A method for pre-determined active disturbance suppression for a non-holonomic mobile robot
By designing a predetermined time active interference suppression method in an incomplete mobile robot, a yaw rate disturbance and velocity observer is constructed, which solves the problems of slow convergence time and high uncertainty in the prior art, and improves the trajectory tracking performance and observer The accuracy of the problem enables more flexible uncertainty estimation.
Patent Information
- Application Number
- CN202210726847.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-24
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-06-24
AI Technical Summary
In the prior art, when designing uncertain estimation methods for incomplete mobile robots, the convergence time is not fast enough and the uncertainty is high, making it difficult to meet the trajectory tracking requirements in complex environments.
By adopting the predetermined time active interference suppression method, a predetermined time observer of yaw rate disturbance, lateral linear velocity and longitudinal sliding speed is constructed, combined with sliding mode control and predetermined time control theory, a rapid convergence and adjustable uncertain estimation method is designed. .
The convergence speed and upper limit estimation accuracy of the observer are improved, and the singularity problem of the observer in sliding mode design is solved, making the sliding mode surface function simpler. Users can adjust the robot's uncertainty of the observer's estimation time according to their needs. Improve observation flexibility.
Smart Images

Figure CN115421480B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of automatic control, and in particular relates to a method for actively suppressing interference at a predetermined time of a non-holonomic mobile robot. Background Art
[0002] In recent years, the path tracking problem of nonholonomic mobile robots has attracted increasing attention due to its wide application in engineering fields such as space exploration, hospital missions, search and rescue, etc. In order to enable nonholonomic mobile robots to complete various complex tasks, appropriate control laws, such as finite-time / fixed-time sliding mode control, adaptive control, neural network control, etc., need to be designed to meet their specific needs. In addition, nonholonomic mobile robots usually need to work in various complex environments, such as curved roads, slippery roads, forest paths, etc. Therefore, designing control laws for nonholonomic mobile robots with interference suppression capabilities for uncertainty is of great significance for improving tracking performance. In order to solve the unknown slip problem of nonholonomic mobile robots, a simple way to deal with interference and uncertainty is to compensate for them by designing a controller with robust performance. This idea directly promoted the vigorous development of observer design, including the design of state observers, adaptive observers, disturbance observers, finite-time observers, etc. However, the initial conditions of the finite-time observer affect the convergence time, and as the initial error increases, the initial conditions become larger. This limits its application in practice in some cases. Fixed-time stability can solve the finite-time stability problem. Therefore, a special fixed-time stability theory has become a research hotspot for scientists, namely, predefined-time stability. Although the predefined-time stability theory has been applied to many systems, it has not yet been applied to the uncertainty estimation problem of non-holonomic mobile robots. In summary, it is very meaningful to study an uncertainty estimation method with fast convergence, adjustable convergence time and more accurate upper limit estimation of convergence time, taking the uncertainty in non-holonomic mobile robots as the estimation and observation object. Summary of the invention
[0003] The object of the present invention is to provide a method for actively suppressing interference at a predetermined time for a non-holonomic mobile robot, which is conducive to adjusting the convergence time and improving the convergence speed of the observer and the estimation accuracy of the upper limit of the convergence time.
[0004] To achieve the above object, the technical solution adopted by the present invention is: a method for actively suppressing interference at a predetermined time for a non-holonomic mobile robot, comprising the following steps:
[0005] Step S1: Establishing a kinematic model of an uncertain nonholonomic mobile robot and a kinematic model of a desired trajectory;
[0006] Step S2: Based on the model established in step S1, establish the posture error equation of the uncertain nonholonomic mobile robot;
[0007] Step S3: Based on the posture error equation established in step S2, a yaw rate disturbance predetermined time observer is constructed in combination with sliding mode control theory and predetermined time control theory;
[0008] Step S4: Based on the yaw rate disturbance predetermined time observer constructed in step S3, in combination with sliding mode control theory and predetermined time control theory, a corresponding lateral linear velocity predetermined time observer and a longitudinal sliding velocity predetermined time observer are constructed.
[0009] Furthermore, the step S1 is specifically as follows:
[0010] For a non-holonomic mobile robot consisting of two driving wheels and two universal wheels, the motion control of the entire mobile robot is achieved by controlling the two driving wheels; the mobile robot not only rolls during movement, but may also shift sideways; the kinematic model of the mobile robot is expressed as:
[0011]
[0012] Among them, the position and posture of the mobile robot is p = [xy θ] T ; Control input is [vw] T , where v represents the linear velocity of the mobile robot, w represents the angular velocity of the mobile robot; μ is the lateral linear velocity; η w is the yaw rate disturbance; η v is the longitudinal sliding speed;
[0013] The kinematic model of the desired trajectory of the mobile robot is expressed as:
[0014]
[0015] Among them, p d =[x d y d θ d ] T is the desired position of the mobile robot; the control input is [v d w d ] T , where v d represents the desired linear velocity, w d represents the desired angular velocity.
[0016] Furthermore, the step S2 is specifically as follows:
[0017] The pose error vector of the mobile robot is expressed as p e =[x e ye θ e ] T , when the posture error converges to zero, the trajectory tracking control of the mobile robot is realized; d =[x d y d θ d ] T is the expected position of the mobile robot in the local coordinate system. According to the coordinate change formula, we can get:
[0018]
[0019] because
[0020] x e = cosθ(x d -x)+sinθ(y d -y) (4)
[0021] x e Taking the derivative, we get:
[0022]
[0023] because
[0024] y e = -sinθ(x d -x)+cosθ(y d -y),
[0025] For e Taking the derivative, we get:
[0026]
[0027] because
[0028] θ e =θ d -θ
[0029] For θ e Taking the derivative, we get:
[0030]
[0031] Therefore, p e The derivative of is:
[0032]
[0033] Thus, the posture error equation of the mobile robot is obtained.
[0034] Furthermore, the step S3 is specifically as follows:
[0035] Based on the cascade control theory, the yaw rate disturbance η is constructed by the backward back-pushing design method. w The scheduled time observer for is:
[0036]
[0037] where k 01 and k 02 is the designed gain; T c01 are the preset parameters, It is obtained by the following formula:
[0038]
[0039] x 32 ∈R 3 is the auxiliary design variable of the yaw rate disturbance predetermined time observer, which is designed as:
[0040]
[0041] where x 12 is the auxiliary design variable of the fixed-time observer, is θ e The estimated value of η w Under the action of the predetermined time observer, accurate estimation is achieved, that is, The convergence time is T c01 ;
[0042] For x 32 ,get:
[0043]
[0044] Choose the Lyapunov function as:
[0045] V 01 =|x 32 |
[0046] Taking the derivative of the above formula, we get:
[0047]
[0048] Satisfying the scheduled time theorem, we have get:
[0049]
[0050] and
[0051]
[0052] Satisfying the fixed time theorem, we have When t→T 01 ,get and x 12 →0, where T 01 =T c01 +T fix ;
[0053] The result obtained by the yaw rate disturbance predetermined time observer is as follows: 01 ,get as well as At this point, we get:
[0054]
[0055]
[0056] Furthermore, the step S4 specifically includes the following steps:
[0057] Step S41: constructing a transverse linear speed predetermined time observer as:
[0058]
[0059] Where T c02 is the preset parameter, x 33 ∈R 3 is the auxiliary design variable of the transverse linear velocity predetermined time observer, which is designed as:
[0060]
[0061] where x 13 is the auxiliary design variable of the fixed-time observer, for y e The estimate of μ is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 ;
[0062] For x 33 ,get:
[0063]
[0064] Choose the Lyapunov function as:
[0065] V 02 =|x 33 |
[0066] Taking the derivative of the above formula, we get:
[0067]
[0068] Satisfying the scheduled time theorem, we have get:
[0069]
[0070] and
[0071]
[0072] Satisfying the fixed time theorem, we have When t→T 02 ,get and x 13 →0, where:
[0073] T 02 =T 01 +T c02 +T fix =T c01 +T c02 +2T fix
[0074] Based on the transverse linear velocity predetermined time observer, when t→T 02 ,get as well as
[0075] Step S42: constructing a longitudinal sliding speed predetermined time observer as:
[0076]
[0077] Where T c02 is the preset parameter, x 31 ∈R 3 is the auxiliary design variable of the longitudinal sliding velocity predetermined time observer, which is designed as:
[0078]
[0079] where x 11 is the auxiliary design variable of the fixed-time observer, For x e The estimated value of η v The longitudinal sliding velocity is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 ;
[0080] For x 31 ,get:
[0081]
[0082] Choose the Lyapunov function as:
[0083] V 03 =|x 31 |
[0084] Taking the derivative of the above formula, we get:
[0085]
[0086] Satisfying the scheduled time theorem, we have get:
[0087]
[0088] and
[0089]
[0090] Satisfying the fixed time theorem, we have When t→T 02 ,get and x 11 →0, based on the longitudinal sliding velocity predetermined time observer, when t→T 02 ,get as well as
[0091] Compared with the prior art, the present invention has the following beneficial effects: the present method improves the convergence speed of the observer, improves the upper limit estimation accuracy of the convergence time of the observer, and solves the singularity problem of the observer when performing sliding mode design, making the constructed sliding surface function simpler. In addition, the present method adopts a backward back-pushing design method, first designs a predetermined time observer of the yaw rate disturbance, and on this basis, designs a predetermined time observer of the lateral linear velocity and a predetermined time observer of the longitudinal sliding velocity at the same time, so that the user can adjust the uncertain observer estimation time of the robot according to the needs of the actual environment, and the observation flexibility is better. BRIEF DESCRIPTION OF THE DRAWINGS
[0092] Figure 1 A block diagram of a method implementation of an embodiment of the present invention;
[0093] Figure 2 Schematic diagram of the position and posture error of a non-holonomic mobile robot containing uncertainty in an embodiment of the present invention;
[0094] Figure 3 In the embodiment of the present invention, when T 01 =1 hour Tracking η w The evolution curve of
[0095] Figure 4 In the embodiment of the present invention, when T 02 =1.5 and T 02 =2 o'clock Track the evolution curve of μ;
[0096] Figure 5 In the embodiment of the present invention, when T 02=1.5 and T 02 =2 o'clock Tracking η v The evolution curve of
[0097] Figure 6 In the embodiment of the present invention, when T 01 =1, T 02 =1.5 and T 02 =2. Evolution curves of the tracking errors of the three observers. DETAILED DESCRIPTION
[0098] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0099] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.
[0100] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.
[0101] like Figure 1 As shown, this embodiment provides a method for actively suppressing interference at a predetermined time for a non-holonomic mobile robot, comprising the following steps:
[0102] Step S1: Establish a kinematic model of the uncertain nonholonomic mobile robot and a kinematic model of the expected trajectory.
[0103] In this embodiment, the step S1 is specifically as follows:
[0104] For a non-holonomic mobile robot consisting of two drive wheels and two universal wheels, the motion control of the entire mobile robot is achieved by controlling the two drive wheels. The mobile robot not only rolls during movement, but may also shift sideways. The kinematic model of the mobile robot is expressed as:
[0105]
[0106] Among them, the position and posture of the mobile robot is p = [xy θ] T ; Control input is [vw] T , where v represents the linear velocity of the mobile robot, w represents the angular velocity of the mobile robot; μ is the lateral linear velocity; ηw is the yaw rate disturbance; η v is the longitudinal sliding speed.
[0107] The kinematic model of the desired trajectory of the mobile robot is expressed as:
[0108]
[0109] Among them, p d =[x d y d θ d ] T is the desired position of the mobile robot; the control input is [v d w d ] T , where v d represents the desired linear velocity, w d represents the desired angular velocity.
[0110] Step S2: Based on the model established in step S1, establish the uncertainty non-holonomic mobile robot posture error equation.
[0111] In this embodiment, the posture error of the mobile robot is as follows: Figure 2 The step S2 is specifically as follows:
[0112] The pose error vector of the mobile robot is expressed as p e =[x e y e θ e ] T , when the posture error converges to zero, the trajectory tracking control of the mobile robot is realized. d =[x d y d θ d ] T is the expected position of the mobile robot in the local coordinate system. According to the coordinate change formula, we can get:
[0113]
[0114] because
[0115] x e = cosθ(x d -x)+sinθ(y d -y) (4)
[0116] x e Taking the derivative, we get:
[0117]
[0118] because
[0119] y e = -sinθ(x d -x)+cosθ(y d -y),
[0120] For e Taking the derivative, we get:
[0121]
[0122] because
[0123] θ e =θ d -θ
[0124] For θ e Taking the derivative, we get:
[0125]
[0126] Therefore, p e The derivative of is:
[0127]
[0128] Thus, the posture error equation of the mobile robot is obtained.
[0129] Step S3: Based on the posture error equation established in step S2, a yaw rate disturbance predetermined time observer is constructed in combination with sliding mode control theory and predetermined time control theory.
[0130] In this embodiment, step S3 is specifically as follows:
[0131] Based on the cascade control theory, the yaw rate disturbance η is constructed by the backward back-pushing design method. w The scheduled time observer for is:
[0132]
[0133] where k 01 and k 02 is the designed gain; T c01 are the preset parameters, It is obtained by the following formula:
[0134]
[0135] x 32 ∈R 3 is the auxiliary design variable of the yaw rate disturbance predetermined time observer, which is designed as:
[0136]
[0137] where x 12 is the auxiliary design variable of the fixed-time observer, is θ e The estimated value of η w Under the action of the predetermined time observer, accurate estimation is achieved, that is, The convergence time is T c01 .
[0138] For x 32 ,get:
[0139]
[0140] Choose the Lyapunov function as:
[0141] V 01 =|x 32 |
[0142] Taking the derivative of the above formula, we get:
[0143]
[0144] Satisfying the scheduled time theorem, we have get:
[0145]
[0146] and
[0147]
[0148] Satisfying the fixed time theorem, we have When t→T 01 ,get and x 12 →0, where T 01 =T c01 +T fix .
[0149] The result obtained by the yaw rate disturbance predetermined time observer is as follows: 01 ,get as well as At this point, we get:
[0150]
[0151]
[0152] Step S4: Based on the yaw rate disturbance predetermined time observer constructed in step S3, in combination with sliding mode control theory and predetermined time control theory, a corresponding lateral linear velocity predetermined time observer and a longitudinal sliding velocity predetermined time observer are constructed.
[0153] In this embodiment, step S4 specifically includes the following steps:
[0154] Step S41: constructing a transverse linear speed predetermined time observer as:
[0155]
[0156] Where T c02 is the preset parameter, x 33 ∈R 3 is the auxiliary design variable of the transverse linear velocity predetermined time observer, which is designed as:
[0157]
[0158] where x 13 is the auxiliary design variable of the fixed-time observer, for y e The estimate of μ is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 .
[0159] For x 33 ,get:
[0160]
[0161] Choose the Lyapunov function as:
[0162] V 02 =|x 33 |
[0163] Taking the derivative of the above formula, we get:
[0164]
[0165] Satisfying the scheduled time theorem, we have get:
[0166]
[0167] and
[0168]
[0169] Satisfying the fixed time theorem, we have When t→T 02 ,get and x 13 →0, where:
[0170] T 02 =T 01 +T c02 +T fix =Tc01 +T c02 +2T fix
[0171] Based on the transverse linear velocity predetermined time observer, when t→T 02 ,get as well as
[0172] Step S42: constructing a longitudinal sliding speed predetermined time observer as:
[0173]
[0174] Where T c02 is the preset parameter, x 31 ∈R 3 is the auxiliary design variable of the longitudinal sliding velocity predetermined time observer, which is designed as:
[0175]
[0176] where x 11 is the auxiliary design variable of the fixed-time observer, For x e The estimated value of η v The longitudinal sliding velocity is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 .
[0177] For x 31 ,get:
[0178]
[0179] Choose the Lyapunov function as:
[0180] V 03 =|x 31 |
[0181] Taking the derivative of the above formula, we get:
[0182]
[0183] Satisfying the scheduled time theorem, we have get:
[0184]
[0185] and
[0186]
[0187] Satisfying the fixed time theorem, we have When t→T 02 ,get and x 11 →0, based on the longitudinal sliding velocity predetermined time observer, when t→T 02 ,get as well as
[0188] In order to more intuitively demonstrate the effectiveness and feasibility of this method, this embodiment uses a set of MATLAB / Simulink simulations to verify this method.
[0189] Using this method and based on MATLAB / Simulink, a sliding mode control simulation of an uncertain nonholonomic mobile robot is performed, and the initial posture error is set to [x e (0),y e (0),θ e (0)] T =[2,2,π / 4] T , the ideal angular velocity is w d =1.0rad / s, the ideal linear velocity is v d =5.0m / s, tracking position is p d =[x d ,y d ,θ d ] T , then the ideal tracking curve is:
[0190] x d (t) = rcosw d t
[0191] y d (t) = rsinw d t
[0192] θ d (t) = w d t
[0193] The yaw rate disturbance η is calculated by using the yaw rate disturbance time observer. w Real-time tracking observation, the observer parameters are selected as: α=5, β=2, η=1, γ=1,∈ 01 =∈ 11 =0.4. Similarly, the lateral linear velocity preset time observer is used to track and observe the lateral linear velocity μ in real time, and the longitudinal sliding velocity preset time observer is used to track and observe the longitudinal sliding velocity η. v Real-time tracking observation, the observer parameters are selected as: ∈ 02 =∈ 12 =0.7,∈ 03 =∈ 13 =2. The uncertain interference is set to: η w =0.7sin(t),ηv =0.4cos(10t),μ=2. The simulation example results are as follows Figure 3-Figure 6 As shown, Figure 3 is the tracking result of the yaw rate disturbance preset time observer, Figure 4 The tracking result of the predetermined time observer for the lateral linear velocity, Figure 5 Tracking result of the time observer for the longitudinal sliding velocity, Figure 6 It is obvious that the yaw rate disturbance time observer, the lateral linear velocity time observer, and the longitudinal sliding velocity time observer all complete the tracking error curve of the uncertain yaw rate disturbance η within their respective expected time. w , transverse linear velocity μ, longitudinal sliding velocity η v The above specific example results are in line with the research expectations of this method.
[0194] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.
Claims
1. A method for actively suppressing interference at a predetermined time for a non-holonomic mobile robot, characterized in that: The following steps are involved: Step S1: Establishing a kinematic model of an uncertain nonholonomic mobile robot and a kinematic model of a desired trajectory; Step S2: Based on the model established in step S1, establish the posture error equation of the uncertain nonholonomic mobile robot; Step S3: Based on the posture error equation established in step S2, a yaw rate disturbance predetermined time observer is constructed in combination with sliding mode control theory and predetermined time control theory; Step S4: Based on the yaw rate disturbance predetermined time observer constructed in step S3, a corresponding lateral linear velocity predetermined time observer and a longitudinal sliding velocity predetermined time observer are constructed in combination with sliding mode control theory and predetermined time control theory; The step S1 is specifically as follows: For a non-holonomic mobile robot consisting of two driving wheels and two universal wheels, the motion control of the entire mobile robot is achieved by controlling the two driving wheels; the mobile robot not only rolls during movement, but may also shift sideways; the kinematic model of the mobile robot is expressed as: Among them, the position and posture of the mobile robot is p = [xy θ] T ; Control input is [vw] T , where v represents the linear velocity of the mobile robot, w represents the angular velocity of the mobile robot; μ is the lateral linear velocity; η w is the yaw rate disturbance; η v is the longitudinal sliding speed; The kinematic model of the desired trajectory of the mobile robot is expressed as: Among them, p d =[x d y d θ d ] T is the desired position of the mobile robot; the control input is [v d w d ] T , where v d represents the desired linear velocity, w d represents the desired angular velocity; The step S2 is specifically as follows: The pose error vector of the mobile robot is expressed as p e =[x e y e θ e ] T , when the posture error converges to zero, the trajectory tracking control of the mobile robot is realized; d =[x d y d θ d ] T is the expected position of the mobile robot in the local coordinate system. According to the coordinate change formula, we can get: because x e =cosθ(x d -x)+sinθ(y d -y) (4) x e Taking the derivative, we get: because and e =-sinθ(x d -x)+cosθ(y d -and), For e Taking the derivative, we get: because i e =θ d -θ For θ e Taking the derivative, we get: Therefore, p e The derivative of is: Thus, the posture error equation of the mobile robot is obtained; The step S3 is specifically as follows: Based on the cascade control theory, the yaw rate disturbance η is constructed by the backward back-pushing design method. w The scheduled time observer for is: where k 01 and k 02 is the designed gain; T c01 are the preset parameters, It is obtained by the following formula: x 32 ∈R 3 is the auxiliary design variable of the yaw rate disturbance predetermined time observer, which is designed as: where x 12 is the auxiliary design variable of the fixed-time observer, is θ e The estimated value of η w Under the action of the predetermined time observer, accurate estimation is achieved, that is, The convergence time is T c01 ; For x 32 ,get: Choose the Lyapunov function as: V 01 =|x 32 | Taking the derivative of the above formula, we get: Satisfying the scheduled time theorem, we have get: and Satisfying the fixed time theorem, we have When t→T 01 ,get and x 12 →0, where T 01 =T c01 +T fix ; The result obtained by the yaw rate disturbance predetermined time observer is as follows: 01 ,get as well as At this point, we get: The step S4 specifically comprises the following steps: Step S41: constructing a transverse linear speed predetermined time observer as: Where T c02 is the preset parameter, x 33 ∈R 3 is the auxiliary design variable of the transverse linear velocity predetermined time observer, which is designed as: where x 13 is the auxiliary design variable of the fixed-time observer, for y e The estimate of μ is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 ; For x 33 ,get: Choose the Lyapunov function as: V 02 =|x 33 | Taking the derivative of the above formula, we get: Satisfying the scheduled time theorem, we have get: and Satisfying the fixed time theorem, we have When t→T 02 ,get and x 13 →0, where: T 02 =T 01 +T c02 +T fix =T c01 +T c02 +2T fix Based on the transverse linear velocity predetermined time observer, when t→T 02 ,get as well as Step S42: constructing a longitudinal sliding speed predetermined time observer as: Where T c02 is the preset parameter, x 31 ∈R 3 is the auxiliary design variable of the longitudinal sliding velocity predetermined time observer, which is designed as: where x 11 is the auxiliary design variable of the fixed-time observer, For x e The estimated value of η v The longitudinal sliding velocity is accurately estimated under the action of the predetermined time observer, and the convergence time is T c02 ; For x 31 ,get: Choose the Lyapunov function as: V 03 =|x 31 | Taking the derivative of the above formula, we get: Satisfying the scheduled time theorem, we have get: and Satisfying the fixed time theorem, we have When t→T 02 ,get and x 11 →0, based on the longitudinal sliding velocity predetermined time observer, when t→T 02 ,get as well as
