint and construct a constraint following framework to achieve a tight formation. In order to avoid collision, the spatial measure between the two robots is creatively transformed by differential homeomorphism, so that the spatial measure is subject to positive constraints, while the transformed variables (i.e., group variables) are not restricted, thereby freeing the control design from state constraints.

[0137] 3. The present invention takes the ideal group performance as the expected constraint, expresses the group control problem as a constraint following task, and defines the β measure as the control object of the constraint following error. By selecting the parameters of the adaptive robust controller, the β measure is made uniformly bounded and uniformly ultimately bounded without considering the uncertainty, driving the controlled system to approximately follow the expected constraints, thereby presenting the ideal group performance.

[0138] 4. The present invention proposes a control algorithm to achieve the consistency of a multi-mobile robot system with time delay. Under this algorithm, the consistency conditions of the multi-mobile robot system with time delay are determined.

[0139] 5. The present invention adopts an adaptive robust constrained following controller to study an adaptive robust group control scheme for multi-robot collaborative control systems, and specially considers the uncertainty of time-varying but bounded multi-robot collaborative systems. For the case where the uncertainty boundary is unknown, an adaptive law is proposed to instantly estimate its comprehensive impact on the system performance, and it can adapt to the unknown but bounded uncertainty in the system. By adjusting the control based on the tracking error and designing an adaptive law, the comprehensive control effect of the controller on the multi-robot collaborative system based on constraint following is improved while ensuring the stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0140] Figure 1 It is a schematic diagram of the formation control method of the present invention;

[0141] Figure 2 It is a schematic diagram of the process of the present invention;

[0142] Figure 3 This is a schematic diagram of the delay-resistant multi-mobile robot collaborative system based on constraint following of the present invention. DETAILED DESCRIPTION

[0143] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0144] See also Figure 1-3 The present invention provides a technical solution: a time-delay-resistant multi-mobile robot formation control method based on constraint following, comprising the following steps:

[0145] S1. Establish a two-dimensional dynamic model of mobile robot swarm with time-varying bounded uncertainty, introduce the constraint following idea, take the ideal group performance as the expected constraint, and construct the dynamic model of multi-mobile robot system;

[0146] S2. Transform the formation control problem into a constrained following problem, and perform a differential homeomorphic transformation on the spatial measurement between robots so that the spatial measurement is subject to positive constraints;

[0147] S3. Design an adaptive robust control strategy for approximate constraint following, define the performance index β as the control object for evaluating the constraint following error, optimize the control parameters in the adaptive robust constraint following controller, and substitute the obtained optimal parameters into the controller to make the multi-mobile robot system uniformly bounded and uniformly ultimately bounded;

[0148] S4. Conduct consistency analysis on multi-mobile robot systems with time delays, and propose necessary and sufficient conditions and control algorithms for multi-mobile robot systems with time delays to achieve consistency under the network topology of directed spanning trees.

[0149] S5. Finally, analyze the overall control effect of the system and draw conclusions.

[0150] The present invention expands the formation control problem of a multi-robot collaborative system from a traditional linear problem to a generalized planar problem, and on this basis, studies the two-dimensional formation control problem, while taking into account longitudinal and lateral motion, so that the application of autonomous multi-robot swarm systems is no longer limited to robots on the road; at the same time, the trajectory is no longer limited to preset roads, but may be any route determined by the work scenario, broadening the actual use scope of multi-robot collaborative operations and making the system more autonomous and adaptable to location and environment.

[0151] Specifically, in step S1, the ideal group performance is used as the expected constraint to construct a dynamic model of the robot group system, in which the leader is regarded as a virtual robot, and its dynamic equation is described as:

[0152]

[0153] in, is acceleration, t∈R represents time, M0∈R 2 , M0>0 represents the inertia matrix;

[0154] Let X0 = [x0 y0] T , X i :=[x i y i ] T Represent the positions of the leader and follower, respectively. Denote the control inputs of the leader and the follower respectively, then the dynamic equation of the multi-robot collaborative system is described as:

[0155]

[0156] in, Indicates speed, represents acceleration, is a time-varying and bounded uncertain parameter, is bounded and uncertain, and is used to represent the system uncertainty parameter σ i The boundary of M i ∈R 2 , M i >0 is the inertia matrix of the system, represents the aerodynamic resistance, Represents the combined rolling resistance, acceleration resistance and other external disturbances. In addition, M i (·),C i (·) and F i (·) are all continuous functions;

[0157] The actual horizontal distance between robot i and the preceding robot i-1 is defined as:

[0158]

[0159] The actual vertical distance is:

[0160]

[0161] Number the robots naturally, that is, assume that robot i comes after robot i-1, and get x i-1 -x i >0, where |x i-1 (t)-x i (t)| represents the horizontal distance between two adjacent robots with their center of gravity as the reference point, |y i-1 (t)-y i (t)| represents the vertical distance between two adjacent robots with their center of gravity as the reference point, represents the robot radius of robot i-1 with its center of gravity, represents the robot radius with robot i as the center of gravity.

[0162] Specifically, in step S1, the horizontal space metric is introduced as follows:

[0163]

[0164] The vertical space measurement is introduced as:

[0165]

[0166] As an indirect measurement of the space between the two robots, the spacing error can be expressed as:

[0167]

[0168]

[0169] in, represents the expected horizontal distance of robot i, represents the expected vertical distance of robot i;

[0170] Therefore, the spacing error obtained based on the horizontal or vertical space measurement can be expressed as:

[0171]

[0172]

[0173] in, is the horizontal spacing error, is the vertical spacing error.

[0174] Specifically, in step S2, an indirect control method for the horizontal space metric and the vertical space metric is included, namely, the collision avoidance state transformation, which can be expressed as:

[0175]

[0176] in, is an additional variable, and

[0177] After inverse transformation of the spacing error, we get:

[0178]

[0179] Substituting the horizontal and vertical spatial metrics into the inverse transform expression for the spacing error yields:

[0180]

[0181] Take x i and i The second-order derivative of t is:

[0182]

[0185] in, Defined as Defined as and and All are positive scalars;

[0186] make:

[0187]

[0188] get:

[0189]

[0190] set up q i ∈R 2 , so there is a first-order derivative And the second-order derivative

[0191] in,

[0192]

[0193] Therefore, (17) can be converted into matrix form as follows:

[0194]

[0195] Combining the system dynamics equation, we can get:

[0196]

[0197] The dynamic equation of the multi-mobile robot swarm system is rewritten as:

[0198]

[0199] Among them, M i ∈R 2 , M i >0 is the inertia matrix of the system, represents the aerodynamic resistance, Represents the combined rolling resistance, acceleration resistance and other external disturbances. In addition, M i (·),C i (·) and F i (·) are all continuous functions.

[0200] The present invention proposes to use the ideal group performance of a multi-robot collaborative system as the expected constraint and construct a constraint following framework to achieve tight formation. In order to avoid collision, the spatial measure between the two robots is creatively transformed by differential homeomorphism, so that the spatial measure is subject to positive constraints, while the transformed variables (i.e., group variables) are not restricted, thereby freeing the control design from state constraints.

[0201] Specifically, in step S3, the ideal motion performance is used as the desired constraint, and the performance index β used to evaluate the constraint following error is designed:

[0202] The ideal motion performance is taken as the first-order constraint on the robot:

[0203]

[0204] Among them, q i ∈R n is the state vector, G ij (·):R n ×R n →R represents the influence of agent j on i;

[0205] make:

[0206]

[0207] in, q∈R 2N ;

[0208] Take the ideal group performance as the desired constraint:

[0209]

[0210] Taking its first-order derivative with respect to time t, the second-order constraint is:

[0211]

[0212] in, Represents G i (q) the first derivative with respect to t;

[0213] The constrained following error is defined as:

[0214]

[0215] in, It represents the degree of expected constraint following. For complete expected following, β i ≡0;

[0216] Take the first derivative of β with respect to t:

[0217]

[0218] The β dynamics equation is described as:

[0219]

[0220] Specifically, in step S3, for uncertainty processing, M i , C i and F i The decomposition is as follows:

[0221]

[0222]

[0223]

[0224] Among them, M i ∈R 2 is the inertia matrix, is the aerodynamic drag, To combine rolling resistance, acceleration resistance and other external disturbances, and is the nominal part, ΔM i ,ΔC i and ΔF i is the uncertainty part;

[0225] when Symmetrical and positive timing, ΔM i (·), ΔC i (·), and ΔF i (·) are all continuous functions, let:

[0226]

[0227]

[0228]

[0229]

[0230] therefore,

[0231]

[0232] The decomposed nominal control items are expressed as:

[0233]

[0234] For uncertain system control, the β dynamics equation can be rewritten as:

[0235]

[0236] Among them, M i ∈R 2 is the inertia matrix, is the aerodynamic drag, To combine rolling resistance, acceleration resistance and other external disturbances, and is the nominal part, ΔM i ,ΔC i and ΔF i The uncertainty part.

[0237] Specifically, in step S3, an adaptive robust control strategy of approximate constraint following is designed:

[0238]

[0239] in,

[0240]

[0241] Among them, λ i >0 is a constant design parameter, function Parameters in The leaky adaptive law is as follows:

[0242]

[0243] Among them, the scalar is the design parameter, α i represents the comprehensive impact of uncertainty on system performance, Represents the adaptive parameter α i Adaptive robust action.

[0244] Specifically, in step S3, the control parameters in the adaptive robust constraint following controller are optimized and selected so that the multi-robot collaborative system is uniformly bounded and uniformly ultimately bounded:

[0245] Assumption 1: There exists a constant κ i ∈(0,∞) and function f i (κ i ,·):R 2 ×R→R 2 , so that the Lyapunov function V i (·):R 2 ×R→R + and a strictly increasing function R + →R + satisfy:

[0246]

[0247] For any

[0248]

[0249]

[0250] Assumption 2: There is a possibly unknown constant For all (X i ,t)∈R 2 ×R exists:

[0251]

[0252] Among them, λ m (·) is the minimum eigenvalue of the incidence matrix, is usually an unknown constant and

[0253] Assumption 3: There is an unknown constant and an unknown function Π i (·): For any σ i ∈∑i exist:

[0254]

[0255] For all function Existence, for any have:

[0256]

[0257] in, A constant parameter that makes each component of the function non-decreasing;

[0258] Select the controller parameters as follows:

[0259]

[0260] Among them, M i ∈R 2 is the inertia matrix, is the aerodynamic drag, To combine rolling resistance, acceleration resistance and other external disturbances, and is the nominal part;

[0261] Based on assumptions 1-3 and controller parameter selection, the multi-robot collaborative system exhibits the following performance:

[0262] 1) Uniform boundedness: for any r>0, there exists d(r)<∞ such that when ‖δ(t0)‖≤r, for any t≥t0, there exists ‖δ(t)‖≤d(r);

[0263] 2) Uniformly bounded: For any r>0 and ‖δ(t0)‖≤r, there exists and So that when When, for any exist

[0264] Specifically, in step S4, a consistency analysis is performed on the multi-mobile robot system with time delay:

[0265] The time-delay system is expressed as:

[0266]

[0267] where A1∈R n×n and A2∈R n×n is a constant matrix, x∈R n Represents state variables;

[0268] After the neutral model transformation, the time-delay system is obtained:

[0269]

[0270] For any s∈C+, when When , the stability of system (50) is the same as that of system (49);

[0271] Among them, A1∈R n×n and A2∈R n×n is a constant matrix, x∈R n represents the state variable, λ i (A) represents the i-th eigenvalue of matrix A, C+ represents the set of complex numbers with positive real part;

[0272] For any initial condition, the state of each robot in the system satisfies the consistency constraint:

[0273]

[0274] Among them, q i (t)-q j (t) represents the displacement of each mobile robot, represents the speed of each mobile robot;

[0275] When there is time delay in the mobile robot system, the robot consistency constraint in the system is expressed as:

[0276]

[0277] It can be further expressed as a vector:

[0278]

[0279] Where D∈R N×N , A∈R N×N They represent the in-degree matrix and the adjacent-order matrix of the directed graph G respectively, represents the Kronecker product, I n represents the n×n identity matrix, and τ represents the time lag.

[0280] The present invention takes the ideal group performance as the expected constraint, expresses the group control problem as a constraint following task, and defines the β measure as the control object of the constraint following error. By selecting the parameters of the adaptive robust controller, the β measure is made uniformly bounded and uniformly ultimately bounded without considering the uncertainty, and the controlled system is driven to approximately follow the expected constraints, thereby presenting the ideal group performance.

[0281] Specifically, in step S4, under the network topology of a directed spanning tree, the necessary and sufficient conditions and control algorithm for the delay-resistant multi-mobile robot system to reach consensus are proposed:

[0282] In order to eliminate the constraint error in the control algorithm, The values ​​of k1 and k2 are appropriately selected so that the vector is globally asymptotically stable;

[0283] Taking k1=α, k2=β, we can get the second-order formal constraints of the mobile robot system:

[0284]

[0285] Among them, α and β are positive constants representing the coupling strength;

[0286] The control algorithm for achieving consistency of the multi-mobile robot system with time delay is obtained from the second-order constraints and the system dynamics equations as follows:

[0287]

[0288] Where M(q,t)∈R n×n is the inertia matrix, q∈R n are generalized coordinates, is a known force, N∈R n×n is a given positive definite matrix;

[0289] Let N = M -2 get:

[0290]

[0291] The i-th component can be expressed as:

[0292]

[0293] Substitute the control algorithm into the system dynamics equation and express it in matrix form as:

[0294]

[0295] Where D∈R N×N , A∈R N×N They represent the in-degree matrix and the adjacent-order matrix of the directed graph G respectively, represents the Kronecker product, I n represents the n×n identity matrix, τ represents the time delay, α and β represent the coupling strength;

[0296] Assumption 4: The directed graph G has a directed spanning tree;

[0297] Lemma 1: Under Assumption 4, the Laplacian matrix of the directed graph G is has one eigenvalue 0, and all other eigenvalues ​​have positive real parts;

[0298] When Assumption 4 holds, the consistency condition for system (58) without time delay τ = 0 is expressed as:

[0299]

[0300] Among them, ν i (i=2,3,...,N) is the Laplace matrix Non-zero eigenvalues ​​of ;

[0301] According to Lemma 1, when there is a time delay τ>0, the singular vector decomposition is:

[0302]

[0303] Among them, the last column of W is 1 N , W -1 The last line of p T , 1 N and p T are the Laplacian matrices associated with zero eigenvalues The left and right eigenvectors of ;

[0304] From A=DL, we get:

[0305]

[0306] in,

[0307] Decoupling the dynamic equations of the multi-mobile robot system with time delay yields:

[0308]

[0309]

[0310] in, represents the first N-1 rows of r(t), represents the first N-1 rows of ν(t), represents the last row of r(t), The last row of ν(t) is represented by I N-1 represents the N-1×N-1 identity matrix, D∈R N×N represents the in-degree matrix of the directed graph G;

[0311] Therefore, under Assumption 4, the necessary and sufficient condition for a multi-mobile robot system with time delay τ∈[0,τ0) to reach consistency is:

[0312]

[0313]

[0314]

[0315] in, represents the set of complex numbers with positive real part, P,S∈R (2N-2)×(2N-2) represents the preferred positive definite matrix, (A1+A2) T P+P(A1+A2) represents a preferred negative definite matrix;

[0316] The present invention proposes a control algorithm to achieve the consistency of a multi-mobile robot system with time delay. Under the algorithm, the consistency condition of the multi-mobile robot system with time delay is determined.

[0317] By substituting the parameters into the controller and configuring the required hardware interface modules in Matlab / Simulink, a graphical model of the control method is built; secondly, the Matlab / Simulink model is automatically converted into executable code and downloaded to the main controller; then, the test data measured by the sensor is observed by the host computer; finally, the obtained test data is saved, processed and analyzed.

[0318] The present invention adopts an adaptive robust constraint following controller to study an adaptive robust group control scheme for a multi-robot collaborative control system, and specially considers the uncertainty of a time-varying but bounded multi-robot collaborative system. For the case where the uncertainty boundary is unknown, an adaptive law is proposed to instantly estimate its comprehensive impact on the system performance, and it is able to adapt to the unknown but bounded uncertainty in the system. By adjusting the control by designing an adaptive law based on tracking error, the comprehensive control effect of the controller on the multi-robot collaborative system based on constraint following is improved while ensuring the stability of the system.

[0319] The above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them.

Claims

1. Anti-delay multi-mobile robot formation control method based on constraint following, It is characterized in that The following steps are involved: S1. Establish a two-dimensional dynamic model of mobile robot swarm with time-varying bounded uncertainty, introduce the constraint following idea, take the ideal group performance as the expected constraint, and construct the dynamic model of multi-mobile robot system; S2. Transform the formation control problem into a constrained following problem, and perform a differential homeomorphic transformation on the spatial measurement between robots so that the spatial measurement is subject to positive constraints; S3. Design an adaptive robust control strategy for approximate constraint following, define the performance index β as the control object for evaluating the constraint following error, optimize the control parameters in the adaptive robust constraint following controller, and substitute the obtained optimal parameters into the controller to make the multi-mobile robot system uniformly bounded and uniformly ultimately bounded; S4. Conduct consistency analysis on multi-mobile robot systems with time delays, and propose necessary and sufficient conditions and control algorithms for multi-mobile robot systems with time delays to achieve consistency under the network topology of directed spanning trees. S5. Finally, analyze the overall control effect of the system and draw conclusions.

2. According to claim 1, the time-delay resistant multi-mobile robot formation control method based on constraint following, Features: In step S1, the ideal group performance is used as the expected constraint to construct a dynamic model of the robot group system, in which the leader is regarded as a virtual robot, and its dynamic equation is described as: The dynamic equation of the multi-robot collaborative system is described as: The actual horizontal distance between robot i and the preceding robot i-1 is defined as: The actual vertical distance is: Number the robots naturally, that is, assume that robot i comes after robot i-1, and get x i-1 -x i >0.

3. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 2, Features: In step S1, the horizontal spatial metric is introduced as: The vertical space measurement is introduced as: As an indirect measurement of the space between the two robots, the spacing error can be expressed as: Therefore, the spacing error obtained based on the horizontal or vertical space measurement can be expressed as:

4. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 1, Features: In step S2, an indirect control method for the horizontal space metric and the vertical space metric is included, namely, the collision avoidance state transformation, which can be expressed as: After inverse transformation of the spacing error, we get: Substituting the horizontal and vertical spatial metrics into the inverse transform expression for the spacing error yields: Take x i and i The second-order derivative of t is: make: get: set up So there is a first-order derivative And the second-order derivative in, Therefore, (17) can be converted into matrix form as follows: Combining the system dynamics equation, we can get: The dynamic equation of the multi-mobile robot swarm system is rewritten as:

5. The delay-resistant multi-mobile robot formation control method based on constraint following according to claim 1, Features: In step S3, the ideal motion performance is used as the desired constraint, and the performance index β used to evaluate the constraint following error is designed: The ideal motion performance is taken as the first-order constraint on the robot: make: in, Take the ideal group performance as the desired constraint: Taking its first-order derivative with respect to time t, the second-order constraint is: The constrained following error is defined as: in, It represents the degree of expected constraint following. For complete expected following, β i ≡0; Take the first derivative of β with respect to t: The β dynamics equation is described as:

6. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 5, Features: In step S3, for uncertainty processing, M i , C i and F i The decomposition is as follows: when Symmetrical and positive timing, ΔM i (·), ΔC i (·), and ΔF i (·) are all continuous functions, let: therefore, The decomposed nominal control items are expressed as: For uncertain system control, the β dynamics equation can be rewritten as:

7. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 6, Features: In step S3, an adaptive robust control strategy that approximates constraint following is designed: in, Among them, λ i >0 is a constant design parameter, function Parameters in The leaky adaptive law is as follows: Among them, the scalar is the design parameter, α i represents the comprehensive impact of uncertainty on system performance, Represents the adaptive parameter α i Adaptive robust action.

8. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 7, Features: In step S3, the control parameters in the adaptive robust constraint following controller are optimized and selected so that the multi-robot collaborative system is uniformly bounded and uniformly ultimately bounded.

9. The time-delay-resistant multi-mobile robot formation control method based on constraint following according to claim 1, Features: In step S4, the consistency analysis of the multi-mobile robot system with time delay is performed: The time-delay system is expressed as: After the neutral model transformation, the time-delay system is obtained: For any s∈C+, when When , the stability of system (50) is the same as that of system (49); For any initial condition, the state of each robot in the system satisfies the consistency constraint: When there is time delay in the mobile robot system, the robot consistency constraint in the system is expressed as: It can be further expressed as a vector:

10. The delay-resistant multi-mobile robot formation control method based on constraint following according to claim 9, Features: In step S4, under the network topology of directed spanning tree, the necessary and sufficient conditions and control algorithm for the delay-resistant multi-mobile robot system to reach consensus are proposed: In order to eliminate the constraint error in the control algorithm, The values ​​of k1 and k2 are appropriately selected so that the vector is globally asymptotically stable; Take k 1 =α,k 2 =β to obtain the second-order formal constraints of the mobile robot system: The control algorithm for achieving consistency of the multi-mobile robot system with time delay is obtained from the second-order constraints and the system dynamics equations as follows: Let N = M -2 get: The i-th component can be expressed as: Substitute the control algorithm into the system dynamics equation and express it in matrix form as:

Citation Information

Cited By

  • Mechanical arm predefined time trajectory tracking control method and system based on time delay estimation

    CN120985656A

  • Table tennis playing robot preset performance control method based on optimal human decision

    CN121635104A