Industrial robot control method, device, equipment and storage medium

Abstract

The application discloses an industrial robot control method, device and equipment and a storage medium, relates to the technical field of intelligent control, and comprises the following steps: acquiring physical characteristic data and a preset performance index of an industrial robot; constructing a constrained system model of the industrial robot according to the physical characteristic data and the preset performance index; performing equivalent system conversion on the constrained system model to obtain an unconstrained system model; constructing a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial robot; and solving the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method to obtain an optimal control law, so that the industrial robot is controlled according to the optimal control law. The application solves the problem that the existing industrial robot control method cannot realize optimal control while meeting the constraint of the preset performance index, and realizes high-precision control of the industrial robot.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent control, and particularly relates to an industrial robot arm control method, device, equipment and storage medium. BACKGROUND

[0002] With the continuous development of technology and the continuous improvement of production efficiency and operation precision, industrial robot arms have been widely used in the industries of machining, assembly and welding with high requirements for trajectory tracking precision. In order to improve the control precision of the industrial robot arm, it is usually necessary to propose requirements for performance indicators, so it is of great significance to combine the preset performance indicators in the controller design process. Optimal control is a control strategy considering system control performance and energy saving effect, and the motion process of the industrial robot arm is a highly coupled nonlinear system, which brings great challenges to the traditional optimal control method.

[0003] Therefore, it is an urgent problem to design an industrial robot arm controller that meets the preset performance index constraint and can realize optimal control. SUMMARY

[0004] The main purpose of the present application is to provide an industrial robot arm control method, device, equipment and storage medium, which aims to solve the technical problem that the existing industrial robot arm control method cannot realize optimal control while meeting the preset performance index constraint.

[0005] To achieve the above purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides an industrial robot arm control method, which comprises:

[0007] obtaining physical characteristic data of an industrial robot arm and a preset performance index;

[0008] constructing a constrained system model of the industrial robot arm according to the physical characteristic data and the preset performance index;

[0009] performing equivalent system conversion on the constrained system model to obtain an unconstrained system model;

[0010] constructing a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial robot arm;

[0011] solving the Hamilton-Jacobi-Bellman equation by an adaptive dynamic programming method to obtain an optimal control law, so as to control the industrial robot arm according to the optimal control law.

[0012] Optionally, in the industrial robot arm control method, the step of constructing the constrained system model of the industrial robot arm according to the physical characteristic data and the preset performance index comprises:

[0013] constructing a state space equation of the industrial robot arm according to the physical characteristic data;

[0014] defining a preset performance function according to the preset performance index;

[0015] obtaining the constrained system model according to the state space equation and the preset performance function.

[0016] Optionally, in the industrial robot arm control method, the step of constructing a state space equation of the industrial robot arm according to the physical characteristic data comprises:

[0017] modeling the physical characteristic data to obtain a dynamics model of the industrial robot arm;

[0018] transforming the dynamics model to obtain the state space equation of the industrial robot arm.

[0019] Optionally, in the industrial robot arm control method, the step of constructing the Hamilton-Jacobi-Bellman equation according to the unconstrained system model and the optimal control strategy of the industrial robot arm comprises:

[0020] defining a cost function according to the unconstrained system model and the optimal control strategy of the industrial robot arm;

[0021] constructing the Hamilton-Jacobi-Bellman equation according to the cost function.

[0022] Optionally, in the industrial robot arm control method, the step of defining a cost function according to the unconstrained system model and the optimal control strategy of the industrial robot arm comprises:

[0023] defining a position tracking error and a hybrid error according to the unconstrained system model;

[0024] obtaining an error vector according to the position tracking error and the hybrid error;

[0025] defining a cost function according to the error vector and the optimal control strategy of the industrial robot arm.

[0026] Optionally, in the industrial robot arm control method, the step of constructing the Hamilton-Jacobi-Bellman equation according to the cost function comprises:

[0027] defining a Hamilton function and an optimal cost function according to the cost function;

[0028] solving the optimal cost function by using Bellman's optimality principle to obtain an optimal solution of the optimal cost function;

[0029] substituting the optimal solution into the Hamilton function to obtain the Hamilton-Jacobi-Bellman equation.

[0030] Optionally, in the industrial robot control method described above, the step of solving the Hamilton-Jacobi-Bellman equation by using the adaptive dynamic programming method to obtain the optimal control law comprises:

[0031] The Hamilton-Jacobi-Bellman equation is solved by using an adaptive dynamic programming method based on a neural network architecture to obtain the optimal control law.

[0032] In a second aspect, the present application provides an industrial robot control device, which comprises:

[0033] a data acquisition module configured to acquire physical characteristic data of the industrial robot and a preset performance index;

[0034] a model construction module configured to construct a constrained system model of the industrial robot according to the physical characteristic data and the preset performance index;

[0035] a system conversion module configured to perform equivalent system conversion on the constrained system model to obtain an unconstrained system model;

[0036] an equation construction module configured to construct a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial robot;

[0037] an optimal control module configured to solve the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method to obtain an optimal control law, and to control the industrial robot according to the optimal control law.

[0038] In a third aspect, the present application provides an industrial robot control device, which comprises a processor and a memory, and the memory stores an industrial robot control program, and the industrial robot control program, when executed by the processor, implements the industrial robot control method described above.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium, and the storage medium stores a computer program, and the computer program, when executed by one or more processors, implements the industrial robot control method described above.

[0040] The above one or more technical solutions provided by the present application can have the following advantages or at least achieve the following technical effects:

[0041] The industrial robot control method, device, equipment and storage medium provided by the application, by constructing a constrained system model of the industrial robot according to the physical characteristic data of the industrial robot and the preset performance index, performing equivalent system conversion on the constrained system model to obtain an unconstrained system model, constructing a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and the optimal control strategy of the industrial robot, solving the Hamilton-Jacobi-Bellman equation by an adaptive dynamic programming method to obtain an optimal control law, and controlling the industrial robot according to the optimal control law, optimal control is performed on the basis of guaranteeing the preset performance index, and high-precision control of the industrial robot is realized; the method can make the system output of the industrial robot effectively track a reference signal, and make tracking errors and the like meet preset requirements, improve control precision, and has an energy-saving effect. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0043] Figure 1 The flowchart of the industrial robot control method of the present application;

[0044] Figure 2 The hardware structure diagram of the industrial robot control device related to the present application;

[0045] Figure 3 The physical characteristic diagram of the single-link robot arm in embodiment one of the present application;

[0046] Figure 4 The convergence curve diagram of the evaluation network weight in embodiment one of the present application;

[0047] Figure 5 The curve diagram of the reference signal y d and the system state x1 in embodiment one of the present application;

[0048] Figure 6 The curve diagram of the reference signal and the system state x2 in embodiment one of the present application;

[0049] Figure 7 The curve diagram of the tracking error and the preset performance boundary in embodiment one of the present application;

[0050] Figure 8 The function module diagram of the industrial robot control device of the present application.

[0051] The objectives, functional characteristics and advantages of the present application will be further described with reference to the embodiments in conjunction with the accompanying drawings. DETAILED DESCRIPTION

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0053] It should be noted that, in the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive containing, so that the process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or system. Without more limitations, the elements defined by the statement "comprising" do not exclude the presence of other identical elements in the process, method, article or system including the element. In addition, in the present application, the suffixes such as "module", "component" or "unit" used to represent elements are only for the convenience of description of the present application, and have no specific meaning. Therefore, "module", "component" or "unit" can be used mixedly. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, the technical solutions of various embodiments can be combined with each other, but it is based on the fact that the technical solutions can be realized by those of ordinary skill in the art. When the combination of technical solutions contradicts each other or cannot be realized, it should be considered that the combination of technical solutions does not exist, and is not within the protection scope of the present application.

[0054] It is found through analysis of the prior art that fast response and high-precision position tracking control has always been a research hotspot of industrial robot. At present, most of the control methods for industrial robots are still at the stage of ensuring the progressive convergence of tracking error, and there are problems such as slow reaction speed, excessive overshoot, etc. In order to further improve the control accuracy of industrial robots, it is usually necessary to require performance indicators such as convergence rate, maximum overshoot and steady-state error. Therefore, it is of great significance to combine preset performance indicators in the controller design process.

[0055] In addition, while focusing on high-precision control of industrial robot arms, the industry also puts forward higher requirements for the energy consumption of industrial robot arms. Reducing the control cost of industrial robot arms and reducing energy consumption is particularly important for current industrial development. Optimal control is a control strategy that considers system control performance and energy saving effects. Research shows that the movement process of an industrial robot arm is a highly coupled nonlinear system, which poses a huge challenge to traditional optimal control methods.

[0056] The adaptive dynamic programming method is a solution to the optimal control problem of a strongly nonlinear system. Based on the adaptive dynamic programming design technique, the optimal controller obtained can ensure the stability of the controlled system while achieving optimal system performance. However, at present, for the high-precision tracking control problem of industrial robot arms, designing an optimal controller that meets the preset performance index constraint is still a problem to be solved.

[0057] In view of the technical problem in the prior art that the control method of the industrial robot arm cannot achieve optimal control while meeting the preset performance index constraint, the present application provides an industrial robot arm control method, the general idea of which is as follows:

[0058] Physical property data and a preset performance index of an industrial robot arm are obtained, a constrained system model of the industrial robot arm is constructed according to the physical property data and the preset performance index, an equivalent system conversion is performed on the constrained system model to obtain an unconstrained system model, a Hamilton-Jacobi-Bellman equation is constructed according to the unconstrained system model and an optimal control strategy of the industrial robot arm, the Hamilton-Jacobi-Bellman equation is solved by an adaptive dynamic programming method to obtain an optimal control law, and the industrial robot arm is controlled according to the optimal control law.

[0059] Through the above technical solution, optimal control is performed on the basis of ensuring the preset performance index, high-precision control of the industrial robot arm is achieved, the method of the present application can make the system output of the industrial robot arm effectively track the reference signal and make the tracking error satisfy the preset requirements, thereby improving the control precision and also having an energy saving effect.

[0060] The industrial robot arm control method, device, equipment and storage medium provided by the present application will be described in detail below with reference to the accompanying drawings, specific embodiments and implementation modes.

[0061] Embodiment one

[0062] Referring to Figure 1 the flowchart, the first embodiment of the industrial robot arm control method of the present application is proposed, and the industrial robot arm control method is applied to an industrial robot arm control device.

[0063] The industrial robot arm control device refers to a terminal device or a control device capable of realizing data transmission, which can be a mobile phone, a computer, an embedded industrial computer, etc. terminal device, or a controller, a processor, etc. control device located in the system.

[0064] As shown in Figure 2 , it is a hardware structure diagram of the industrial robot arm control device. The industrial robot arm control device can include a processor 1001, such as a CPU (Central Processing Unit), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005.

[0065] Specifically, the communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 is used to connect the client and communicate data with the client, and the user interface 1003 can include an output unit such as a display screen and an input unit such as a keyboard; the network interface 1004 is used to connect the background server and communicate data with the background server, and the network interface 1004 can include an input / output interface, such as a standard wired interface, a wireless interface, such as a Wi-Fi interface; the memory 1005 is used to store various types of data, which can include, for example, instructions of any application program or method in the industrial robot arm control device, and application program related data, and the memory 1005 can be a high-speed RAM memory or a stable memory, such as a disk memory; optionally, the memory 1005 can also be a storage device independent of the processor 1001, and referring to Figure 2 , the memory 1005 can include an operating system, a network communication module, a user interface module, and an industrial robot arm control program;

[0066] The processor 1001 is used to call the industrial robot arm control program stored in the memory 1005 and perform the following operations:

[0067] Obtain the physical characteristic data of the industrial robot arm and the preset performance index;

[0068] According to the physical characteristic data and the preset performance index, a constrained system model of the industrial robot arm is constructed;

[0069] The constrained system model is converted into an equivalent system to obtain an unconstrained system model;

[0070] According to the unconstrained system model and the optimal control strategy of the industrial robot arm, a Hamilton-Jacobi-Bellman equation is constructed;

[0071] The Hamilton-Jacobi-Bellman equation is solved by an adaptive dynamic programming method to obtain an optimal control law, so as to control the industrial robot arm according to the optimal control law.

[0072] Based on the above industrial robot control device, the industrial robot control method of the embodiment is described in detail below in combination with the flowchart shown in the figure. The method can include the following steps: Figure 1 The method can include the following steps:

[0073] Step S100: Obtain physical characteristic data of the industrial robot and a preset performance index.

[0074] Specifically, the industrial robot can be a single-link robot arm, such as a type of nonlinear single-link robot arm. The physical characteristic data refers to the mechanical characteristics corresponding to the hardware structure of the industrial robot, such as the mass of the payload of the industrial robot, the length of the robot arm, the moment of inertia, the friction coefficient, the rotation angle, and the control input, etc. The preset performance index refers to the parameter index for achieving the preset performance control of the industrial robot, such as the tracking error requirement of the industrial robot. The tracking error is the difference between the output signal of the industrial robot and the input signal to be tracked, and the tracking error index can be a range value.

[0075] Step S200: Construct a constrained system model of the industrial robot according to the physical characteristic data and the preset performance index.

[0076] Specifically, after the control device obtains the physical characteristic data and the preset performance index of the industrial robot, it can construct a constrained system model of the industrial robot. The model can be a dynamics model or a state equation model.

[0077] Further, step S200 can include:

[0078] Step S210: Construct a state space equation of the industrial robot according to the physical characteristic data.

[0079] Specifically, when constructing the state space equation according to the physical characteristic data of the industrial robot, a dynamics model can be first established and then converted into the form of a state space equation, or the physical characteristic data can be directly input to obtain the corresponding state space equation through a set program or method, such as some existing modeling software.

[0080] Further, step S210 can include:

[0081] Step S211: Model the physical characteristic data to obtain a dynamics model of the industrial robot.

[0082] In this embodiment, a type of single-link robot arm is taken as an example, as shown in Figure 1A physical characteristic diagram of the single-link robot arm is shown, and the physical characteristics of the single-link robot arm are obtained, including the mass M of the payload of the single-link robot arm, the gravitational acceleration g, the length H of the single-link robot arm, the rotation angle θ(t), the moment of inertia G, and the control input u(t) of the control device to the single-link robot arm. The physical characteristics of the single-link robot arm are modeled, and the obtained dynamic model is:

[0083]

[0084] wherein M represents the mass of the payload of the single-link robot arm, g represents the gravitational acceleration, H represents the length of the single-link robot arm, G represents the moment of inertia, D represents the friction coefficient, θ(t) represents the rotation angle, u(t) represents the control input, and t represents time.

[0085] Step S212: converting the dynamic model to obtain a state space equation of the industrial robot arm.

[0086] In this embodiment, according to the physical characteristics of the single-link robot arm, the equation of the dynamic model obtained by modeling can be converted into a state space equation. Let x1=θ(t), The state space equation of the single-link robot arm is obtained as:

[0087]

[0088]

[0089] y=x1

[0090] wherein y represents an output signal of the industrial robot arm.

[0091] Suppose that the input signal to be tracked of the single-link robot arm is y d , the first-order derivative , and the second-order derivative all exist, then the compact structure of the state space equation of the single-link robot arm can be further obtained as:

[0092]

[0093] y=x1

[0094] wherein

[0095] Step S220: defining a preset performance function according to the preset performance index;

[0096] Preset performance control is a practical technology that can predetermine dynamic performance indicators such as convergence speed and control accuracy, enabling the tracking error of an industrial robotic arm to be maintained within a finite range composed of two specified performance functions, thus ensuring the high dynamic performance of the industrial robotic arm.

[0097] In this embodiment, the tracking error of the single-link robotic arm is the difference between the output signal y and the input signal y to be tracked d . Assuming that the tracking error requirement of the single-link robotic arm is between b and B, the preset performance index is b < y - y d < B. Then, the preset performance function determined according to the preset performance index of the single-link robotic arm is:

[0098]

[0099]

[0100] where δ1, l1, γ1, δ2, l2, and γ2 are all positive design parameters that can be set according to actual needs.

[0101] Step S230: Obtain the constrained system model according to the state space equation and the preset performance function.

[0102] In this embodiment, according to the preset performance function, the output constraint of the single-link robotic arm can be obtained as b + y d < y < B + y d . Combining with the aforementioned state space equation, the constrained system model can be obtained.

[0103] Step S300: Perform an equivalent system transformation on the constrained system model to obtain an unconstrained system model.

[0104] In this embodiment, to achieve the goal of high-precision control, a system transformation technology based on the preset performance function is introduced. For the output signal of the single-link robotic arm, a nonlinear mapping function is defined as:

[0105] ρ1 = T(x1, B, b, y d )

[0106] where a tanh(·) represents the arctangent function;

[0107] Taking the derivative of this nonlinear mapping function, we can obtain:

[0108]

[0109] Let Continuing, we get:

[0110]

[0111]

[0112] wherein,

[0113]

[0114] The preset performance constraint problem is processed by using a system transformation technique. By defining a nonlinear mapping function based on a preset performance index, a constrained system can be transformed into an equivalent unconstrained system. A control strategy designed for the transformed unconstrained system can make the tracking error of the industrial robot arm remain within a limited range composed of two preset performance functions, for example, within the ranges of b and B described above, thereby effectively improving the control precision.

[0115] Step S400: constructing a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and the optimal control strategy of the industrial robot arm.

[0116] Further, step S400 can include:

[0117] Step S410: defining a cost function according to the unconstrained system model and the optimal control strategy of the industrial robot arm.

[0118] Specifically, to balance the control precision and the energy consumed by the control input, thereby achieving the goal of optimal control, a cost function is defined.

[0119] Further, step S410 can include:

[0120] Step S411: defining a position tracking error and a hybrid error according to the unconstrained system model.

[0121] In this embodiment, for the transformed unconstrained system, the position tracking error z is defined as:

[0122] z = p1 - s d

[0123] wherein,

[0124] and the hybrid error θ is defined as:

[0125]

[0126] wherein, is a positive constant, which can be set according to actual needs, represents the rotational rate tracking error, is the first derivative of s. d

[0127] ​Step S412: obtaining an error vector according to the position tracking error and the mixing error.

[0128] In this embodiment, a new error vector X is defined based on the position tracking error z and the mixing error θ defined above:

[0129]

[0130] Thus, the dynamic equation of the error vector X can be obtained as:

[0131]

[0132] wherein,

[0133]

[0134] Step S413: defining a cost function according to the error vector and the optimal control strategy of the industrial robot arm.

[0135] In this embodiment, the cost function defined according to the error vector and the optimal control strategy of the industrial robot arm is:

[0136]

[0137] wherein, is a positive definite matrix, and R is a normal number which can be set according to actual needs, and R>0.

[0138] Step S420: constructing a Hamilton-Jacobi-Bellman equation according to the cost function.

[0139] Specifically, the Hamilton-Jacobi-Bellman equation is derived through the cost function defined based on the optimal control strategy, so that the optimal control problem can be converted into solving the Hamilton-Jacobi-Bellman equation.

[0140] Further, step S420 can include:

[0141] Step S421: defining a Hamilton function and an optimal cost function according to the cost function.

[0142] In this embodiment, the Hamilton function is defined according to the cost function defined in step S413 as:

[0143]

[0144] wherein,

[0145] During optimal control, the cost function needs to be minimized to obtain the desired control accuracy through the minimum control input. In this embodiment, the optimal cost function is defined as:

[0146]

[0147] wherein Ω represents a set of allowed control strategies for a single-link robot arm, V * (X) satisfies V * (0)=0.

[0148] Step S422: solving the optimal cost function by using the Bellman optimality principle to obtain an optimal solution of the optimal cost function.

[0149] According to the Bellman optimality principle, the optimal solution of the optimal cost function can be obtained as

[0150]

[0151] wherein,

[0152] The optimal solution of the optimal cost function can be obtained as

[0153]

[0154] Step S423: substituting the optimal solution into the Hamilton function to obtain the Hamilton-Jacobi-Bellman equation.

[0155] In this embodiment, the optimal solution u * (t) is substituted into the Hamilton function defined in step S421, and the Hamilton-Jacobi-Bellman equation can be obtained as

[0156]

[0157] Step S500: solving the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method to obtain an optimal control law, so as to control the industrial robot arm according to the optimal control law.

[0158] Specifically, due to the strong nonlinear property of the Hamilton-Jacobi-Bellman equation, it is difficult to directly solve the optimal cost function and the optimal control law. Therefore, the adaptive dynamic programming method of a single network can be used to approximately solve the Hamilton-Jacobi-Bellman equation.

[0159] Further, step S500 can include:

[0160] Step S510: solving the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method based on a neural network architecture to obtain an optimal control law.

[0161] In this embodiment, an evaluation network of a single network is established based on a neural network architecture, and the optimal cost function V * ​(X) can be evaluated as a network approximation:

[0162]

[0163]

[0164] wherein, is an ideal weight vector, is a basis function vector, is an approximation error, m represents a number of neural network nodes, and

[0165] Let represent an estimated value of the ideal weight, the optimal cost function can be estimated as:

[0166]

[0167]

[0168] The approximate optimal control law can be obtained as:

[0169]

[0170] Thus, the control device can correspondingly control the single-link robot arm according to the optimal control law, ensure optimal control on the basis of the preset performance index, and achieve high-precision control of the single-link robot arm.

[0171] The single-network adaptive dynamic programming method is used to approximately obtain the optimal control law, control the industrial robot arm, and the single evaluation network is used to approximately obtain the optimal cost function, which helps to reduce the calculation amount and memory requirement compared with the traditional execution-evaluation double-network structure.

[0172] Optionally, the weight update law of the designed evaluation network can be:

[0173]

[0174] wherein β is a positive design parameter, which can be set according to actual needs,

[0175] In order to verify the effectiveness of the industrial robot arm control method provided in the embodiment, the following simulation experiment is performed:

[0176] In the simulation experiment, the control target is set to make the output signal y of the single-link robot arm track the reference signal y d = 0.2sin(t) in an optimal manner. According to the actual system of the single-link robot arm, the physical characteristic data are respectively taken as M = 1 kg and g = 9.8 m / s 2, l = 1 m, D = 2N m s / rad. The preset performance function is b = -0.1 - e -t and B = 0.1 + e -t The initial value of the state of the single-link robot arm is x1(0) = 0.6, x2(0) = -1.5. The function of the evaluation network is σ c (X) = [z 2 , θ T , z θ] c The initial value of the weight of the evaluation network is w T (0) = [100, 250, 50] d In addition, other parameters are set as Q = [800, 0; 0, 800], R = 1, β = 2.

[0177] The result is analyzed in the way of taking the time derivative of the Lyapunov function The parameters are designed based on the actual situation in the simulation experiment, which will not be exemplified here. After the analysis, it can be obtained that According to the Lyapunov stability theorem, it can be known that the tracking error z and the mixed error θ and the weight estimation error of the evaluation network are consistent and ultimately bounded, that is, the output signal y of the single-link robot arm can track the reference signal y d , and the weight of the evaluation network can converge close to the ideal value, as shown in the convergence curve of the evaluation network weight in Figure 4 The horizontal axis represents time in seconds (s), and the vertical axis represents the value of the evaluation network weight. It can be seen from the graph that the evaluation network can accurately approximate the cost function, so that the control input u(t) obtained can be considered optimal.

[0178] As shown in Figure 5 , the curve graph of the reference signal y d and the system state x1 in this embodiment is shown. The horizontal axis represents time in seconds (s), and the vertical axis represents the value of each curve. As shown in Figure 6 , the curve graph of the reference signal and the system state x2 in this embodiment is shown. The horizontal axis represents time in seconds (s), and the vertical axis represents the value of each curve. As shown in Figure 7 , the curve graph of the tracking error and the preset performance boundary in this embodiment is shown. The horizontal axis represents time in seconds (s), and the vertical axis represents the value of each curve, wherein the preset performance boundary is the curve corresponding to the value of the preset performance index. It can be seen from Figures 5-7 that the system state of the single-link robot arm corresponding to the output signal y is close to the reference signal y The control method keeps consistent, the tracking effect is good, and the tracking error meets the requirement of the preset performance index, so that high-precision control of the single-link mechanical arm can be realized.

[0179] The industrial mechanical arm control method provided by the embodiment constructs a constrained system model of the industrial mechanical arm according to physical characteristic data of the industrial mechanical arm and a preset performance index, performs equivalent system conversion on the constrained system model to obtain an unconstrained system model, constructs a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial mechanical arm, solves the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method to obtain an optimal control law, and controls the industrial mechanical arm according to the optimal control law, so that optimal control is performed on the basis of guaranteeing the preset performance index, high-precision control of the industrial mechanical arm is realized, the system output of the industrial mechanical arm can effectively track a reference signal, and the tracking error and the like meet preset requirements, the control precision is improved, and the energy-saving effect is achieved.

[0180] Embodiment Two

[0181] Based on the same inventive concept, referring to Figure 8 , the first embodiment of the industrial mechanical arm control device of the present application is proposed, which can be a virtual device and is applied to an industrial mechanical arm control equipment.

[0182] The industrial mechanical arm control device provided by the embodiment will be described in detail below with reference to the functional module schematic diagram shown in Figure 8 The device can include:

[0183] A data acquisition module is configured to acquire physical characteristic data of the industrial mechanical arm and a preset performance index.

[0184] A model construction module is configured to construct a constrained system model of the industrial mechanical arm according to the physical characteristic data and the preset performance index.

[0185] A system conversion module is configured to perform equivalent system conversion on the constrained system model to obtain an unconstrained system model.

[0186] An equation construction module is configured to construct a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial mechanical arm.

[0187] An optimal control module is configured to solve the Hamilton-Jacobi-Bellman equation by using an adaptive dynamic programming method to obtain an optimal control law, and control the industrial mechanical arm according to the optimal control law.

[0188] It should be noted that the functions and technical effects of the modules in the industrial robot arm control device provided in the embodiments can refer to the descriptions of the specific embodiments of the industrial robot arm control method of the present application. For the sake of brevity of the description, the details are not repeated here.

[0189] Embodiment three

[0190] Based on the same inventive concept, referring to the hardware structure diagram of Figure 2 , the present embodiment provides an industrial robot arm control device, which can include a processor and a memory, and the memory stores an industrial robot arm control program. When the processor executes the industrial robot arm control program, all or part of the steps of the embodiments of the industrial robot arm control method of the present application are implemented.

[0191] Specifically, the industrial robot arm control device refers to a terminal device or a control device that can realize data transmission. It can be a terminal device such as a mobile phone, a computer, an embedded industrial computer, etc., or a control device such as a controller, a processor, etc. located in the system.

[0192] Those skilled in the art can understand that Figure 2 the hardware structure shown in the above does not constitute a limitation on the industrial robot arm control device of the present application, and can include more or fewer components than those shown, or combine certain components, or different component arrangements.

[0193] It can be understood that the industrial robot arm control device can also include a communication bus, a user interface and a network interface. Among them, the communication bus is used to realize the connection and communication between these components; the user interface is used to connect the client and communicate data with the client. The user interface can include an output unit such as a display screen and an input unit such as a keyboard; the network interface is used to connect the background server and communicate data with the background server. The network interface can include an input / output interface, such as a standard wired interface, a wireless interface.

[0194] The memory is used to store various types of data, which can include, for example, instructions of any application program or method in the industrial robot control device, and application program related data. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Random Access Memory (RAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic or optical disk, and optionally, the memory can also be a storage device independent of the processor.

[0195] The processor is used to call the industrial robot control program stored in the memory, and execute the industrial robot control method as described above. The processor can be an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic elements, which is used to execute all or part of the steps of each embodiment of the industrial robot control method as described above.

[0196] Embodiment Four

[0197] Based on the same inventive concept, the embodiment provides a computer readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card memory (for example, an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read only memory (ROM), a programmable read only memory (PROM), an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a magnetic memory, a magnetic disk, an optical disk, a server, and the like, which stores a computer program. The computer program can be executed by one or more processors. When the computer program is executed by the processor, all or part of the steps of the industrial robot arm control method embodiments of the present application can be implemented.

[0198] It should be noted that the above-mentioned embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments. The above embodiments are only optional embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent flow conversion within the inventive concept of the present application, or direct or indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. An industrial robot control method, characterized by, The method comprises: acquiring physical characteristic data of an industrial robot arm and a preset performance index; constructing a constrained system model of the industrial robot arm according to the physical characteristic data and the preset performance index; performing equivalent system conversion on the constrained system model to obtain an unconstrained system model; constructing a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial robot arm; solving the Hamilton-Jacobi-Bellman equation by an adaptive dynamic programming method to obtain an optimal control law, and controlling the industrial robot arm according to the optimal control law; the step of constructing the constrained system model of the industrial robot arm according to the physical characteristic data and the preset performance index comprises: constructing a state space equation of the industrial robot arm according to the physical characteristic data; defining a preset performance function according to the preset performance index; obtaining the constrained system model according to the state space equation and the preset performance function.

2. The industrial robot control method of claim 1, wherein, the step of constructing the state space equation of the industrial robot arm according to the physical characteristic data comprises: modeling the physical characteristic data to obtain a dynamics model of the industrial robot arm; performing conversion on the dynamics model to obtain the state space equation of the industrial robot arm.

3. The industrial robot control method of claim 1, wherein, the step of constructing the Hamilton-Jacobi-Bellman equation according to the unconstrained system model and the optimal control strategy of the industrial robot arm comprises: defining a cost function according to the unconstrained system model and the optimal control strategy of the industrial robot arm; constructing the Hamilton-Jacobi-Bellman equation according to the cost function.

4. The industrial robot control method according to claim 3, wherein the step of defining the cost function according to the unconstrained system model and the optimal control strategy of the industrial robot arm comprises: defining a position tracking error and a hybrid error according to the unconstrained system model; obtaining an error vector according to the position tracking error and the hybrid error; defining the cost function according to the error vector and the optimal control strategy of the industrial robot arm.

5. The industrial robot control method according to claim 3, wherein, the step of constructing the Hamilton-Jacobi-Bellman equation according to the cost function comprises: defining a Hamilton function and an optimal cost function according to the cost function; solving the optimal cost function by a Bellman optimality principle to obtain an optimal solution of the optimal cost function; substituting the optimal solution into the Hamilton function to obtain the Hamilton-Jacobi-Bellman equation.

6. The industrial robot control method of claim 1, wherein, the step of solving the Hamilton-Jacobi-Bellman equation by the adaptive dynamic programming method to obtain the optimal control law comprises: solving the Hamilton-Jacobi-Bellman equation by an adaptive dynamic programming method based on a neural network architecture to obtain the optimal control law.

7. An industrial robot control device, characterized by The device comprises: a data acquisition module for acquiring physical characteristic data of an industrial robot arm and a preset performance index; a model construction module for constructing a constrained system model of the industrial robot arm according to the physical characteristic data and the preset performance index; a system conversion module for performing equivalent system conversion on the constrained system model to obtain an unconstrained system model; An equation construction module is configured to construct a Hamilton-Jacobi-Bellman equation according to the unconstrained system model and an optimal control strategy of the industrial robot arm; An optimal control module is configured to solve the Hamilton-Jacobi-Bellman equation by an adaptive dynamic programming method to obtain an optimal control law, and control the industrial robot arm according to the optimal control law. The model construction module is further configured to construct a state space equation of the industrial robot arm according to the physical characteristic data, define a preset performance function according to the preset performance index, and obtain the constrained system model according to the state space equation and the preset performance function.

8. An industrial robot control device, characterized by The device comprises a processor and a memory, and the memory stores an industrial robot arm control program. When the industrial robot arm control program is executed by the processor, the industrial robot arm control method in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program. When the computer program is executed by one or more processors, the industrial robot arm control method in any one of claims 1 to 6 is implemented.

Images