A heavy-duty industrial robot adaptive and compliant large-surface fitting system and control method thereof

Abstract

The present invention discloses a self-adaptive and flexible large-surface fitting system for a heavy-duty industrial robot and a control method thereof, so as to solve the problem of low absolute positioning accuracy of the industrial robot and meet the high-precision assembly requirements of large and complex structural components. The system includes an AGV trolley, a control cabinet, a robot body, a robot bearing connection column, a robot terminal connection flange, a terminal assembly execution component, an assembly component, and a component assembly fixing component; the control method includes: establishing a coordinate system relationship between the robot body and the terminal assembly execution component, building a compensation calculation model and an admittance control model, analyzing the admittance control, and performing control algorithm calculations by real-time monitoring of the component posture, monitoring the external contact force, torque, and the robot terminal acceleration, angular acceleration changes, and correcting the robot posture. The present invention realizes high-precision assembly of heavy-duty industrial robots, performs flexible assembly of customized components, and improves assembly efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of heavy-load industrial robot assembly, and discloses a heavy-load industrial robot self-adaptive and compliant large-curved surface fitting system and a control method thereof. Background Art

[0002] With the continuous development of the manufacturing industry, industrial robots are playing an increasingly important role in the production process. Among them, heavy-duty industrial robots are widely used in the fields of automobile manufacturing, aerospace, etc., undertaking key tasks such as assembly and transportation.

[0003] However, these industrial robots often have difficulty in achieving accurate fitting and stable assembly when faced with complex parts. First, the complex workpieces themselves have certain deformations and deviations, which makes it difficult for the robot to accurately position and fit. Traditional robot end effectors usually use rigid structures that cannot effectively adapt to the ups and downs of the workpiece surface, which can easily lead to loose fitting and even damage to parts. Secondly, the heavy-duty industrial robots themselves have complex structures, and there is a certain amount of error accumulation in the transmission of each joint, which will also affect the accuracy of the final assembly.

[0004] To solve the above problems, patent CN114454060A discloses an unknown surface adaptive constant pressure tracking system and its control method. This method installs one end of the position adaptive actuator on the end of the industrial robot, connects the processing tool on the other end and contacts the unknown surface to be tracked through the processing tool. The three-dimensional unknown surface adaptive constant pressure tracking is achieved through the posture adjustment of the industrial robot and the one-dimensional force control of the actuator. Although this method can achieve constant pressure tracking, it is limited to small planes and small range tracking, and cannot achieve large range tracking. It has great limitations for certain assembly positions.

[0005] Patent CN109910005A discloses a variable admittance control method and system for robots. This method uses a compliant control method of variable admittance to enable a human-machine interactive robot to respond quickly and sensitively to the operator's intentions, and considers the assembly comfort of the assembler for low-speed and low-acceleration interaction between humans and robots. However, no relevant settings are made for the situation where the robot collides with a rigid assembly workpiece during the assembly process, especially for large curved and heavy-loaded workpieces. This method does not involve relevant situations.

[0006] In summary, the flexible end effector is used to achieve passive adaptation to the workpiece surface using adaptive elastic components, but this method is often difficult to accurately track and locate over a large range and is difficult to meet the requirements of large-surface operations. Active compliance control is performed through the admittance control algorithm, but most of the existing algorithms are aimed at human-robot interaction and still have certain limitations for large-surface assembly of heavy-duty industrial robots. Therefore, a new control method is urgently needed for large-surface assembly of heavy-duty industrial robots. Summary of the invention

[0007] In order to solve the problems existing in the above-mentioned prior art, the present invention provides the following technical solutions:

[0008] A heavy-duty industrial robot adaptive and compliant large-curved surface fitting system, comprising:

[0009] AGV trolley, control cabinet, robot body, robot bearing connection column, robot terminal connection flange, terminal assembly actuator, assembly components and component assembly fixing components;

[0010] The AGV trolley is used to receive external position movement instructions and move to a specified position point; the upper front end thereof carries the robot body through the robot carrying connecting column, and the upper rear end directly carries the control cabinet;

[0011] There is a communication connection between the control cabinet and the robot body, which is used to control the robot body to correct the joint posture; the end of the robot body is connected to the end assembly execution component through the robot end connection flange, and the robot body moves the end assembly execution component to the specified assembly posture point according to the control instruction of the control cabinet;

[0012] The terminal assembly execution component absorbs the assembly component through the suction cup component located on its lower side; the component assembly fixing component is fixed at a specified assembly posture point, and its upper side is the surface component to be assembled; when the terminal assembly execution component moves to the specified assembly posture point, the assembly component and the surface component to be assembled achieve curved surface fitting.

[0013] Furthermore, the terminal assembly execution component also includes:

[0014] A force sensor, a force sensor connecting flange, an inertial sensor, an inertial sensor connecting plate, and an actuator body; the actuator body is located in the middle of the terminal assembly actuator, a force sensor is installed on its upper side through the force sensor connecting flange, and an inertial sensor is installed on its side through the inertial sensor connecting plate; the suction cup assembly includes five pairs of ten suction cups, which are symmetrically installed on both sides below the actuator body.

[0015] Furthermore, the component assembly fixing member comprises:

[0016] Surface components to be assembled, rotating components, and positioning support components;

[0017] The positioning support assembly includes: a frame support assembly for supporting and fixing the surface assembly to be assembled and the rotating assembly, four adjustable base support assemblies installed at the four corners of the bottom of the frame support assembly, and four non-adjustable base support assemblies installed at the bottom of both sides of the frame support assembly and distributed symmetrically;

[0018] The rotating assembly is movably mounted on the upper ends of both sides of the frame support assembly;

[0019] The assembly of the surface to be assembled includes a frame assembly of the surface to be assembled and a surface to be assembled installed above the frame assembly; one side below the frame assembly of the surface to be assembled is fixedly connected to the frame support assembly, and the other side below the frame assembly is fixedly connected to the rotating assembly.

[0020] The present invention also provides a control method for a heavy-duty industrial robot adaptive compliant large curved surface fitting system, which specifically comprises the following steps:

[0021] S1. Based on the force sensor, the force sensor receives the three-axis force and the three-axis moment of the components below the force sensor; and the gravity center of gravity parameters of the components below the force sensor are identified, so as to compensate the components below the force sensor in real time and obtain the actual three-axis force and three-axis moment of the components below the force sensor caused by the external force; the components below the force sensor include: a force sensor connecting flange, an inertial sensor, an inertial sensor connecting plate, an actuator body, a suction cup assembly, and an assembly component;

[0022] S2, receiving the three-axis acceleration and three-axis angular velocity of the terminal assembly actuator at the specified assembly posture point based on the inertial sensor; performing zero drift parameter identification on the inertial sensor, thereby performing real-time compensation on the inertial sensor and obtaining the actual three-axis acceleration and three-axis angular velocity;

[0023] S3, real-time monitoring of the actual three-axis force, three-axis torque, three-axis acceleration and three-axis angular velocity after compensation, and continuous correction of the current robot body end position through control algorithm feedback until the assembly parts and the surface to be assembled achieve the expected surface fitting effect.

[0024] Furthermore, step S1 specifically includes:

[0025] S11. Construct world coordinate system O w -X w Y w Z w , the robot body coordinate system is O r -X r Y r Z r , Robot end connection flange coordinate system O f -X f Y f Z f And the force sensor coordinate system is O s -X s Y s Z s ; Set the Z of the world coordinate system wThe axis is parallel to and opposite to the direction of gravity, constructing the attitude transformation matrix of the force sensor coordinate system relative to the world coordinate system The formula is:

[0026]

[0027] in, is the posture transformation matrix of the robot body coordinate system relative to the world coordinate system; The posture transformation matrix of the robot end connection flange coordinate system relative to the robot body coordinate system; It is the posture transformation matrix of the force sensor coordinate system relative to the robot end connection flange coordinate system;

[0028] S12, determine the position point of the component assembly fixing component, the robot body moves near the position point, and stops at six position points, and obtains the coordinates of the robot body of the six position points in the Cartesian coordinate system: (X, Y, Z, A, B, C), where X, Y, Z are the position of the robot in the Cartesian coordinate system; A, B, C are the posture of the robot in the Cartesian coordinate system; receive in real time through the inertial sensor, and after Kalman filtering, obtain the three-axis force and three-axis torque received by the filtered force sensor at the corresponding coordinates;

[0029] S13, according to the coordinate system transformation relationship in step S11 and the three-axis force and three-axis moment obtained in step S12, the gravity G and the center of gravity (G x , G y , G z ), the roll angle φ and pitch angle w of the robot body base, and the zero drift value of the force sensor (F x 、F y 、F z 、M x 、M y 、M z );

[0030] Among them, G x , G y , G z F is the X, Y, and Z components of the gravity center of the components below the force sensor in the Cartesian coordinate system; x 、F y 、F z M is the zero-drift triaxial force of the components below the force sensor; x 、M y 、M z is the zero drift triaxial moment of the components below the force sensor;

[0031] S14, according to the parameters calculated in step S13, combined with the coordinates of the robot body in the Cartesian coordinate system obtained in real time, and the three-axis force and three-axis moment received by the force sensor after filtering at the corresponding coordinates, the actual three-axis force (f x 、f y 、f z ) and actual triaxial moment (t x ,t y ,t z ).

[0032] Furthermore, step S2 specifically includes:

[0033] S21. Construct world coordinate system O w -X w Y w Z w , the robot body coordinate system is O r -X r Y r Z r , Robot end connection flange coordinate system O f -X f Y f Z f and the inertial sensor coordinate system O g -X g Y g Z g , set the Z of the world coordinate system w The axis is parallel to and opposite to the direction of gravity, constructing the attitude transformation matrix of the inertial sensor coordinate system relative to the world coordinate system The formula is:

[0034]

[0035] in, is the posture transformation matrix of the robot body coordinate system relative to the world coordinate system; The posture transformation matrix of the robot end connection flange coordinate system relative to the robot body coordinate system; is the attitude transformation matrix of the inertial sensor coordinate system relative to the robot end connection flange coordinate system;

[0036] S22, determine the position point of the component assembly fixing member, the robot body moves near the position point, and stops at six position points, and obtains the coordinates of the robot body of the six position points in the Cartesian coordinate system in real time: (X, Y, Z, A, B, C); receive in real time through the inertial sensor, and through Kalman filtering, obtain the three-axis acceleration and three-axis angular velocity received by the inertial sensor after filtering at the corresponding coordinates;

[0037] S23, based on the coordinate system conversion relationship constructed in step S21 and the three-axis acceleration and three-axis angular velocity obtained in step S22, the calibration coefficient (K) of the inertial sensor is obtained through the inertial compensation algorithm. x , K y , K z , K a , K b , K c ) and zero drift value (A x , A y , A z , W a , W b , W c ), where K x , K y , K z K is the calibration coefficient of the three-axis acceleration received by the inertial sensor; a , K b , K c A is the calibration coefficient of the three-axis angular velocity received by the inertial sensor; x , A y , A z is the zero drift value of the three-axis acceleration received by the inertial sensor; W a , W b , W c It is the zero drift value of the three-axis angular velocity received by the inertial sensor;

[0038] S24, according to the calibration coefficient and zero drift value obtained in step S23, the inertial sensor is compensated; combined with the coordinates of the robot body in the Cartesian coordinate system obtained in real time in step S22, and the three-axis acceleration and three-axis angular velocity received by the inertial sensor after filtering under the corresponding coordinates, the actual acceleration (a) of the terminal assembly actuator is calculated in real time by the zero drift removal algorithm. x 、a y 、a z ) and the actual three-axis angular velocity (w a 、w b 、w c ), and then further calculate the actual three-axis angular acceleration (a a 、a b 、a c ).

[0039] Furthermore, step S3 specifically includes:

[0040] S31. According to the actual three-axis force, three-axis torque, three-axis acceleration and three-axis angular velocity obtained in step S1 and step S2, an admittance control model is designed as the control algorithm for the assembly task. The formula of the admittance control model when the robot body performs one-dimensional translation motion is expressed as follows:

[0041]

[0042] Among them, m is the virtual mass; b is the virtual damping; k is the virtual stiffness; is the acceleration of the robot end; is the speed of the robot end; x is the position of the robot end; f is the component below the force sensor, that is, the actual force on the robot end; is the expected acceleration of the robot end; is the expected speed of the robot end; x n is the desired position of the robot end; f n is the expected force at the end of the robot body; a is the end assembly actuator, that is, the actual acceleration at the end of the robot body; x n 、f n Set to zero, by calculating x and x n The difference between the two is used to determine the assembly task direction and assembly speed;

[0043] S32. When adjusting the posture, in order to ensure the assembly effect and improve the assembly efficiency, when a collision occurs, the motion rigidity is reduced and the vibration amplitude is reduced; when rebounding, the motion rigidity is increased and the position adjustment time and space are increased, that is, the parameter k is classified and set, and the classification formula is as follows:

[0044]

[0045] Among them, k n is the effective virtual stiffness; λ1 is the virtual stiffness adjustment parameter 1; λ2 is the virtual stiffness adjustment parameter 2; ρ is the acceleration adjustment parameter;

[0046] S33. Design the admittance control model for one-dimensional rotational motion, the formula is expressed as:

[0047]

[0048] Among them, α is the virtual moment of inertia; β is the virtual rotation damping; ψ is the virtual rotation stiffness; is the angular acceleration of the robot end; is the angular velocity of the robot end; θ is the rotation angle of the robot end; t is the component below the force sensor, that is, the actual torque exerted on the robot end; is the expected angular acceleration of the robot end; is the desired angular velocity of the robot end; θ n is the expected rotation angle of the robot end; t n is the desired torque at the end of the robot body; aθ The end assembly actuator is the actual angular acceleration of the end of the robot body; θ n ,t n Set to zero, by calculating θ and θ n The difference between the two is used to determine the assembly task direction and assembly speed;

[0049] S34, continuously adjust the posture and monitor in real time, when the coordinates (X, Y, Z, A, B, C) of the robot body in the Cartesian coordinate system are monitored to be within a certain error fluctuation range within a certain period of Δt, and the actual three-axis force (f x 、f y 、f z ) and triaxial moment (t x ,t y ,t z ) in the corresponding assembly position, corresponding to f z The parameters are maintained within a fixed fluctuation range, and the rest of the f x 、f y ,t x ,t y ,t z The parameter fluctuates around zero; the actual acceleration (a x 、a y 、a z ) and the actual angular acceleration (a a 、a b 、a c ) maintains fluctuations around zero, the assembly task is considered completed.

[0050] Based on the above technical solution, the present invention has the following beneficial effects:

[0051] 1) The heavy-duty industrial robot adaptive large-curved surface bonding system proposed in the present invention achieves high flexibility through the combination of an AGV trolley and a heavy-duty industrial robot system, which can greatly expand the assembly workspace. At the same time, thanks to the use of heavy-duty industrial robots, the system can achieve high-load flexible surface assembly. Combined with flexible end-assembly actuators, large-curved surface parts can be flexibly fixed and clamped, reducing product defects caused by assembly errors, collisions and other factors, and significantly improving assembly accuracy and consistency;

[0052] 2) The heavy-load industrial robot adaptive large-curved surface bonding system proposed in the present invention adopts force sensors and inertial sensors to collect real-time data of the sensor and robot posture, and can monitor the working status of the system in real time. Through the feedback mechanism, the safety risks in the high-load and large-curved surface assembly process are reduced, and the safety of the work is improved;

[0053] 3) The control method of the heavy-duty industrial robot adaptive large-surface bonding system proposed in the present invention greatly shortens the bonding assembly time, improves the overall efficiency of the production line, and reduces labor and time costs through optimized admittance control algorithm and efficient robot adaptive motion planning control algorithm. In particular, the algorithm is optimized for collision events, and the robot posture correction efficiency after each collision event is improved by adjusting the stiffness of the assembly parts before and after the collision, greatly reducing the equipment damage rate caused by the collision event. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is the overall structure diagram of the system of the present invention;

[0055] Figure 2 This is a schematic diagram of the structure of the terminal assembly execution component in the system proposed by the present invention;

[0056] Figure 3 A schematic diagram of the structure of the assembly fixing components in the system proposed by the present invention;

[0057] Figure 4 This is a flow chart of the control method proposed by the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that these specific embodiments are only used to explain the present invention and do not limit its scope. In addition, the technical features involved in each embodiment described in the text can be combined with each other as long as they do not conflict with each other.

[0059] When describing the present invention, it should be particularly noted that, unless otherwise expressly provided or limited, when a component is mentioned to be "connected" to another component, it means that it can be directly connected or indirectly connected through an intermediate medium. Terms such as "installation" and "fixation" should be broadly understood. For example, they can refer to fixed connection, detachable connection or integral connection, mechanical connection or electrical connection, direct connection or indirect connection through an intermediary, and even include the internal connection relationship or interaction between the two components. For ordinary technicians in this field, the specific meanings of these terms in the present invention can be understood according to the specific circumstances.

[0060] In addition, although the steps in the present invention are arranged with numbers, they are not used to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a certain step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used in this article involves and covers any and all possible combinations of one or more of the associated listed items.

[0061] like Figure 1-3 As shown, this embodiment provides a heavy-duty industrial robot adaptive compliant large curved surface fitting system, such as Figure 1 Specifically shown include:

[0062] AGV trolley 1, control cabinet 2, robot body 3, robot bearing connection column 4, robot terminal connection flange 5, terminal assembly execution component 6, assembly component 7 and component assembly fixing component 8.

[0063] The AGV trolley 1 is a trolley with a large range of motion and a high load capacity, which is used to receive external position movement instructions and move to a specified position point; the upper front end thereof carries the robot body 3 through the robot carrying connecting column 4, and the upper rear end directly carries the control cabinet 2;

[0064] The control cabinet 2 and the robot body 3 are a heavy-duty industrial robot system with a large range of motion and multiple degrees of freedom. There is a communication connection between the two, and the control cabinet 2 is used to control the robot body 3 to correct the joint posture. The end of the robot body 3 is connected to the end assembly actuator 6 through the robot end connection flange 5, and the robot body 3 moves the end assembly actuator 6 to the specified assembly posture point according to the control instruction of the control cabinet 2.

[0065] The terminal assembly actuator 6 absorbs the assembly component 7 through the suction cup assembly 66 located on its lower side; the component assembly fixing component 8 is fixed at a specified assembly posture point, and its upper side is the surface component 81 to be assembled; when the terminal assembly actuator 6 moves to the specified assembly posture point, the assembly component 7 and the surface component 81 to be assembled achieve curved surface fitting.

[0066] The assembly component 7 selects different assembly objects according to the assembly task. The assembly component 7 is adsorbed and fixed by the suction cup assembly 66 of the terminal assembly actuator 6, and reaches the assembly posture point through the movement of the robot body 3, and is surface-fitted with the assembly surface 811 of the component assembly fixing component 8 to complete the specified assembly task.

[0067] More specifically, Figure 2 As shown, the terminal assembly execution component 6 also includes:

[0068] A force sensor 61, a force sensor connecting flange 62, an inertial sensor 63, an inertial sensor connecting plate 64, and an actuator body 65; the actuator body 65 is located in the middle of the terminal assembly actuator 6, and a force sensor 65 is installed above it through the force sensor connecting flange 62, and an inertial sensor 63 is installed on its side through the inertial sensor connecting plate 64; the suction cup assembly 66 includes five pairs of ten suction cups, and is symmetrically distributed and installed on both sides below the actuator body 65.

[0069] In this embodiment, the force sensor 61 is a six-dimensional force sensor. The force sensor 61 is connected to the robot body 3 through the robot end connection flange 5, and is connected to the actuator body 65 through the force sensor connection flange 62. It receives the three-axis force and three-axis torque of the components below the force sensor 61, and simultaneously performs real-time data transmission and communication, and transmits the data to the host computer, which performs control algorithm feedback.

[0070] The inertial sensor 63 is an acceleration and angular velocity sensor, which is connected to the actuator body 65 via the inertial sensor connection plate 64. The inertial sensor 63 is installed at a specified position of the actuator body 65, receives the three-axis acceleration and three-axis angular velocity of the terminal assembly actuator 6 at the specified position, and simultaneously performs real-time data transmission and communication, and transmits the data to the host computer, which performs control algorithm feedback.

[0071] The actuator body 65 is a supporting and fixing structure for the terminal assembly actuator 6. The actuator body 65 is connected to the force sensor 61 through the force sensor connecting flange 62, and is connected to the inertial sensor 63 through the inertial sensor connecting plate 64. It mainly plays the role of supporting and fixing the terminal assembly actuator 6, wherein five pairs of suction cup assemblies 66 are distributed in an array and fixed on the actuator body 65.

[0072] The suction cup assembly 66 is a vacuum suction cup assembly, and five pairs of suction cup assemblies 66 are arranged in an array and fixed on the actuator body 65. Since the suction cup assembly 66 is a flexible suction cup and is distributed in five pairs of ten suction cups, it can adsorb the corresponding large curved surface parts. The number and layout of the suction cups can be appropriately adjusted according to the assembly conditions of the parts 7 to complete the specified assembly task.

[0073] More specifically, Figure 3 As shown, the component assembly fixing member 8 includes:

[0074] The surface assembly to be assembled 81, the rotating assembly 82, and the positioning support assembly 83;

[0075] The positioning support assembly 83 includes: a frame support assembly 831 for supporting and fixing the surface assembly 81 to be assembled and the rotating assembly 82, four adjustable base support assemblies 832 installed at the four corners of the bottom of the frame support assembly, and four non-adjustable base support assemblies 833 installed and symmetrically distributed at the bottom of both sides of the frame support assembly; the surface assembly 81 to be assembled and the rotating assembly 82 are supported and fixed by the frame support assembly 831, and the assembly position is moved as needed to complete the specified assembly task.

[0076] The rotating assembly 82 is movably mounted on the upper ends of both sides of the frame support assembly 831 to adjust the angle of the assembly 81 to be assembled as needed;

[0077] The assembly surface assembly includes an assembly surface frame assembly 812 and an assembly surface 811 mounted thereon; one side of the lower side of the assembly surface frame assembly 812 is fixedly connected to the frame support assembly 831, and the other side is fixedly connected to the rotating assembly 82. The assembly surface 811 matching the assembly component 7 can be replaced as needed to meet different assembly requirements and complete the specified assembly task.

[0078] like Figure 4 As shown, this embodiment also provides a control method for the adaptive and compliant large curved surface bonding system of a heavy-duty industrial robot, wherein the AGV trolley 1 carries the equipment and the assembly component 7 to the specified assembly position of the terminal assembly actuator 6, and the force sensor 61 and the inertial sensor 63 perform real-time data transmission and communication, and transmit the data to the host computer, which performs control algorithm feedback, controls the movement of the robot body 3, drives the terminal assembly actuator 6 and the assembly component 7 to move, that is, continuously adjusts the posture, and slowly approaches the surface to be assembled 811 until the surface bonding process is completed. This control method specifically includes the following steps:

[0079] S1, based on the force sensor 61, receiving the three-axis force and three-axis moment of the components below the force sensor 61; and identifying the gravity center of gravity parameters of the components below the force sensor 61, so as to compensate the components below the force sensor 61 in real time, and obtain the actual three-axis force and three-axis moment generated by the external force on the components below the force sensor 61; the components below the force sensor 61 include: a force sensor connecting flange 62, an inertial sensor 63, an inertial sensor connecting plate 64, an actuator body 65, a suction cup assembly 66, and an assembly component 7;

[0080] As a preferred implementation of step S1, S1 specifically includes:

[0081] S11. Construct world coordinate system O w -X w Y w Z w , the robot body 3 coordinate system is O r -X r Y r Z r , robot end connection flange 5 coordinate system O f -X f Y f Z f And the force sensor 61 coordinate system is O s -X s Y s Z s ; Set the Z of the world coordinate system w The axis is parallel to and opposite to the direction of gravity, and the attitude transformation matrix of the force sensor 61 coordinate system relative to the world coordinate system is constructed. The formula is:

[0082]

[0083] in, is the posture transformation matrix of the robot body 3 coordinate system relative to the world coordinate system; The attitude transformation matrix of the robot end connection flange 5 coordinate system relative to the robot body 3 coordinate system; is the posture transformation matrix of the force sensor 61 coordinate system relative to the robot end connection flange 5 coordinate system;

[0084] S12, determine the position point of the component assembly fixing member 8, the robot body 3 moves near the position point, and stops at six position points, and obtains the coordinates of the robot body 3 at the six position points in the Cartesian coordinate system: (X, Y, Z, A, B, C), where X, Y, Z are the positions of the robot in the Cartesian coordinate system; A, B, C are the postures of the robot in the Cartesian coordinate system; receive in real time through the inertial sensor 63, and after Kalman filtering, obtain the three-axis force and three-axis torque received by the filtered force sensor 61 at the corresponding coordinates;

[0085] S13, according to the coordinate system transformation relationship in step S11 and the three-axis force and three-axis moment obtained in step S12, the gravity G and the center of gravity (G x , G y , G z ), the roll angle φ and pitch angle w of the robot body 3 base, and the zero drift value (F x 、F y 、F z 、M x 、M y 、M z );

[0086] Among them, G x , G y , G z F is the X, Y, and Z components of the gravity center of the components below the force sensor 61 in the Cartesian coordinate system; x 、F y 、F z M is the zero drift triaxial force of the components below the force sensor 61; x 、M y 、M z is the zero drift three-axis moment of the components below the force sensor 61;

[0087] In this embodiment, all the parameters calculated above will be used to calculate the actual force and moment generated by the external force on the components below the force sensor 61;

[0088] S14, according to the parameters calculated in step S13, combined with the coordinates of the robot body 3 in the Cartesian coordinate system obtained in real time, and the three-axis force and three-axis moment received by the force sensor 61 after filtering at the corresponding coordinates, the actual three-axis force (f x 、f y 、f z ) and actual triaxial moment (t x ,t y ,t z ).

[0089] S2, receiving the three-axis acceleration and three-axis angular velocity of the terminal assembly actuator 6 at the specified assembly posture point based on the inertial sensor 63; performing zero drift parameter identification on the inertial sensor 63, thereby performing real-time compensation on the inertial sensor 63, and obtaining the actual three-axis acceleration and three-axis angular velocity;

[0090] As a preferred implementation of step S2, S2 specifically includes:

[0091] S21. Construct world coordinate system O w -X w Y w Z w , the robot body 3 coordinate system is O r -X r Y r Z r , robot end connection flange 5 coordinate system O f -X f Y f Z f and the inertial sensor 63 coordinate system O g -X g Y g Z g , set the Z of the world coordinate system w The axis is parallel to and opposite to the direction of gravity, and the attitude transformation matrix of the inertial sensor 63 coordinate system relative to the world coordinate system is constructed. The formula is:

[0092]

[0093] in, is the posture transformation matrix of the robot body 3 coordinate system relative to the world coordinate system; The attitude transformation matrix of the robot end connection flange 5 coordinate system relative to the robot body 3 coordinate system; is the attitude conversion matrix of the inertial sensor 63 coordinate system relative to the robot end connection flange 5 coordinate system;

[0094] S22, determine the position point of the component assembly fixing member 8, the robot body 3 moves near the position point, and stops at six position points, and obtains the coordinates of the robot body 3 of the six position points in the Cartesian coordinate system in real time: (X, Y, Z, A, B, C); receive them in real time through the inertial sensor 63, and obtain the three-axis acceleration and three-axis angular velocity received by the inertial sensor 63 after filtering at the corresponding coordinates through Kalman filtering;

[0095] S23, based on the coordinate system conversion relationship constructed in step S21 and the three-axis acceleration and three-axis angular velocity obtained in step S22, the calibration coefficient (K) of the inertial sensor 63 is obtained through the inertial compensation algorithm. x , K y , K z , K a , K b , K c ) and zero drift value (A x , A y , A z , W a , W b , W c ), where K x , K y , K z K is the calibration coefficient of the three-axis acceleration received by the inertial sensor 63; a , K b , K c A is the calibration coefficient of the three-axis angular velocity received by the inertial sensor 63; x , A y , A z is the zero drift value of the three-axis acceleration received by the inertial sensor 63; W a , W b , W c is the zero drift value of the three-axis angular velocity received by the inertial sensor 63;

[0096] In this embodiment, all the parameters calculated above will be used to calculate the actual acceleration and angular velocity of the terminal assembly actuator 6;

[0097] S24, according to the calibration coefficient and zero drift value obtained in step S23, the inertial sensor 63 is compensated; combined with the coordinates of the robot body 3 in the Cartesian coordinate system obtained in real time in step S22, and the three-axis acceleration and three-axis angular velocity received by the inertial sensor 63 after filtering at the corresponding coordinates, the actual acceleration (a) of the terminal assembly actuator 6 is calculated in real time by the zero drift removal algorithm. x 、a y 、a z ) and the actual three-axis angular velocity (w a 、w b、w c ), and then further calculate the actual three-axis angular acceleration (a a 、a b 、a c ).

[0098] S3, real-time monitoring of the actual three-axis force, three-axis torque, three-axis acceleration and three-axis angular velocity after compensation, and continuous correction of the current end position of the robot body 3 through feedback from the control algorithm until the assembly component 7 and the surface to be assembled 811 achieve the expected surface fitting effect.

[0099] As a preferred implementation of step S3, step S3 specifically includes:

[0100] S31, according to the actual three-axis force, three-axis torque, three-axis acceleration and three-axis angular velocity obtained in step S1 and step S2, design an admittance control model as the control algorithm of the assembly task. The formula of the admittance control model when the robot body 3 performs one-dimensional translation motion is expressed as:

[0101]

[0102] Among them, m is the virtual mass; b is the virtual damping; k is the virtual stiffness; is the acceleration of the robot body 3 end; is the speed of the robot body 3 end; x is the position of the robot body 3 end; f is the component below the force sensor 61, that is, the actual force on the robot body 3 end, is the expected acceleration of the robot body 3 end; is the desired speed of the robot body 3 end; x n is the desired position of the end of the robot body 3; f n is the desired force at the end of the robot body 3; a is the end assembly actuator 6, that is, the actual acceleration at the end of the robot body 3; x n 、f n Set to zero value, f n Setting it to zero will make the assembly process smoother, x n Set to zero, by calculating x and x n The difference is used to determine the assembly task direction and assembly speed; the positive or negative difference reflects the direction, and the size of the difference reflects the speed;

[0103] In this embodiment, because the components below the force sensor 61 are mounted on the end of the robot body 3, the force and torque they are subjected to are considered to be the same; and in actual movement, the acceleration is relatively small, so it is also considered that the acceleration, angular velocity, and angular acceleration of the end assembly actuator 6 are also the same as those of the end of the robot body 3.

[0104] In addition, in this embodiment, setting the parameters to zero values ​​is to ensure that the expected acceleration, expected speed, and expected force are all 0 during and after the veneer process. This is because it is not desired to generate unnecessary vibrations at the end of the robot during and after the veneer process, so that the veneer effect is smooth and smooth, and the expected position may not be set to 0; in addition, the parameters of the end of the robot body 3, except for the actual acceleration of the end of the robot body 3 and the actual force of the end of the robot body 3, which are the parameters obtained in step S1 and step S2, are all formula convergence parameters;

[0105] S32. When adjusting the posture, in order to ensure the assembly effect and improve the assembly efficiency, when a collision occurs, the motion rigidity is reduced and the vibration amplitude is reduced; when rebounding, the motion rigidity is increased and the position adjustment time and space are increased, that is, the parameter k is classified and set, and the classification formula is as follows:

[0106]

[0107] Among them, k n is the effective virtual stiffness; λ1 is the virtual stiffness adjustment parameter 1; λ2 is the virtual stiffness adjustment parameter 2; ρ is the acceleration adjustment parameter;

[0108] S33. Design the admittance control model for one-dimensional rotational motion, the formula is expressed as:

[0109]

[0110] Among them, α is the virtual moment of inertia; β is the virtual rotation damping; ψ is the virtual rotation stiffness; is the angular acceleration of the robot body 3 end; is the angular velocity of the end of the robot body 3; θ is the rotation angle of the end of the robot body 3; t is the actual torque applied to the components below the force sensor 61, that is, the end of the robot body 3; is the expected angular acceleration of the robot body 3 end; is the desired angular velocity of the robot body 3 end; θ n is the desired rotation angle of the robot body 3 end; t n is the desired torque at the end of the robot body 3; a θ The end assembly actuator 6, that is, the actual angular acceleration of the end of the robot body 3; θ n ,t n Set to zero, by calculating θ and θ n The difference is used to determine the assembly task direction and assembly speed; the positive or negative difference reflects the direction, and the size of the difference reflects the speed;

[0111] S34, continuously adjust the posture and monitor in real time, when the coordinates (X, Y, Z, A, B, C) of the robot body 3 in the Cartesian coordinate system are monitored to be within a certain error fluctuation range within a certain period of Δt, and the actual three-axis force (f x 、f y 、f z ) and triaxial moment (t x ,t y ,t z ) in the corresponding assembly position, corresponding to f z The parameters are maintained within a fixed fluctuation range, and the rest of the f x 、f y ,t x ,t y ,t z The parameter fluctuates around zero; the actual acceleration (a x 、a y 、a z ) and the actual angular acceleration (a a 、a b 、a c ) maintains fluctuations around zero, the assembly task is considered completed.

[0112] In summary, the adaptive and flexible large-surface fitting system for heavy-duty industrial robots and the control method thereof proposed in the present invention can solve the problem of low absolute positioning accuracy of industrial robots and meet the assembly requirements of high-precision large-scale complex structural components; at the same time, it takes into account the flexible assembly of customized components and improves assembly efficiency.

[0113] In this specification, the terms "one embodiment", "some embodiments", "examples", "specific examples" or "some examples" refer to at least one embodiment or example described in combination with specific features, structures, materials or characteristics. These described specific features, structures, materials or characteristics can be combined in an appropriate manner in one or more embodiments or examples. In addition, the technician can combine and combine different embodiments or examples and their features described in this specification without contradiction.

[0114] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary rather than restrictive from any point of view, and the scope of the present invention is defined by the appended claims rather than the above description. It is therefore intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A control method for a heavy-duty industrial robot's adaptive, compliant, large-surface fitting system, characterized in that: The specific steps include: S1, based on the force sensor (61), receiving the three-axis force and three-axis moment exerted on the components below the force sensor (61); and identifying the gravity center of gravity parameters of the components below the force sensor (61), thereby performing real-time compensation on the components below the force sensor (61), and obtaining the actual three-axis force and three-axis moment exerted on the components below the force sensor (61) by the external force; the components below the force sensor (61) include: a force sensor connecting flange (62), an inertial sensor (63), an inertial sensor connecting plate (64), an actuator body (65), a suction cup assembly (66), and an assembly component (7); S2, receiving the three-axis acceleration and three-axis angular velocity of the terminal assembly execution component (6) at the specified assembly posture point based on the inertial sensor (63); performing zero drift parameter identification on the inertial sensor (63), thereby performing real-time compensation on the inertial sensor (63) to obtain actual three-axis acceleration and three-axis angular velocity; S3, real-time monitoring of the actual three-axis force, three-axis torque, three-axis acceleration and three-axis angular velocity after compensation, and continuously correcting the current end position of the robot body (3) through control algorithm feedback until the assembly component (7) and the surface to be assembled (811) achieve the expected surface fitting effect.

2. The control method of a heavy-duty industrial robot adaptive compliant large curved surface fitting system according to claim 1, characterized in that: Step S1 specifically includes: S11. Construct world coordinate system O w -X w Y w Z w , the robot body (3) coordinate system is O r -X r Y r Z r 、Robot end connection flange (5) coordinate system O f -X f Y f Z f And the force sensor (61) coordinate system is O s -X s Y s Z s ; Set the Z of the world coordinate system w The axis is parallel to and opposite to the direction of gravity, and the attitude transformation matrix of the force sensor (61) coordinate system relative to the world coordinate system is constructed. The formula is: in, is the posture transformation matrix of the robot body (3) coordinate system relative to the world coordinate system; The attitude conversion matrix of the robot end connection flange (5) coordinate system relative to the robot body (3) coordinate system; is a posture conversion matrix of the force sensor (61) coordinate system relative to the robot end connection flange (5) coordinate system; S12, determine the position point of the component assembly fixing member (8), the robot body (3) moves near the position point, and stops at six position points, and obtains the coordinates of the robot body (3) at the six position points in the Cartesian coordinate system: (X, Y, Z, A, B, C), where X, Y, Z are the positions of the robot in the Cartesian coordinate system; A, B, C are the postures of the robot in the Cartesian coordinate system; receive in real time through the inertial sensor (63), and after Kalman filtering, obtain the three-axis force and three-axis torque received by the filtered force sensor (61) at the corresponding coordinates; S13, according to the coordinate system transformation relationship in step S11 and the three-axis force and three-axis moment obtained in step S12, the gravity G and the center of gravity (G x , G y , G z ), the roll angle φ and pitch angle w of the base of the robot body (3), and the zero drift value (F x 、F y 、F z 、M x 、M y 、M z ); Among them, G x , G y , G z F is the X, Y, and Z components of the gravity center of the components below the force sensor (61) in the Cartesian coordinate system; x 、F y 、F z M is the zero-drift triaxial force of the components below the force sensor (61); x 、M y 、M z is the zero drift three-axis moment of the components below the force sensor (61); S14, based on the parameters calculated in step S13, combined with the coordinates of the robot body (3) in the Cartesian coordinate system acquired in real time, and the three-axis forces and three-axis moments received by the force sensor (61) after filtering at the corresponding coordinates, the actual three-axis forces (f) generated by the external force on the components below the force sensor (61) are calculated by a zero drift removal algorithm. x 、f y 、f z ) and actual triaxial moment (t x ,t y ,t z ).

3. The control method of a heavy-duty industrial robot adaptive compliant large-surface fitting system according to claim 1, characterized in that: Step S2 specifically includes: S21. Construct world coordinate system O w -X w Y w Z w , the robot body (3) coordinate system is O r -X r Y r Z r 、Robot end connection flange (5) coordinate system O f -X f Y f Z f and the inertial sensor (63) coordinate system O g -X g Y g Z g , set the Z of the world coordinate system w The axis is parallel to and opposite to the direction of gravity, and the attitude transformation matrix of the inertial sensor (63) coordinate system relative to the world coordinate system is constructed. The formula is: in, is the posture transformation matrix of the robot body (3) coordinate system relative to the world coordinate system; The attitude conversion matrix of the robot end connection flange (5) coordinate system relative to the robot body (3) coordinate system; is the attitude conversion matrix of the inertial sensor (63) coordinate system relative to the robot end connection flange (5) coordinate system; S22, determining the position point of the component assembly fixing member (8), the robot body (3) moves near the position point, and stops at six position points, and obtains the coordinates of the robot body (3) at the six position points in the Cartesian coordinate system in real time: (X, Y, Z, A, B, C); receiving in real time through the inertial sensor (63), and through Kalman filtering, obtaining the three-axis acceleration and three-axis angular velocity received by the inertial sensor (63) after filtering at the corresponding coordinates; S23, based on the coordinate system conversion relationship constructed in step S21 and the three-axis acceleration and three-axis angular velocity obtained in step S22, the calibration coefficient (K) of the inertial sensor (63) is obtained through an inertial compensation algorithm. x , K y , K z , K a , K b , K c ) and zero drift value (A x , A y , A z , W a , W b , W c ), where K x , K y , K z K is the calibration coefficient of the three-axis acceleration received by the inertial sensor (63); a , K b , K c A is the calibration coefficient of the three-axis angular velocity received by the inertial sensor (63); x , A y , A z is the zero drift value of the three-axis acceleration received by the inertial sensor (63); W a , W b , W c is the zero drift value of the three-axis angular velocity received by the inertial sensor (63); S24, compensating the inertial sensor (63) according to the calibration coefficient and zero drift value obtained in step S23; combining the coordinates of the robot body (3) in the Cartesian coordinate system obtained in real time in step S22, and the three-axis acceleration and three-axis angular velocity received by the inertial sensor (63) after filtering at the corresponding coordinates, and calculating the actual acceleration (a) of the terminal assembly actuator (6) in real time through a zero drift removal algorithm. x 、a y 、a z ) and the actual three-axis angular velocity (w a 、w b 、w c ), and then further calculate the actual three-axis angular acceleration (a a 、a b 、a c ).

4. The control method of a heavy-load industrial robot adaptive compliant large curved surface fitting system according to claim 1, characterized in that: Step S3 specifically includes: S31. According to the actual three-axis forces, three-axis torques, three-axis accelerations and three-axis angular velocities obtained in step S1 and step S2, an admittance control model is designed as a control algorithm for the assembly task. The formula of the admittance control model when the robot body (3) performs one-dimensional translation motion is expressed as follows: Among them, m is the virtual mass; b is the virtual damping; k is the virtual stiffness; is the acceleration of the robot body (3); is the speed of the end of the robot body (3); x is the position of the end of the robot body (3); and f is the component below the force sensor (61), that is, the actual force exerted on the end of the robot body (3); is the expected acceleration of the end of the robot body (3); is the desired speed of the robot body (3) at the end; x n is the desired position of the end of the robot body (3); n is the desired force at the end of the robot body (3); a is the end assembly actuator (6), that is, the actual acceleration at the end of the robot body (3); x n 、f n Set to zero, by calculating x and x n The difference between the two is used to determine the assembly task direction and assembly speed; S32. When adjusting the posture, in order to ensure the assembly effect and improve the assembly efficiency, when a collision occurs, the motion rigidity is reduced and the vibration amplitude is reduced; when rebounding, the motion rigidity is increased and the position adjustment time and space are increased, that is, the parameter k is classified and set, and the classification formula is as follows: Among them, k n is the effective virtual stiffness; λ1 is the virtual stiffness adjustment parameter 1; λ2 is the virtual stiffness adjustment parameter 2; ρ is the acceleration adjustment parameter; S33. Design the admittance control model for one-dimensional rotational motion, the formula is expressed as: Among them, α is the virtual moment of inertia; β is the virtual rotation damping; ψ is the virtual rotation stiffness; is the angular acceleration of the robot body (3) at the end; is the angular velocity of the end of the robot body (3); θ is the rotation angle of the end of the robot body (3); t is the actual torque applied to the components below the force sensor (61), that is, the end of the robot body (3); is the expected angular acceleration of the end of the robot body (3); is the desired angular velocity of the robot body (3) end; θ n is the desired rotation angle of the end of the robot body (3); t n is the desired torque at the end of the robot body (3); θ The actual angular acceleration of the end of the robot body (3) is the end assembly actuator (6); θ n ,t n Set to zero, by calculating θ and θ n The difference between the two is used to determine the assembly task direction and assembly speed; S34, continuously adjust the posture and monitor in real time, when the coordinates (X, Y, Z, A, B, C) of the robot body (3) in the Cartesian coordinate system are monitored to be maintained within a certain error fluctuation range within a certain period of time Δt, and the actual three-axis force (f x 、f y 、f z ) and triaxial moment (t x ,t y ,t z ) in the corresponding assembly position, corresponding to f z The parameters are maintained within a fixed fluctuation range, and the rest of the f x 、f y ,t x ,t y ,t z The parameter fluctuates around zero; the actual acceleration (a x 、a y 、a z ) and the actual angular acceleration (a a 、a b 、a c ) maintains fluctuations around zero, the assembly task is considered completed.

5. The application system of the control method according to any one of claims 1 to 4, characterized in that: Specifically include: AGV trolley (1), control cabinet (2), robot body (3), robot bearing connection column (4), robot terminal connection flange (5), terminal assembly execution component (6), assembly component (7) and component assembly fixing component (8); The AGV vehicle (1) is used to receive an external position movement instruction and move to a designated position point; The upper front end thereof carries the robot body (3) via a robot carrying connecting column (4), and the upper rear end directly carries the control cabinet (2); There is a communication connection between the control cabinet (2) and the robot body (3) for controlling the robot body (3) to correct the joint posture; the end of the robot body (3) is connected to the end assembly execution component (6) through the robot end connection flange (5), and the robot body (3) moves the end assembly execution component (6) to a specified assembly posture point according to the control instruction of the control cabinet (2); The terminal assembly execution component (6) absorbs the assembly component (7) through the suction cup component (66) located on the lower side thereof; the component assembly fixing component (8) is fixed at a specified assembly posture point, and the upper side thereof is a surface component (81) to be assembled; when the terminal assembly execution component (6) moves to the specified assembly posture point, the assembly component (7) and the surface component (81) to be assembled achieve curved surface fitting.

6. The application system according to claim 5, characterized in that: The terminal assembly execution component (6) further comprises: A force sensor (61), a force sensor connecting flange (62), an inertial sensor (63), an inertial sensor connecting plate (64), an actuator body (65), and a suction cup assembly (66); the actuator body (65) is located in the middle of the terminal assembly actuator (6), the force sensor (61) is installed on the upper side thereof via the force sensor connecting flange (62), and the inertial sensor (63) is installed on the side thereof via the inertial sensor connecting plate (64); the suction cup assembly (66) includes five pairs of ten suction cups in total, and is symmetrically distributed and installed on both sides below the actuator body (65).

7. The application system according to claim 5, characterized in that: The component assembly fixing member (8) comprises: A surface assembly to be assembled (81), a rotating assembly (82), and a positioning support assembly (83); The positioning support assembly (83) comprises: a frame support assembly (831) for supporting and fixing the surface assembly (81) to be assembled and the rotating assembly (82), four adjustable base support assemblies (832) installed at four corners of the bottom of the frame support assembly, and four non-adjustable base support assemblies (833) installed at the bottom of both sides of the frame support assembly and distributed symmetrically; The rotating assembly (82) is movably mounted on the upper ends of both sides of the frame supporting assembly (831); The assembly surface assembly comprises an assembly surface frame assembly (812) and an assembly surface (811) mounted thereon; one side below the assembly surface frame assembly (812) is fixedly connected to the frame support assembly (831), and the other side below the assembly surface frame assembly (812) is fixedly connected to the rotating assembly (82).

Images