A method for optimal control of grasping force of multi-fingered dexterous hand based on quadratic programming

Abstract

The application provides a multi-fingered dexterous hand optimal grasping force control method based on a quadratic programming, comprising the following steps: 1) obtaining contact point coordinate information and friction coefficient information of a multi-fingered dexterous hand and a target object; 2) obtaining a grasping force conversion matrix from an object coordinate system to a world coordinate system according to the contact point coordinate information; 3) designing a quadratic programming scheme of the multi-fingered dexterous hand optimal grasping force control; 4) converting the quadratic programming scheme into the form of a nonlinear equation set; 5) using the proposed optimal grasping force solver based on neural dynamics to solve the problem; and 6) controlling the multi-fingered dexterous hand to perform a grasping task. The application provides a reliable control scheme for a grasping task with complex grasping scenes and various grasping targets, controls the multi-fingered dexterous hand to grasp the object with the minimum force while eliminating the influence of external disturbances, and avoids the conditions of damaging the object or unstable grasping.

Description

Technical Field

[0001] This invention relates to the field of optimal control of grasping force in multi-finger dexterous hands, and specifically to an optimal control method for grasping force in multi-finger dexterous hands based on quadratic programming. Background Technology

[0002] Multi-fingered dexterous hands are an important research direction in the field of robotics. Compared to traditional parallel grippers, multi-fingered dexterous hands have higher degrees of freedom, giving them more diverse grasping methods. At the same time, multi-fingered dexterous hands can precisely control the force applied to the target object. This advantage enables them to perform better in grasping tasks with increasingly diverse grasping targets and scenarios.

[0003] Optimizing gripping force is one of the key issues in the optimized control of multi-finger dexterity hands. When the target object is fragile or prone to slipping, the gripping task requires the multi-finger dexterity hand to apply appropriate force to the object to avoid damage or insecure gripping. Therefore, during the gripping task, the force applied by the multi-finger dexterity hand to the target object should be as small as possible while preventing relative slippage. Furthermore, multi-finger dexterity hands are used in various gripping scenarios due to their diverse gripping methods. It is worth noting that most existing gripping force optimization control methods do not consider the influence of external disturbances. In traditional control methods, once external disturbances occur during the execution of tasks by the multi-finger dexterity hand, the target object may shift relative to the multi-finger dexterity hand, or even detach from it, leading to task failure. To address this, a gripping control method that can resist external disturbances is needed. This invention proposes a method for optimizing the grasping force of a multi-fingered dexterous hand based on quadratic programming. This method can resist external disturbances and control the multi-fingered dexterous hand to grasp objects with optimal force, while avoiding damage to objects or insecure grasping. These advantages have significant practical implications for industrial production. In conclusion, this invention patent possesses novelty and practicality. Summary of the Invention

[0004] This invention provides a method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming. The aim is to eliminate the influence of external disturbances, control the multi-finger dexterous hand to grasp objects with optimal force, and avoid damage to objects or insecure grasping.

[0005] The first aspect of this invention provides a quadratic programming scheme for optimizing the grasping force control of a multi-finger dexterous hand;

[0006] The quadratic programming scheme is as follows:

[0007] Its performance metrics are designed to be minimized. Subject to , ,in, It indicates that dexterity is... The force exerted on an object at all times; Represents a parameter vector; Describe a symmetric positive definite matrix; This represents the transformation matrix that converts the force from the contact point coordinate system to the object coordinate system; Indicates in The external force acting on an object at any given moment; Represents the transpose of a matrix or vector; ,in Its physical meaning is to ensure that the multi-fingered dexterous hand does not slip relative to the target object at the point of contact.

[0008] The second aspect of this invention provides a method for solving the optimal grasping force based on neurodynamics; the method proposes an optimal grasping force solver that is resistant to disturbances, and outputs the optimal grasping force of a quadratic programming scheme while eliminating the influence of external disturbances;

[0009] The solution method involves matrixing the quadratic programming scheme and adding an anti-disturbance term, resulting in the final solution formula as follows:

[0010]

[0011] in, This represents the compacted parameter matrix; This represents the compacted parameter vector; This represents the compacted variable vector; External noise can be constant noise, linear noise, random noise, or a superposition of them. This represents the convergence parameter. Attached Figure Description

[0012] Figure 1 This is a flowchart of the present invention;

[0013] Figure 2 To realize the three-dimensional view, front view and top view of the multi-fingered dexterous hand of the present invention;

[0014] Figure 3 To achieve the energy consumption variation diagram under the application of this invention;

[0015] Figure 4 To realize the variation diagram of the force exerted by each finger on the target object in the multi-finger dexterity hand under the application of the present invention;

[0016] Figure 5 A schematic diagram illustrating a method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming, as described in this invention. Detailed Implementation

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of the invention. First, the coordinates of the contact point between the multi-fingered dexterous hand and the target object, as well as the friction coefficient information, are obtained. Then, based on the contact point coordinates, a grasping force transformation matrix from the object coordinate system to the world coordinate system is obtained. Next, an optimal performance index for the grasping force of the multi-fingered dexterous hand is designed, generating a quadratic optimization control method. Next, the quadratic programming scheme is transformed into a nonlinear equation system. Simultaneously, the proposed disturbance-resistant optimal grasping force solver based on neurodynamics is used to solve the equations. Finally, the multi-fingered dexterous hand is controlled to complete the grasping task.

[0019] Figure 2 To realize the multi-finger dexterous hand of the present invention, it is a three-finger large-scale robot DH-3 electric gripper. The multi-finger dexterous hand consists of three fingers, each of which consists of joint one (1), joint two (2), joint three (3) and force sensor (4).

[0020] The quadratic programming scheme for optimizing the grasping force of the multi-finger dexterous hand designed in this invention is as follows:

[0021] Minimize:

[0022] Constraints:

[0023]

[0024] in, It indicates that dexterity is... The force exerted on an object at all times; Represents a parameter vector; Describe a symmetric positive definite matrix; ,in Its physical meaning is to ensure that the multi-fingered dexterous hand does not slip relative to the target object at the point of contact; This represents the transformation matrix that converts the force from the contact point coordinate system to the object coordinate system; Indicates in The external force acting on an object at any given moment; Represents the transpose of a matrix or vector.

[0025] The perturbation-resistant optimal grasping force solver based on neurodynamics designed in this invention is as follows:

[0026] First, the quadratic programming schemes (1)-(3) are transformed into:

[0027]

[0028] in, and For Lagrange multipliers. For ease of expression, the transformed optimization scheme (4) is rearranged into a compact form.

[0029]

[0030] in

[0031]

[0032] Design the error function based on equation (5).

[0033]

[0034] in, express The L2 norm. Error function. right The gradient can be expressed as

[0035]

[0036] Considering that noise is unavoidable in the actual task performance of multi-fingered dexterous hands, an integral term containing historical information is added to eliminate the influence of noise. The derivative of the improved error function (9) is then as follows:

[0037]

[0038] in External noise can be constant noise, linear noise, random noise, or a superposition of them. The parameters are used to control the convergence speed of the error. Substituting the error function (9) into formula (11), we can obtain the following solution for the optimal gripping force based on neurodynamics:

[0039]

[0040] This invention has two key features. First, it considers the diversity of grasping targets in grasping tasks and designs an optimized grasping force control scheme. This avoids damage to objects or unstable grasping during the grasping process. Second, it considers the diversity of grasping scenarios and designs a disturbance-resistant optimal grasping force solver based on neurodynamics. This allows multi-finger dexterity hands to successfully perform grasping tasks even under the influence of external disturbances. In summary, this invention provides a highly adaptable multi-finger dexterity hand control scheme for grasping tasks with increasingly diverse grasping scenarios and targets.

[0041] The workflow of this invention will now be described with reference to a specific example.

[0042] Using MATLAB software, an experimental simulation of the method of this invention was conducted, taking the DH-3 multi-fingered dexterous hand of the Dahuan robot grasping a square object and performing a rotation operation as an example. Specific parameter settings are as follows: the target object is a cube with a side length of 0.1 meters; the coefficient of friction between the multi-fingered dexterous hand and the target object... The coordinates of the point of contact between the two in the object coordinate system are: , and Meters; control error convergence speed parameter and Noise set to This is a mixture of constant noise and Gaussian noise. The optimal grasping force is solved using a neurodynamic-based perturbation-resistant optimal grasping force solver, and the calculated results are transmitted to the multi-fingered dexterous hand, thereby controlling the multi-fingered dexterous hand to complete the grasping task.

[0043] Figure 3 To achieve the energy consumption variation diagram under the application of this invention. From Figure 3 It can be seen that energy consumption quickly stabilized after the mission began. The magnitude indicates that the multi-fingered dexterous hand achieved optimal force grasping of the target object in a short period of time.

[0044] Figure 4 This diagram illustrates the variation in the force exerted by each finger on a target object in a multi-finger dexterous hand, as described in this invention. Figure 4 It can be seen that the force exerted by each finger on the target object converges to the optimal value in a very short time and is maintained until the end of the grasping task, indicating that the proposed neurodynamic-based anti-disturbance grasping force optimization control method can successfully achieve optimal force grasping of the target object under external disturbances.

[0045] Figure 5 This diagram illustrates a method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming, as described in this invention. Figure 5 As shown, the control device completes the transmission of control signals from the solver to the multi-finger dexterous hand through information transmission between the modules of the control device, and finally completes the goal of grasping the target object with the optimal grasping force.

[0046] The control device includes:

[0047] The multi-finger dexterity hand information acquisition module 501 is used to acquire the contact force magnitude information of the multi-finger dexterity hand;

[0048] The grasping force coordinate system transformation module 502 is used to obtain the coordinate information of the contact point between the multi-finger dexterity hand and the target object, and to transform the grasping force from the object coordinate system to the world coordinate system.

[0049] The expected information acquisition module 503 is used to acquire the expected grasping force and transmit the information to the quadratic programming solver;

[0050] The model building module 504 is used to construct a quadratic programming scheme for optimal control of the grasping force of the multi-finger dexterity hand according to preset rules based on the contact point coordinates between the multi-finger dexterity hand and the target object and the expected grasping force information.

[0051] The control signal determination module 505 is used to determine the control signal of the multi-fingered dexterous hand in the environment of external interference by using the proposed neurodynamic-based anti-disturbance optimal grasping force solver.

[0052] Information transmission module 506 is used to acquire the control signal of the multi-finger dexterous hand and transmit the control signal to the lower-level machine;

[0053] The multi-finger dexterous hand control module 507 is used to control the multi-finger dexterous hand according to the control signal of the multi-finger dexterous hand, so that the multi-finger dexterous hand can grasp the target object.

Claims

1. A method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming, characterized in that, The quadratic programming scheme is determined based on the expected movement trajectory information of the target object and the coordinate information of the contact point between the multi-finger dexterity hand and the target object. The quadratic programming scheme is as follows: Its performance metrics are designed to be minimized. Subject to , ,in, It indicates that dexterity is... The force exerted on an object at all times; Represents a parameter vector; Describe a symmetric positive definite matrix; This represents the transformation matrix that converts the force from the contact point coordinate system to the object coordinate system; Indicates in The external force acting on an object at any given moment; Represents the transpose of a matrix or vector; ,in Its physical meaning is to ensure that the multi-fingered dexterous hand does not slip relative to the target object at the point of contact; A perturbation-resistant optimal grasping force solver based on neurodynamics is designed. By proposing a perturbation-resistant optimal grasping force solver based on neurodynamics, the control of multi-fingered dexterity hands can be achieved while eliminating the influence of external perturbations. The solution method is to compactify the quadratic programming scheme and add a perturbation-resistant term, resulting in the final solution formula as follows: in, This represents the compacted parameter matrix; This represents the compacted parameter vector; This represents the compacted variable vector; External noise can be constant noise, linear noise, random noise, or a superposition of them. This represents the convergence parameter.

2. The method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming according to claim 1, characterized in that, The solution results of the quadratic programming solver can be converted into the control signals required for motor drive, driving the motors of each joint so that the multi-fingered dexterous hand can grasp the target object with optimal grasping force.

3. The method for optimizing the grasping force of a multi-finger dexterous hand based on quadratic programming according to claim 2, characterized in that, The control signal is transmitted from the solver to the multi-finger dexterous hand through information transmission between the modules of the control device, and finally the goal of grasping the target object with the optimal grasping force is achieved. The control device includes: A multi-finger dexterity hand information acquisition module is used to acquire information about the magnitude of the contact force of the multi-finger dexterity hand; The grasping force coordinate system transformation module is used to obtain the coordinate information of the contact point between the multi-finger dexterity hand and the target object, and to transform the grasping force from the object coordinate system to the world coordinate system; The expected information acquisition module is used to acquire the expected grasping force and transmit this information to the quadratic programming solver. The model building module is used to construct a quadratic programming scheme for optimizing the grasping force of the multi-finger dexterity hand according to preset rules, based on the contact point coordinates between the multi-finger dexterity hand and the target object and the expected grasping force information. The control signal determination module is used to determine the control signal of the multi-fingered dexterous hand in an environment of external interference by using the proposed neurodynamic-based anti-disturbance optimal grasping force solver. The information transmission module is used to acquire the control signals of the multi-fingered dexterous hand and transmit the control signals to the lower-level machine; The multi-finger dexterous hand control module is used to control the multi-finger dexterous hand according to the control signal of the multi-finger dexterous hand, so that the multi-finger dexterous hand can grasp the target object.

Images