A robot high-precision shaft hole assembly method, system, terminal and storage medium

Abstract

The application discloses a kind of robot high-precision shaft hole assembly method, system, terminal and storage medium, method includes: establishing position and the first impedance control based on quaternion attitude, design state observer based on environmental parameter estimation and the optimizer based on linear quadratic regulator, obtain second impedance control;Sample data under the shaft is not entered hole in second impedance control is collected, and contact state recognition model based on support vector machine is trained;According to contact state recognition model and pre-designed hole entry adjustment model, pose adjustment is carried out until assembly into hole is completed;Realize hole assembly adjustment based on first impedance control and second impedance control control robot.The application divides assembly into two steps of hole entry and in-hole adjustment, refines the process of assembly, designs contact state recognition and adjustment strategy, and realizes high-precision fast hole entry.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of robot assembly, and more particularly relates to a robot high-precision shaft hole assembly method, system, terminal and storage medium. BACKGROUND

[0002] Robot assembly has the advantages of high automation, low cost, and adaptability to harsh environments, and its combination with AGV can enhance adaptability, so using robots for assembly is the current development trend. Most of the assembly robots currently in practical application are based on position control, lack flexibility, and can handle some assembly tasks with large gaps, but lack flexibility when facing high-precision shaft hole assembly tasks, and are prone to cause high force to damage parts and robots.

[0003] To solve the above problems, the prior art usually increases impedance control algorithm based on position to improve the flexibility of the robot. However, the traditional impedance control algorithm cannot be perfectly applied to high-precision shaft hole assembly tasks. First, the force tracking error of the traditional impedance control is related to unknown environmental information, and it is difficult to maintain a stable desired force during assembly. Second, during shaft hole assembly, the workpiece has a process from free assembly to contact, and if the parameters of the impedance control are not properly selected, high force impact will occur, causing damage to the parts and the robot. Finally, since the initial positioning of the assembly usually has errors, this will cause the assembly shaft to fail to enter the hole, and the impedance control cannot handle this problem.

[0004] Based on the above reasons, the present application provides a robot high-precision shaft hole assembly technology based on compliance control and contact state recognition, which realizes high-precision automatic assembly of shaft hole parts. SUMMARY

[0005] In view of the above defects or improvement needs of the prior art, the present application provides a robot high-precision shaft hole assembly method, system, terminal and storage medium, which divides the assembly into two steps of entering the hole and adjusting in the hole, refines the assembly process, designs a contact state recognition and adjustment strategy, and realizes high-precision and fast entry into the hole.

[0006] To achieve the above purpose, according to one aspect of the present application, a robot high-precision shaft hole assembly method is provided, comprising:

[0007] A first impedance control based on position and quaternion attitude is established, a state observer based on environmental parameter estimation and an optimizer based on a linear quadratic regulator are designed, and a second impedance control is obtained;

[0008] Sample data of the shaft not entering the hole under the second impedance control is collected, and a contact state recognition model based on support vector machine is trained;

[0009] According to the contact state recognition model and the pre-designed hole adjustment model, the pose adjustment is performed until the assembly into the hole is completed.

[0010] The robot is controlled based on the first impedance control and the second impedance control to realize the hole assembly adjustment.

[0011] Further, the first impedance control based on the position and the quaternion attitude includes:

[0012] The end trajectory tracking impedance control of the position can be expressed as:

[0013]

[0014] Wherein, M, B, K are 3*3 mass, damping and stiffness diagonal matrices respectively; X is a 3*1 position vector, X r is a 3*1 reference trajectory vector, F ext is a 3*1 external force vector, F d is a 3*1 expected force vector;

[0015] The attitude trajectory tracking impedance control equation based on the quaternion is:

[0016]

[0017] Wherein, M o ,B o ,K o are corresponding 3*3 attitude mass, damping and stiffness diagonal matrices respectively, Q is the current attitude, Q d is the expected attitude, ω is the angular velocity corresponding to the axis angle, ω d is the expected angular velocity, is the angular acceleration, is the expected angular acceleration, T ext is the external torque, T d is the expected torque, is the quaternion multiplication, and log(·) represents the logarithmic mapping.

[0018] Further, the state observer based on the environment parameter estimation includes:

[0019]

[0020] Wherein is the environment parameter estimation value, f d is the expected force, is the external force estimation value, f ext is the actual external force measured by the sensor, λ1, λ2 are parameters, and x is the actual position.

[0021] Further, the designed optimizer based on the linear quadratic regulator includes:

[0022] determining state space equations of impedance control according to the first impedance controller;

[0023] obtaining an algebraic Riccati equation according to the linear quadratic regulator;

[0024] solving the algebraic Riccati equation to obtain a feedback matrix.

[0025] Further, the sample data of the shaft not entering the hole under the second impedance control is collected, and a contact state recognition model based on a support vector machine is trained, comprising:

[0026] The vertical hole surface direction is taken as the z-axis direction, and the remaining directions are taken as the x and y-axis directions. The forces, torques, position deviations and attitude deviations in the x and y directions at different positions and attitudes of the shaft not entering the hole are collected.

[0027] Further, the pose adjustment is performed according to the contact state recognition model and a pre-designed hole entry adjustment model until the assembly into the hole is completed, comprising:

[0028] The position and attitude deviation directions of the shaft and the hole under the forces and torques are recognized according to the contact state recognition model;

[0029] The pre-designed hole entry adjustment model adjusts the position and attitude of the robot according to the recognized position and attitude deviation directions;

[0030] The forces and torques of the shaft and the hole after the adjustment are collected, and the contact state recognition model and the hole entry adjustment model are recognized again. The above process is repeated until the assembly into the hole is completed.

[0031] Further, the pre-designed hole entry adjustment model adjusts the position and attitude of the robot according to the recognized position and attitude deviation directions, comprising:

[0032] The hole entry adjustment model is:

[0033]

[0034]

[0035] wherein y p-cur is the current position model output, y o-cur is the current attitude model output, ξx, ξy are random position increments, ξrx, ξry are random attitude increments (i.e. angle increments), y p1 , y p2 , y p3 , y p4 are pre-specified position offset states, y o1 , y o2 , y o3 , yo4 is a pre-specified position offset state;

[0036] According to the identified position and attitude deviation direction, the robot moves δ x mm along the x-axis and δ y mm along the y-axis, while rotating δ rx ° and δ ry ° around the x-axis and y-axis, respectively.

[0037] According to a second aspect of the present application, a robot high-precision shaft-hole assembly system is provided, comprising:

[0038] A first main module is configured to establish a first impedance control based on a position and a quaternion-based attitude, design a state observer based on environment parameter estimation, and design an optimizer based on a linear quadratic regulator to obtain a second impedance control;

[0039] A second main module is configured to collect sample data of a shaft not entering a hole under the second impedance control, and train a contact state recognition model based on a support vector machine;

[0040] A third main module is configured to perform pose adjustment according to the contact state recognition model and a pre-designed hole-entry adjustment model until the assembly into the hole is completed;

[0041] A fourth main module is configured to control the robot to realize in-hole assembly adjustment based on the first impedance control and the second impedance control.

[0042] According to a third aspect of the present application, an electronic device is provided, comprising a processor and a memory, which are connected to each other;

[0043] The memory is configured to store a computer program;

[0044] The processor is configured to execute the robot high-precision shaft-hole assembly method when the computer program is called.

[0045] According to a fourth aspect of the present application, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to perform the robot high-precision shaft-hole assembly method.

[0046] Overall, compared with the prior art, the above technical solutions conceived by the present application can achieve the following beneficial effects:

[0047] 1.The robot high-precision shaft-hole assembly method of the present application, based on the traditional impedance controller, designs a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to obtain a second impedance control; based on the sample data of the shaft not entering the hole under the second impedance control, a contact state recognition model based on support vector machine is trained; according to the contact state recognition model and the hole entry adjustment model, the pose adjustment is carried out to realize the pose adjustment of the shaft hole entry process; and then according to the traditional impedance control and the second impedance control, the robot is controlled to realize the hole assembly adjustment, and the whole process of the shaft hole assembly is completed. The method of the present application divides the assembly into two steps of hole entry and hole adjustment, refines the assembly process, and realizes the high-precision automatic robot shaft-hole assembly.

[0048] 2.The robot high-precision shaft-hole assembly method of the present application, based on the traditional impedance controller, designs a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to optimize, the improved impedance control can improve the stability of the controller, maintain the stable expected force, reduce the impact and vibration during contact, and at the same time can reduce the dimension of the input space during contact state recognition, and improve the training efficiency.

[0049] 3.The robot high-precision shaft-hole assembly method of the present application, the hole entry task is realized by establishing a contact state model through SVM, and a hole entry adjustment model considering model error is designed, which has high robustness and can quickly and accurately complete the hole entry task. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 It is an adaptive impedance control diagram based on environment parameter estimation of the embodiment of the present application;

[0051] Figure 2 It is a linear quadratic regulator state space block diagram of the embodiment of the present application;

[0052] Figure 3 It is a contact state model hole entry strategy diagram based on support vector machine of the embodiment of the present application;

[0053] Figure 4 It is a data sampling diagram of the embodiment of the present application;

[0054] Figure 5 It is a robot high-precision shaft-hole assembly method flow chart based on compliant control and contact state recognition of the embodiment of the present application;

[0055] Figure 6 It is a robot high-precision shaft-hole assembly method flow chart of the embodiment of the present application;

[0056] Figure 7 It is a schematic diagram of an electronic device of the embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. In addition, the technical features involved in the various embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0058] In the description of the present application, unless otherwise explicitly specified and limited, the terms "connected", "connected", "fixed" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or it can be integrated; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0059] Those skilled in the art can understand that, unless otherwise specified, the singular forms "a", "an" and "the" used herein also include the plural forms. It should be further understood that the phrase "comprising" used in the specification of the present application means that the features, integers, steps, operations, elements and / or components exist, but does not exclude the existence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0060] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as that generally understood by those skilled in the art to which the present application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood as having meanings consistent with those in the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as in the embodiments of the present application.

[0061] The robot high-precision shaft hole assembly method of the present application is applied to industrial robot shaft hole assembly in the fields of aerospace, automobile, energy and the like, and is used for realizing high-precision automatic assembly of shaft holes.

[0062] The state observer is used to estimate the stiffness parameter of the environment and the initial position parameter of the environment, and to generate a reference trajectory of impedance control. The output reference trajectory of the state observer is input as a reference trajectory of impedance control.

[0063] Wherein, LQR (linear quadratic regulator) is a linear system in state space form given in modern control theory, and the objective function is a quadratic function of the object state and control input. LQR optimal design refers to the state feedback controller K designed to minimize the quadratic objective function J, and K is determined by the weight matrix Q and R, so the selection of Q and R is particularly important.

[0064] In the application, the Z axis is the direction of the assembly hole, and the x axis and the y axis are generally not distinguished, and can be the direction of the three-dimensional coordinate axis established according to the right-hand rule, or the direction obtained by other similar methods.

[0065] As shown in Figure 5 and 6 , the application provides a high-precision robot shaft hole assembly method, which divides the assembly into two steps of hole entering and hole adjustment, refines the assembly process, and realizes high-precision automatic robot shaft hole assembly. The method comprises steps S100-S400.

[0066] Step S100, establish a position and a first impedance control based on a quaternion attitude, design a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator, and obtain a second impedance control;

[0067] To ensure six degrees of freedom compliant control, the impedance control of the attitude is added in the impedance control, the differential of the quaternion is realized, and the impedance control of the quaternion attitude is completed. The impedance control of the position is obtained in the same way, and the impedance control of the position and the impedance control based on the quaternion attitude are finally obtained.

[0068] That is, the position end trajectory tracking impedance control equation:

[0069]

[0070] Wherein, M, B and K are 3x3 mass, damping and stiffness diagonal matrices; X is a 3x1 position vector, X r is a 3x1 reference trajectory vector, F ext is a 3x1 external force vector, F d is a 3x1 expected force vector;

[0071] The attitude trajectory tracking impedance control equation based on the quaternion is:

[0072]

[0073] In the formula, M o ,B o ,K oare the corresponding 3x3 pose mass, damping and stiffness diagonal matrices, Q is the current pose, Q d is the desired pose, ω is the angular velocity corresponding to the axis angle, ω d is the desired angular velocity, is the angular acceleration, is the desired angular acceleration, T ext is the external torque, T d is the desired torque, is the quaternion multiplication, and log(·) represents the logarithmic mapping.

[0074] To ensure that the steady-state force error of the system is 0, improve the accuracy of force control, and simplify the subsequent assembly contact state recognition process, when the shaft hole is assembled, constant force control is performed on the z-axis, and the state observer is used to estimate the stiffness parameter and initial position parameter of the environment parameter, and then the reference trajectory of the impedance controller is given, and the impedance control after adding the state observer of the environment parameter estimation is as shown in Figure 1 .

[0075] The state observer based on environment parameter estimation comprises:

[0076]

[0077] In the formula is the environment parameter estimation value, f d is the desired force, is the external force estimation value, f ext is the actual external force measured by the sensor, λ1 and λ2 are parameters, x is the actual position, and t is time.

[0078] In order to reduce the impact generated in the transition state and the vibration of the system under environmental constraints, and further ensure the safety of the assembly process, the optimizer of the linear quadratic regulator is designed. By adjusting the feedback matrix K, the appropriate impedance parameter gain is determined, so that the tracking effect of the system in the transition state and under the environmental constraints can be improved.

[0079] The state space equation of the impedance control is determined according to the first impedance controller;

[0080] The algebraic Riccati equation is obtained according to the linear quadratic regulator;

[0081] The feedback matrix is obtained by solving the algebraic Riccati equation.

[0082] Specifically, the state space equation of the impedance control can be obtained according to the first impedance control equation:

[0083] u=f ext -f d +kxr

[0084] Where m, b, and k are the mass, damping, and stiffness coefficients, respectively, and x is the actual trajectory. For velocity, x r For reference trajectory;

[0085] Based on the principle of the linear quadratic regulator, the algebraic Riccati equation can be obtained:

[0086] A T P+PA+Q-PBR -1 B T P=0

[0087] Where Q and R are parameter matrices, and T is the matrix transpose. Solving this equation yields matrix P, which can then be further solved using K = R. - 1 B T P yields the feedback matrix K. For impedance control:

[0088] K = [k1 k2]

[0089] It should be noted that the solution process for the algebraic Riccati equation is based on existing technology and will not be elaborated here.

[0090] To address the issue of assembly failure due to pose errors in shaft hole positioning during assembly, the SVM method is used to obtain the mapping relationship between pose offset (i.e., position and orientation offset) and force and torque, thereby obtaining the contact state model during shaft hole assembly. Furthermore, an assembly adjustment strategy for the robot is designed to compensate for the pose errors generated during object positioning and complete the assembly task.

[0091] Step S200: Collect sample data of shafts not entering holes under the second impedance control, and train a contact state recognition model based on support vector machine;

[0092] Under normal circumstances, the force information collected by the force sensor is six-dimensional, including force and torque in three directions: x, y, and z (the z-axis direction is perpendicular to the hole surface). The torque in the z-axis can be ignored during shaft-hole assembly. Under the control of the second impedance, the desired force in the z-direction is a constant value, so the force in the z-axis can be ignored.

[0093] like Figure 4 As shown, the positioning error range is determined, and samples are uniformly collected within the positioning error range. Force and torque on the x-axis and y-axis, position offset (marking the offset direction on the x-axis and y-axis), and attitude offset (marking the offset direction on the x-axis and y-axis) are collected to obtain two main categories of sample data. The first category is those that successfully entered the hole, and the second category is those that failed to enter the hole. During training, only the samples that failed to enter the hole are trained.

[0094] Specifically, in the contact state recognition model training, the input feature vector is:

[0095] X = (f x x y m x m y ) T

[0096] Where f x , m x , f y , m y are the force and moment of the x and y axes respectively.

[0097] The output space of the position offset is:

[0098] Y pos = [y p1 y p2 y p3 y p4 ] T

[0099] With + indicating positive offset along the axis, - indicating negative offset along the axis, then y p1 , y p2 , y p3 , y p4 respectively represent (-x, -y), (-x, +y), (+x, -y), (+x, +y).

[0100] The output space of the attitude offset is:

[0101] Y ori = [y o1 y o2 y o3 y o4 ] T

[0102] With + indicating positive offset along the axis, - indicating negative offset along the axis, then y o1 , y o2 , y o3 , y o4 respectively represent (-r x , -r y ), (-r x , +r y ), (+r x , -r y ), (+r x , +r y ).

[0103] In one embodiment of the present application, a position and a posture contact control model are respectively established based on the sample data of the unsuccessful hole entry by using the support vector machine method.

[0104] In another embodiment of the present application, a position and a posture contact control model are established based on the sample data of the unsuccessful hole entry by using the support vector machine method, which is used for simultaneously and directly identifying the position and the posture.

[0105] The contact control model is trained by the support vector machine method, and a mapping relationship between the relative position and posture (position and posture) of the shaft hole when failing to enter the hole and the force and the torque can be obtained. Based on the mapping relationship, the position and the posture deviation of the shaft hole when failing to enter the hole can be determined by the force and the torque at different positions collected by the force sensor.

[0106] In step S300, the posture is adjusted according to the contact state identification model and the pre-designed hole entry adjustment model until the assembly into the hole is completed, as shown in Figure 3 .

[0107] Specifically, step S300 includes steps S301-S303.

[0108] In step S301, the position and the posture deviation direction under the force and the torque when the shaft contacts with the hole are identified according to the contact state identification model.

[0109] Specifically, the trained contact state identification model of the position and the posture can output the current position and posture deviation direction (indicated by "+", "-") according to the force and the torque collected by the force sensor.

[0110] In step S302, the pre-designed hole entry adjustment model adjusts the position and the posture of the robot according to the identified position and posture deviation direction.

[0111] In order to realize the assembly of the robot into the hole, the adjustment strategy of the robot when failing to assemble into the hole needs to be designed, and the position and the posture of the robot are adjusted.

[0112] When the shaft fails to enter the hole during the assembly, the adjustment into the hole strategy according to the current contact state is that the robot is respectively moved by δ x mm along the x-axis and by δ y mm along the y-axis, and is respectively rotated by δ rx ° around the x-axis and by δ ry ° around the y-axis. On this basis, based on the model error, the randomness is increased to obtain the hole entry adjustment model:

[0113]

[0114]

[0115] wherein, y p-cury is the current position model output o-cur ξx, ξy are random position increments, and ξrx, ξry are random attitude increments (i.e., angle increments).

[0116] It should be noted that, taking the position adjustment of the robot assembly shaft into the hole as an example, for the above hole entry adjustment model, there are four states output by the contact state recognition model, which correspond to y p1 ,y p2 ,y p3 ,y p4 Correspondingly, the hole entry adjustment model controls the robot to move along the x-axis by δ x mm and the y-axis by δ y mm, respectively, according to the state (when the position deviation direction output by the contact state recognition model is y p1 , When the position deviation direction output by the contact state recognition model is y p2 , δx = 0.4 + ξ x , δy = -0.4 + ξ y , and the rest of the position adjustment is the same) The attitude adjustment method of the robot assembly shaft into the hole is the same, and will not be repeated here.

[0117] In an embodiment of the present application, the ranges of ξx, ξy, ξrx, and ξry are [-0.2, +0.2].

[0118] Specifically, the hole entry adjustment model outputs the adjustment parameters of the position and attitude according to the position and attitude deviation direction, adjusts the position and attitude based on the parameters, and continues to move downward based on the adjusted position and attitude until the shaft contacts the hole.

[0119] Step S303, based on the force and torque when the adjusted shaft contacts the hole, the contact state recognition model and the hole entry adjustment model are identified again, and the above process is repeated until the assembly into the hole is completed.

[0120] After the position and attitude are adjusted, the force and torque change, and the contact state recognition model outputs the position and attitude deviation direction again according to the force and torque measured by the force sensor, and the hole entry adjustment model adjusts the position and attitude again until the assembly into the hole is completed.

[0121] It should be noted that in the above process, the force and torque of different positions and attitudes are measured by the force sensor.

[0122] After the assembly into the hole is completed, the shaft needs to be further inserted into the hole, and the position and attitude adjustment in the hole is needed, otherwise the situation of being stuck may occur.

[0123] Step S400, based on the first impedance control and the second impedance control, the six-dimensional force of the robot is controlled to realize the adjustment of the assembly in the hole.

[0124] Specifically, the step S400 includes steps S401-S402.

[0125] The step S401 controls the z-axis direction force of the impedance controller by using the second impedance control.

[0126] The step S402 controls the remaining five dimensions of the six-dimensional force of the impedance controller by using the first impedance control.

[0127] After entering the hole, the z-axis direction force of the impedance controller uses the second impedance control described above, and the remaining five dimensions can use the traditional impedance controller (the first impedance control) to provide compliance because the desired force and the desired torque are usually set to 0.

[0128] The embodiment of the present application also provides a robot high-precision shaft hole assembly system for realizing the robot high-precision shaft hole assembly method of the embodiment of the present application.

[0129] The first main module is used for establishing a position and a first impedance control based on a quaternion attitude, designing a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator, and obtaining a second impedance control.

[0130] The second main module is used for collecting sample data of a shaft not entering a hole under the second impedance control, and training a contact state recognition model based on a support vector machine.

[0131] The third main module is used for adjusting a pose according to the contact state recognition model and a pre-designed hole entry adjustment model until the assembly into the hole is completed.

[0132] The fourth main module is used for controlling the robot to realize the in-hole assembly adjustment based on the first impedance control and the second impedance control.

[0133] It should be noted that the robot high-precision shaft hole assembly system provided by the embodiment of the present application can also be a computer program (including program code) running in a computer device, for example, the robot high-precision shaft hole assembly system is an application program, which can be used to execute corresponding steps in the above method provided by the embodiment of the present application.

[0134] In some possible implementation manners, the robot high-precision shaft-hole assembly system provided in the embodiment can be implemented in a combination of software and hardware. For example, the direct injection engine oil pressure control system provided in the embodiment can be a hardware decoding processor programmed to execute the robot high-precision shaft-hole assembly method provided in the embodiment. For example, the hardware decoding processor can be one or more of an application specific integrated circuit (ASIC), a digital signal processor (DSP), a programmable logic device (PLD), a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), or other electronic elements.

[0135] In some possible implementation manners, the robot high-precision shaft-hole assembly system provided in the embodiment can be implemented in a software manner, which can be software in the form of a program and a plug-in, and include a series of modules.

[0136] The robot high-precision shaft-hole assembly system provided in the embodiment establishes a first impedance control based on a position and a quaternion attitude, designs a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator, and obtains a second impedance control. Sample data of a shaft not entering a hole under the second impedance control is collected, and a contact state recognition model based on a support vector machine is trained. A pose is adjusted according to the contact state recognition model and a pre-designed hole entry adjustment model until assembly into the hole is completed. The robot is controlled to realize in-hole assembly adjustment based on the first impedance control and the second impedance control. The assembly is divided into two steps of hole entry and in-hole adjustment, the assembly process is refined, a contact state recognition and adjustment strategy is designed, and high-precision and rapid hole entry is achieved.

[0137] The embodiment of the present application further provides an electronic device, Figure 7 is a structural schematic diagram of the electronic device of the embodiment, like Figure 7As shown, the electronic device 1000 in this embodiment may include: a processor 1001, a network interface 1004, and a memory 1005. Furthermore, the electronic device 1000 may also include: a user interface 1003, and at least one communication bus 1002. The communication bus 1002 is used to implement communication between these components. The user interface 1003 may include a display screen and a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1004 may be a high-speed RAM or non-volatile memory, such as at least one disk storage device. The memory 1005 may optionally be at least one storage device located remotely from the processor 1001. Figure 7 As shown, the memory 1005, which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a device control application.

[0138] like Figure 7 In the illustrated electronic device 1000, the network interface 1004 provides network communication functionality; the user interface 1003 primarily provides an input interface for the user; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

[0139] A first impedance control based on position and quaternion-based posture is established. A state observer based on environmental parameter estimation and an optimizer based on a linear quadratic regulator are designed to obtain a second impedance control. Sample data of the shaft not yet inserted into the hole under the second impedance control are collected, and a contact state recognition model based on a support vector machine is trained. The pose is adjusted according to the contact state recognition model and the pre-designed hole insertion adjustment model until the assembly into the hole is completed. The robot is controlled based on the first and second impedance controls to achieve in-hole assembly adjustment. This invention divides the assembly into two steps: hole insertion and in-hole adjustment, refines the assembly process, and designs contact state recognition and adjustment strategies to achieve high-precision and rapid hole insertion.

[0140] It should be understood that in some possible implementations, the processor 1001 described above can be a central processing unit (CPU), and the processor can also be other general-purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The memory can include read-only memory and random access memory, and provide instructions and data to the processor. Part of the memory can also include non-volatile random access memory. For example, the memory can also store device type information.

[0141] In a specific implementation, the electronic device 1000 described above can perform the implementation manner provided by each of the steps described above through each functional module built therein, and specific implementation manners can be referred to the implementation manners provided by each of the steps described above, which will not be described here. Figure 6

[0142] The electronic device provided in the embodiment establishes a first impedance control based on a position and a quaternion attitude, designs a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to obtain a second impedance control; sample data of an axis not entering a hole under the second impedance control is collected, and a contact state recognition model based on a support vector machine is trained; pose adjustment is performed according to the contact state recognition model and a pre-designed hole entry adjustment model until assembly into the hole is completed; and a robot is controlled to realize in-hole assembly adjustment based on the first impedance control and the second impedance control. The present application divides assembly into two steps of hole entry and in-hole adjustment, refines the assembly process, designs a contact state recognition and adjustment strategy, and realizes high-precision and rapid hole entry.

[0143] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method provided by each of the steps described above. Specific implementation manners can be referred to the implementation manners provided by each of the steps described above, which will not be described here. Figure 6

[0144] The computer storage medium establishes a first impedance control based on a position and a quaternion attitude, designs a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to obtain a second impedance control; sample data of an axis not entering a hole under the second impedance control is collected, and a contact state recognition model based on a support vector machine is trained; pose adjustment is performed according to the contact state recognition model and a pre-designed hole entry adjustment model until assembly into the hole is completed; and a robot is controlled to realize in-hole assembly adjustment based on the first impedance control and the second impedance control. The present application divides assembly into two steps of hole entry and in-hole adjustment, refines the assembly process, designs a contact state recognition and adjustment strategy, and realizes high-precision and rapid hole entry. ​​

[0145] It should be understood that although the steps in the flowcharts of the drawings are shown in sequential order, such that each individual step is executed after the preceding step has been completed, the steps do not have to be executed in this order. Unless specifically stated in the specification, the execution of the steps is not limited to a strict sequential order, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the drawings can comprise multiple sub-steps or stages, which do not have to be executed at the same time, but can be executed at different times, and the order of execution of which is not necessarily sequential, but can be round-robin or alternating with at least some of the other steps or sub-steps or stages of other steps.

[0146] Those skilled in the art will easily understand that the above-mentioned is only the preferred embodiment of the present application, and is not intended to limit the present application, and any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A robot high-precision shaft-hole assembly method, characterized by, The method comprises the following steps: establishing a first impedance control based on position and quaternion attitude, designing a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to obtain a second impedance control; collecting sample data of the shaft not entering the hole under the second impedance control, and training a contact state recognition model based on a support vector machine; adjusting the position and attitude according to the contact state recognition model and a pre-designed hole entry adjustment model until the assembly into the hole is completed; controlling the robot to realize the assembly adjustment in the hole based on the first impedance control and the second impedance control.

2. The robot high-precision shaft hole assembly method according to claim 1, characterized in that, The first impedance control based on position and quaternion attitude comprises: the end trajectory tracking impedance control of the position is: where M, B, K are 3x3 mass, damping and stiffness diagonal matrices, respectively; X is a 3x1 position vector, X r is a 3x1 reference trajectory vector, F ext is a 3x1 external force vector, F d is a 3x1 desired force vector; the attitude trajectory tracking impedance control equation based on the quaternion is: where M o , B o , K o are the corresponding 3x3 attitude mass, damping and stiffness diagonal matrices, respectively, Q is the current attitude, is the desired attitude, ω is the angular velocity corresponding to the axis angle, ω d is the desired angular velocity, is the angular acceleration, is the desired angular acceleration, T ext is the external torque, T d is the desired torque, is the quaternion multiplication, log(·) denotes the logarithmic mapping.

3. The robot high-precision shaft hole assembly method according to claim 1, characterized in that, The state observer based on the environment parameter estimation comprises: wherein is an environmental parameter estimate, f d is a desired force, is an external force estimate, f ext is a sensor measured actual external force, λ1, λ2 are parameters, x is an actual position.

4. The robot high-precision shaft hole assembly method according to claim 1, characterized in that, The optimizer based on the linear quadratic regulator comprises: determining the state space equation of the impedance control according to the first impedance controller; obtaining the algebraic Riccati equation according to the linear quadratic regulator; solving the algebraic Riccati equation to obtain the feedback matrix.

5. The robot high-precision shaft hole assembly method according to claim 1, wherein The collection of sample data of the shaft not entering the hole under the second impedance control comprises: taking the direction perpendicular to the hole surface as the z-axis direction, and collecting the forces, moments, position deviations and attitude deviations in the x and y directions at different positions and attitudes when the shaft does not enter the hole.

6. The robot high-precision shaft hole assembly method according to claim 1, wherein The adjustment of the position and attitude according to the contact state recognition model and the pre-designed hole entry adjustment model until the assembly into the hole is completed comprises: identifying the position and attitude deviation directions under the forces and moments when the shaft contacts the hole according to the contact state recognition model; the pre-designed hole entry adjustment model adjusts the position and attitude of the robot according to the identified position and attitude deviation directions; collecting the forces and moments when the shaft contacts the hole after the adjustment, and identifying the adjustment again according to the contact state recognition model and the hole entry adjustment model, and repeating the above process until the assembly into the hole is completed.

7. The robot high-precision shaft hole assembly method according to claim 6, characterized in that, The pre-designed hole entry adjustment model adjusts the position and attitude of the robot according to the identified position and attitude deviation directions comprises: the hole entry adjustment model is: where y p-cur is the current position model output, y o-cur is the current pose model output, ξx, ξy are random position increments, ξrx, ξry are random pose increments, y p1 , y p2 , y p3 , y p4 is a pre-specified position offset state, y o1 , y o2 , y o3 , y o4 is a pre-specified position offset state; According to the identified position and orientation deviation direction, the robot moves δ x mm along the x-axis and δ y mm along the y-axis, while rotating δ rx ° around the x-axis and δ ry ° around the y-axis, respectively.

8. A robotic high precision shaft hole assembly system, characterized by, comprises: a first main module for establishing a first impedance control based on position and quaternion attitude, designing a state observer based on environment parameter estimation and an optimizer based on a linear quadratic regulator to obtain a second impedance control; a second main module for collecting sample data of the shaft not entering the hole under the second impedance control, and training a contact state recognition model based on a support vector machine; a third main module for adjusting the position and attitude according to the contact state recognition model and a pre-designed hole entry adjustment model until the assembly into the hole is completed; a fourth main module for controlling the robot to realize the assembly adjustment in the hole based on the first impedance control and the second impedance control.

9. An electronic device, comprising: comprises a processor and a memory, and the processor and the memory are connected to each other; the memory is used for storing a computer program; the processor is configured to execute the robot high-precision shaft hole assembly method as claimed in any one of claims 1 to 7 when the computer program is called.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the robot high-precision shaft hole assembly method as claimed in any one of claims 1 to 7.

Images