Compliant assembly method and system based on admittance control and ideal trajectory fitting

Abstract

The application provides a compliant assembly method and system based on admittance control and ideal trajectory fitting, which comprises the following steps: after completing the preliminary work, determining the series stiffness of the assembly object containing the reference part and the part to be assembled; after moving the part to be assembled to the docking preparation position through human-computer cooperation, taking the docking preparation position as the starting point of the trajectory, determining the assembly reference trajectory; controlling the robot to drive the part to be assembled to feed according to the assembly reference trajectory; in the feeding process, determining the target position of the trajectory point after the force-position relationship coordination through the admittance controller, and preliminarily correcting the assembly reference trajectory according to the target position; fitting to generate an ideal trajectory, and iteratively correcting the subsequent feeding process according to the ideal trajectory; until the feeding termination condition is met, the robot stops the assembly feeding. The compliant assembly process can be optimized on the basis of admittance control, the ideal trajectory can be solved in real time and actively corrected, so that the robot can more efficiently and high-quality realize the compliant assembly process.

Description

Technical Field

[0001] This invention relates to the field of assembly manufacturing technology, and in particular to a compliant assembly method and system based on admittance control and ideal trajectory fitting. Background Technology

[0002] Compliant assembly refers to an assembly technique that adapts to contact forces during the assembly process, adjusting the assembly trajectory to avoid abnormal conditions such as large contact forces or jamming, allowing the work to continue. For the docking process of single-section components, it mainly has the following characteristics:

[0003] 1) Due to working conditions, the reference part is placed on a movable frame. After the frame is manually pushed to the docking position, the wheels are locked to complete the positioning. The target pose cannot be accurately known.

[0004] 2) The main fitting feature is pin hole docking, which is relatively tight. If the positional deviation is large, docking is not possible.

[0005] 3) Due to limitations of space and process, it is not possible to measure the position of the reference part before assembly;

[0006] 4) In order to protect the product, the contact force and torque during the assembly and docking process must be controlled within a certain range and should not be too large.

[0007] Under the above constraints, the assembly process cannot be automated using ordinary robots. It typically requires manual assembly, necessitating at least four people working simultaneously, which is time-consuming, labor-intensive, and inefficient. Because the vehicle can only be positioned within a general range each time, random positional deviations exist, resulting in different trajectories and final positions for each assembly. Furthermore, the lack of measurement capabilities makes it difficult to guarantee assembly accuracy. While traditional compliant assembly methods such as impedance / admittance control and force / position hybrid control can complete the assembly process, the absence of an ideal reference assembly trajectory leads to prolonged assembly time and higher contact stress due to this passive, compromise-based approach. Summary of the Invention

[0008] This invention provides a compliant assembly method and system based on admittance control and ideal trajectory fitting, which solves the problems of low assembly efficiency and low assembly accuracy of traditional assembly schemes.

[0009] On one hand, the present invention provides a compliant assembly method based on admittance control and ideal trajectory fitting, comprising:

[0010] After moving the reference component to the assembly docking station and installing the component to be assembled on the robot end effector, the series stiffness of the assembly object including the reference component and the component to be assembled is determined.

[0011] After the part to be assembled is moved to the docking preparation position through human-machine collaboration, the docking preparation position is used as the trajectory starting point. Based on the pose information of the trajectory starting point, the expected assembly time and the assembly feed distance, the assembly reference trajectory is determined.

[0012] The robot is controlled to drive the part to be assembled to feed along the assembly reference trajectory;

[0013] During the feeding process, based on the robot's pose data and the contact force data collected by the six-dimensional force sensor on the robot, the target position of the trajectory point after force-position relationship coordination is determined by the admittance controller, and the assembly reference trajectory is initially corrected based on the target position.

[0014] Based on the actual position of historical trajectory points during the feeding process, the contact force data, and the series stiffness, an ideal trajectory is fitted and generated, and the subsequent feeding process is iteratively corrected based on the ideal trajectory.

[0015] The robot is controlled to stop assembly feeding until the feed termination condition is met.

[0016] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, the pose information of the trajectory starting point is obtained through the following process:

[0017] Based on the structure of the component to be assembled, establish a robot tool coordinate system;

[0018] Based on the robot tool coordinate system, determine the measurement coordinate system corresponding to the six-dimensional force sensor on the robot;

[0019] Based on the robot tool coordinate system and the measurement coordinate system, the workpiece to be loaded is pulled to the docking preparation position, with the docking preparation position as the starting point of the trajectory.

[0020] Determine the position coordinates of the trajectory starting point in the robot's base coordinate system;

[0021] Using the positive direction of the coordinate axis corresponding to the assembly feed direction in the robot tool coordinate system as the trajectory extension direction, the direction vector of the trajectory starting point in the base coordinate system is determined.

[0022] The position coordinates and the direction vector are used as the pose information of the starting point of the trajectory.

[0023] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, determining the measurement coordinate system corresponding to the six-dimensional force sensor on the robot based on the robot tool coordinate system includes:

[0024] In the robot tool coordinate system, the initial position and initial orientation of the measurement origin of the six-dimensional force sensor are calibrated to obtain the calibration results;

[0025] Based on the calibration results, the initial orientation of the coordinate axes and the position of the measurement origin of the original coordinate system corresponding to the six-dimensional force sensor are adjusted to obtain a measurement coordinate system that matches the robot tool coordinate system.

[0026] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, determining the assembly reference trajectory based on the pose information of the trajectory starting point, the expected assembly time, and the assembly feed distance includes:

[0027] The number of trajectory points is determined based on the expected assembly time and the preset control cycle;

[0028] The single-step feed distance is determined based on the assembly feed distance and the number of trajectory points;

[0029] Based on the single-step feed distance and the direction vector of the trajectory starting point, the increment of each trajectory point in each direction is determined;

[0030] Based on the position coordinates of the starting point of the trajectory and the increments of each trajectory point in each direction, the position coordinates of each trajectory point are determined.

[0031] The position coordinates of all trajectory points are combined to obtain the assembly reference trajectory.

[0032] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, determining the series stiffness of the assembly object including the reference member and the part to be assembled includes:

[0033] Using the reference component and the component to be assembled as assembly objects, a series stiffness model of the assembly objects is constructed.

[0034] During the process of direct or indirect contact between the reference component and the component to be installed, the first deformation at the end of the component to be installed, the second deformation at the end of the reference component, the deformation correlation amount, and the elastic force value are measured in the series stiffness model.

[0035] The series stiffness is calculated based on the first deformation, the second deformation, the deformation correlation amount, and the elastic force value.

[0036] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, the step of determining the target position of the trajectory point after force-position relationship coordination by an admittance controller based on the robot's pose data and contact force data collected by the six-dimensional force sensor on the robot includes:

[0037] Based on the admittance control principle and the robot's pose data, a correspondence is established between the contact force data and the deviation data of the trajectory points; wherein, the deviation data includes the positional deviation between the actual position and the target position of the trajectory point, the velocity deviation between the actual velocity and the target velocity, and the acceleration deviation between the actual acceleration and the target acceleration.

[0038] The correspondence between the contact force data and the deviation data of the trajectory points is discretized by the admittance controller, and then integrated to obtain the target position of the trajectory points after the force-position relationship is coordinated.

[0039] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, the step of fitting and generating an ideal trajectory based on the true position of historical trajectory points during the feeding process, the contact force data, and the series stiffness includes:

[0040] For each historical trajectory point, the contact force data is multiplied by the reciprocal of the series stiffness to obtain intermediate parameters;

[0041] The ideal position of the historical trajectory point is determined by subtracting the intermediate parameter from the actual position of the historical trajectory point, thus obtaining the ideal trajectory point.

[0042] The ideal trajectory is generated by fitting all the ideal trajectory points within the target segment.

[0043] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, after controlling the robot to stop the assembly feed, the method further includes:

[0044] The robot is controlled to move the parts to be installed to contact the reference parts in a preset direction and cause deformation.

[0045] The contact force data collected by the six-dimensional force sensor during the contact and deformation process is acquired;

[0046] Based on the contact force data, the assembly gap between the part to be assembled and the reference part is adjusted.

[0047] According to the compliant assembly method based on admittance control and ideal trajectory fitting provided by the present invention, the step of adjusting the assembly gap between the part to be assembled and the reference part according to the contact force data includes:

[0048] Determine whether the contact force data reaches the first preset range;

[0049] When it is determined that the contact force data has reached the first set range, the position coordinates of the first key point are determined;

[0050] Determine whether the contact force data reaches the second preset range;

[0051] When the contact force data is determined to reach the second preset range, the position coordinates of the second key point are determined;

[0052] Based on the position coordinates of the first key point and the position coordinates of the second key point, determine the midpoint of the line connecting the first key point and the second key point;

[0053] The robot is controlled to move the component to be installed to the position of the midpoint of the connecting line.

[0054] On the other hand, the present invention also provides a compliant assembly system based on admittance control and ideal trajectory fitting, comprising: a control device and a robot equipped with a six-dimensional force sensor;

[0055] The control device includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, it implements any of the above-described compliant assembly methods based on admittance control and ideal trajectory fitting.

[0056] The compliant assembly method and system based on admittance control and ideal trajectory fitting provided by this invention can initially correct the assembly reference trajectory using the target position during the assembly process, and can iteratively correct the subsequent feed process using the fitted ideal trajectory. It can optimize the compliant assembly process based on admittance control, and can calculate the ideal trajectory in real time and actively correct it, so as to improve the assembly efficiency. This allows the robot to more accurately achieve compliant assembly between the reference part and the part to be assembled, making the assembly process more efficient and accurate. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0058] Figure 1 This is one of the flowcharts of the compliant assembly method based on admittance control and ideal trajectory fitting provided in the embodiments of the present invention;

[0059] Figure 2 This is a schematic diagram showing the installation positions of the reference parts and the parts to be installed during the preparation phase of the working condition.

[0060] Figure 3 This is a schematic diagram illustrating the measurement principle of the gasket method;

[0061] Figure 4 This is a schematic diagram illustrating the measurement principle of the contact method;

[0062] Figure 5 This is a schematic diagram illustrating the principle of determining the Y-axis coordinates of an ideal trajectory point;

[0063] Figure 6 This is a flowchart illustrating the process of coordinating contact stress and assembly clearance along the Y-axis.

[0064] Figure 7 This is the second flowchart of the compliant assembly method based on admittance control and ideal trajectory fitting;

[0065] Figure 8 This is a schematic diagram of the compliant assembly system based on admittance control and ideal trajectory fitting provided in an embodiment of the present invention;

[0066] Figure 9 This is a schematic diagram of the control equipment. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0068] This embodiment relates to the field of assembly manufacturing, and can be specifically applied to scenarios where robots are used to assemble objects. In industrial manufacturing, automated robot assembly technology has been widely used. For assembly processes with lower difficulty, such as when the accurate pose of the object to be assembled is known, the assembly features are simple, and there are no obstacles or other special situations, the robot can be directly programmed. The robot can complete the assembly task by executing preset trajectories and actions. This situation is most typical in the automotive industry's production and manufacturing processes.

[0069] Some assembly scenarios are quite complex, and direct assembly may result in abnormal contact between the assembled objects, leading to damage or incomplete assembly. These scenarios include those where the target object lacks accurate positioning, the objects are easily damaged, the assembly process has strict requirements for contact stress, and there are also precision assembly scenarios involving shafts and holes with tight fits. In these cases, robots need to possess a certain degree of compliance to achieve automated assembly. Compliant assembly methods offer advantages such as not relying on measurement steps and controllable contact forces during the process. They perform well in addressing these problems, facilitating the automation of assembly in complex scenarios and improving assembly quality and efficiency.

[0070] Compliant assembly refers to assembly techniques that adapt to contact forces during the assembly process, adjusting the assembly trajectory to avoid abnormal states such as large contact forces or jamming, allowing the work to continue. Currently, there are compliant assembly methods such as passive compliant techniques, active compliant techniques, admittance / impedance control, force-position hybrid control, and compliant assembly techniques based on deep reinforcement learning. Each of these methods has its own advantages and disadvantages, and each has its own main applicable scenarios.

[0071] Passive compliance technology relies on elastic elements to give the robot end effector compliance, but the dimensions and effects of compliance are limited, and the flexible end effector has poor reusability for different objects, requiring targeted design. Active compliance technology usually gives the robot flexibility through force sensors and compliance control algorithms. The most classic one is impedance / admittance control, which usually uses a second-order differential model of "inertia-damping-stiffness" to construct and coordinate the force-position relationship during the robot end effector's motion, thereby achieving compliance with the applied forces. Subsequently, force / position hybrid control emerged. This method decouples the dimensions and defines when to use force control and when to use position control for each dimension according to the requirements. These two are also the most widely used compliance control methods.

[0072] Building upon this, people have further introduced deep learning and reinforcement learning techniques to train models to learn compliant assembly processes, which has also achieved good results. However, the training process often takes a certain amount of time and is not easy to apply quickly in actual working conditions.

[0073] This invention primarily addresses the docking process of a single-section component, and it has the following main characteristics:

[0074] 1) Due to working conditions, the fixed end section is placed on a movable scaffold. After the scaffold is manually pushed to the docking position, the wheels are locked to complete the positioning. The target position cannot be accurately known.

[0075] 2) The main fitting feature is pin hole docking, which is relatively tight. If the positional deviation is large, docking is not possible.

[0076] 3) Due to limitations of space and process, it is not possible to measure the position of the reference end before assembly;

[0077] 4) In order to protect the product, the contact force and torque during the assembly and docking process must be controlled within a certain range and should not be too large.

[0078] Under the above constraints, this process cannot be completed automatically by general robots. For a long time, manual assembly has been used, requiring more than four people to work at the same time, which is time-consuming and labor-intensive.

[0079] Because the vehicle can only be positioned within a general range each time, with random positional deviations, the assembly trajectory and final position are different each time. Furthermore, measurement is not possible. Therefore, obtaining the initial position and the original assembly reference trajectory is the primary problem to be solved. Traditional compliant assembly methods, such as impedance / admittance control and force / position hybrid control, can complete this assembly process, but due to the lack of an ideal reference assembly trajectory, using this passive compromise method leads to longer assembly times and higher contact stress.

[0080] Based on the above-mentioned technical problems, the embodiments of the present invention provide specific solutions, which are described below in conjunction with... Figures 1 to 9 This invention describes the detailed scheme of the compliant assembly method and system based on admittance control and ideal trajectory fitting provided in the embodiments of the present invention.

[0081] Figure 1 This is one of the flowcharts of the compliant assembly method based on admittance control and ideal trajectory fitting provided in the embodiments of the present invention.

[0082] like Figure 1 As shown in the embodiment of the present invention, the compliant assembly method based on admittance control and ideal trajectory fitting can be executed by a control device with data transmission and reception and data processing functions. The method mainly includes the following steps:

[0083] Step 110: After moving the reference part to the assembly docking station and installing the part to be assembled on the robot end effector, determine the series stiffness of the assembly object including the reference part and the part to be assembled.

[0084] In this embodiment, before determining the series stiffness, the process first enters the working condition preparation stage, such as... Figure 2 As shown, during the preparation phase, the frame 220 with the reference part 210 is moved to the assembly docking station and the wheel fixing position is locked. The part to be assembled 230 is connected to the six-dimensional force sensor 260 at the end of the robot 250 through the clamping fixture 240.

[0085] It should be noted that in this embodiment, the part to be assembled, the reference part, the robot, and the fixed end platform of the reference part are taken as the assembly objects, and the series stiffness of the assembly objects is determined.

[0086] Step 120: After moving the part to be assembled to the docking preparation position through human-machine collaboration, take the docking preparation position as the trajectory starting point, and determine the assembly reference trajectory based on the pose information of the trajectory starting point, the expected assembly time, and the assembly feed distance.

[0087] In this embodiment, the operation of moving the part to be assembled to the accurate docking position can be achieved through human-machine collaboration. Specifically, by setting the stiffness value in the admittance control model to zero, the human-machine collaboration effect can be achieved. At this time, the operator can manually pull the part to be assembled to the docking preparation position. In this embodiment, the reference part has two guide pins. Through collaborative traction, the head of the guide pin is brought into the hole of the part to be assembled by about 1mm, so as to ensure that the guide pin can be captured by the hole after the automatic assembly starts, thereby obtaining effective contact force, so that automatic assembly can be carried out. The same applies to other working conditions.

[0088] Step 130: Control the robot to feed the parts to be assembled according to the assembly reference trajectory.

[0089] In this embodiment, the compliant controller can coordinate the force-position relationship and output commands to control the robot's movement. When no contact force is sensed, the robot will carry the part to be assembled along the assembly reference trajectory.

[0090] Step 140: During the feeding process, based on the robot's pose data and the contact force data collected by the six-dimensional force sensor on the robot, the target position of the trajectory point after force-position relationship coordination is determined by the admittance controller, and the assembly reference trajectory is initially corrected based on the target position.

[0091] When contact force is sensed, the admittance controller comes into play and adjusts the force-position relationship. The six-dimensional force sensor installed at the end of the robot feeds back the collected contact force data to the admittance controller. The admittance controller processes and analyzes the current contact state, calculates the target position after the force-position relationship is coordinated, makes preliminary corrections to the assembly reference trajectory based on the target position, and then outputs control commands to the robot so that the robot can assemble and feed according to the preliminary optimized trajectory.

[0092] Step 150: Based on the actual position, contact force data and series stiffness of the historical trajectory points during the feeding process, an ideal trajectory is fitted and generated, and the subsequent feeding process is iteratively corrected based on the ideal trajectory.

[0093] In this embodiment, based on the relevant data of historical trajectory points on the historical trajectory already traversed by the robot, an ideal trajectory can be fitted and generated, and then the subsequent feeding process can be iteratively corrected to further improve the accuracy of the subsequent feeding process.

[0094] Step 160: Control the robot to stop assembly feeding until the feed termination condition is met.

[0095] In practical applications, by continuously cycling through steps 140 and 150, the robot is made to feed continuously along the optimized trajectory until the contact force data of the feed depth and feed direction both reach the set threshold. At this point, the feed termination condition is met, and the robot can be controlled to stop the assembly feed.

[0096] In some embodiments, the set threshold corresponding to the contact force data and the set threshold corresponding to the feed depth can be reasonably selected according to the actual application scenario. In actual applications, the value of the feed depth will change according to the working conditions and structural changes of the assembly object.

[0097] In one embodiment, the pose information of the trajectory starting point can be obtained through the following process:

[0098] The first step is to establish a robot tool coordinate system based on the structure of the part to be assembled.

[0099] In this step, based on the structure of the part to be assembled at the end of the robot, the origin position and orientation of the robot tool coordinate system can be set. The assembly feed direction is set as the positive X-axis, the vertical upward direction is set as the positive Z-axis, and the Y-axis direction is determined based on the right-hand Cartesian coordinate system, thereby establishing the robot tool coordinate system.

[0100] The second step is to determine the measurement coordinate system corresponding to the six-dimensional force sensor on the robot, based on the robot tool coordinate system.

[0101] In a specific implementation, the measurement coordinate system corresponding to the six-dimensional force sensor on the robot is determined based on the robot tool coordinate system, specifically including:

[0102] First, the initial position and initial orientation of the measurement origin of the six-dimensional force sensor are calibrated in the robot tool coordinate system to obtain the calibration results.

[0103] Then, based on the calibration results, the initial orientation of the coordinate axes and the position of the measurement origin of the original coordinate system corresponding to the six-dimensional force sensor are adjusted to obtain a measurement coordinate system that matches the robot tool coordinate system.

[0104] In this step, the position of the default measurement origin of the six-dimensional force sensor in the robot tool coordinate system and its orientation in the robot tool coordinate system can be calibrated. That is, the initial position and initial orientation of the measurement origin of the six-dimensional force sensor are calibrated, and the initial directions of the three coordinate axes and the position of the measurement origin in the original coordinate system of the six-dimensional force sensor are adjusted accordingly to match the measurement coordinate system with the robot tool coordinate system.

[0105] The third step involves collaborative traction based on the robot tool coordinate system and the measurement coordinate system to pull the part to be installed to a suitable docking preparation position, with the docking preparation position as the starting point of the trajectory.

[0106] The fourth step is to determine the position coordinates of the trajectory starting point in the robot's base coordinate system.

[0107] In this step, the docking preparation position of the part to be installed after human-machine collaboration is taken as the trajectory starting point, and the position coordinates of the trajectory starting point in the robot base coordinate system are (x0, y0, z0).

[0108] The fifth step is to determine the direction vector of the trajectory starting point in the base coordinate system, taking the positive direction of the coordinate axis corresponding to the robot tool coordinate system and the assembly feed direction as the trajectory extension direction.

[0109] The positive direction of the coordinate axis corresponding to the robot tool coordinate system and the assembly feed direction is taken as the trajectory extension direction. For example, the positive X-axis of the robot tool coordinate system can be taken as the extension direction, and the direction vector of the trajectory starting point can be denoted as α.

[0110] The sixth step is to use the position coordinates and direction vector as the pose information of the trajectory starting point.

[0111] In one embodiment, an assembly reference trajectory is determined based on the pose information of the trajectory starting point, the expected assembly time, and the assembly feed distance, specifically including:

[0112] The first step is to determine the number of trajectory points based on the expected assembly time and the preset control cycle.

[0113] The second step is to determine the single-step feed distance based on the assembly feed distance and the number of trajectory points.

[0114] The third step is to determine the increment of each trajectory point in each direction based on the single-step feed distance and the direction vector of the trajectory starting point.

[0115] The fourth step is to determine the position coordinates of each trajectory point based on the position coordinates of the trajectory starting point and the increments of each trajectory point in each direction.

[0116] The fifth step is to combine the position coordinates of all trajectory points to obtain the assembly reference trajectory.

[0117] In this embodiment, the desired assembly time is set to t, for example, t can be 20s, and the assembly feed distance is set to d, for example, d can be 35mm. Based on the above information, the assembly reference trajectory is generated.

[0118] Assuming the robot's control cycle is T = 0.012s, the number of segmented trajectory points is k = t / T = 1667, and the single-step feed distance is Δd = d / k = 0.021mm. Therefore, the increment in each direction between each trajectory point is:

[0119] [Δx Δy Δz] T =0.021mm×α (1)

[0120] The coordinates of the i-th trajectory point are:

[0121]

[0122] In trajectory drawing, the attitude reference value is taken from the attitude of the trajectory starting point, so that an assembly reference trajectory can be obtained to guide the robot's automatic assembly process.

[0123] This embodiment uses a human-machine collaboration approach for preliminary guidance before assembly. The part to be assembled is manually pulled and its spatial position is adjusted to meet the docking conditions, and an assembly reference trajectory and endpoint target position are generated, thereby solving the problem of not being able to obtain the target pose through measurement.

[0124] In one embodiment, determining the series stiffness of the assembly object, including the reference member and the part to be assembled, specifically includes:

[0125] The first step is to construct a series stiffness model of the assembled objects.

[0126] The second step involves measuring the first deformation at the end of the component to be installed, the second deformation at the end of the reference component, the deformation correlation amount, and the elastic force value in the series stiffness model during the direct or indirect contact between the reference component and the component to be installed.

[0127] The third step is to calculate the series stiffness based on the first deformation, the second deformation, the deformation correlation, and the elastic force value.

[0128] In practical applications, there are various methods for measuring series stiffness. This embodiment uses the gasket method and the contact method as examples for illustration.

[0129] The measurement principle of the gasket method can be found in [reference needed]. Figure 3 ,like Figure 3 As shown, n metal shims 310 are inserted into the gap between the reference part 210 and the part to be installed 230. Both the end of the part to be installed (robot side) and the end of the reference part (carriage side) deform. Since the metal shims are standard parts, their stiffness is extremely high after compaction, so the deformation of the metal shims is not considered. It is assumed that the first deformation of the end of the part to be installed is Δy. r The second deformation at the reference end is Δy e The thickness of the shim that causes deformation is the deformation-related quantity, denoted as Δy. The elastic force is ΔF. The stiffnesses of the end to be installed and the end to the reference part are K, respectively. r and K e Then we can get:

[0130]

[0131] Combining the above two equations, we can obtain:

[0132]

[0133] In the formula, As the reciprocal of the series stiffness, based on a certain actual measurement scenario, the elastic force value ΔF is 9.7N, the deformation correlation Δy is 0.4mm, and the series stiffness is 24250N / m.

[0134] The measurement principle of the contact method can be found in [reference needed]. Figure 4 ,like Figure 4 As shown, the upper side illustrates the positional state between the reference component 210 and the component to be installed 230 under the critical contact state. After controlling the robot to bring the reference component 210 into direct contact with the component to be installed 230 and gradually increasing the interference, Figure 4 The lower middle section shows the positional states of the reference part 210 and the part to be installed 230 under deformation. The original feedback position is y0, the actual contact position is y1, the robot feedback position is y2, and the first deformation of the part to be installed is Δy. r =y1-y0, the second deformation at the reference end is Δy e =y2-y1, the deformation correlation quantity is Δy=Δy r +Δy e =y2-y0, and the series stiffness can be obtained by measuring the contact force data during this process and performing calculations.

[0135] Understandably, the measurement principle of the contact method is basically the same as that of the gasket method, except that the deformation correlation quantity Δy is no longer the gasket thickness, but rather the positional difference between the robot and the critical deformation state. This data is obtained from the feedback value of the robot control system. In the same actual measurement scenario, the elastic force value ΔF is 7.5N, the deformation correlation quantity Δy is 0.3mm, and the series stiffness is 25000N / m. It is easy to see that the measurement results of the series stiffness obtained by the two strategies mentioned above are similar, with an error within 3%.

[0136] In one embodiment, based on the robot's pose data and contact force data collected by the six-dimensional force sensor on the robot, the target position of the trajectory point after force-position relationship coordination is determined by the admittance controller, specifically including:

[0137] First, based on the admittance control principle and the robot's pose data, a correspondence is established between the contact force data and the deviation data of the trajectory points. The deviation data includes the positional deviation between the actual position and the target position of the trajectory points, the velocity deviation between the actual velocity and the target velocity, and the acceleration deviation between the actual acceleration and the target acceleration.

[0138] Then, the correspondence between the contact force data and the deviation data of the trajectory points is discretized by the admittance controller, and integral calculation is performed to obtain the target position of the trajectory points after the force-position relationship is coordinated.

[0139] In this embodiment, F exM represents the actual external force acting on the controlled object. d Indicates inertial (mass) parameters, such as M d The value can be 300 N / kg, B d This represents the damping parameter of an object, such as B. d The value can be 2000 Ns / m, K d Represents stiffness parameters, such as K. d The value can be 4000 N / m, x, Let x represent the true position, true velocity, and true acceleration, respectively. d , Let x represent the target position, target velocity, and target acceleration, respectively. e =xx d According to the admittance control principle, we can obtain:

[0140]

[0141] In practical applications, robots can use RSI modules for communication and control. The communication and control cycle is T = 12ms. Therefore, the robot control process is essentially a discrete process, requiring discretization of the admittance control principle. All integral elements are approximated by multiplying the control cycle by the change. Let the sampling period count be k. Taking a single direction as an example, the integration process is as follows:

[0142]

[0143] In the above formula, x(k) is the target position after the force-position relationship is coordinated by the compliant controller.

[0144] In one embodiment, an ideal trajectory is fitted and generated based on the actual position of historical trajectory points, contact force data, and series stiffness during the feeding process, specifically including:

[0145] The first step is to multiply the contact force data with the reciprocal of the series stiffness for each historical trajectory point to obtain intermediate parameters.

[0146] The second step is to calculate the difference between the actual position of the historical trajectory point and the intermediate parameters to determine the ideal position of the historical trajectory point, thus obtaining the ideal trajectory point.

[0147] The third step is to fit all the ideal trajectory points within the target segment to generate the ideal trajectory.

[0148] In this embodiment, the ideal trajectory point can be determined by calculating the coordinate values ​​of the X-axis, Y-axis, and Z-axis, respectively. Taking the determination process of the Y-axis coordinate value as an example, as follows... Figure 5As shown, assuming there is a deviation between the reference trajectory and the ideal trajectory, and assuming the reference coordinate value of the Y-axis is y4 and the ideal coordinate value is y1, under the influence of the deviation and environmental stiffness, contact force is generated due to deformation. This will generate an output value coordinated by the admittance controller to control the robot's movement. However, since the robot's own stiffness is not infinite, deformation will also occur. When in equilibrium, let the output value of the admittance controller be y3. Due to deformation, the actual coordinate value of the robot's end effector is not this value, but will be between y1 and y3, denoted as y2. Then, the following relationship exists:

[0149] K r ·(y3-y2)=K e ·(y2-y1)=F (7)

[0150] Where F represents the contact force data acquired by the six-dimensional force sensor, and K... r K e Let y1 be the stiffness of the end to be installed and the stiffness of the reference part, respectively. Then the ideal coordinate value y1 is:

[0151]

[0152] In the formula, y3 is the output value of the admittance controller, which is the position fed back by the robot control system; F is the contact force data collected by the six-dimensional force sensor. It is the reciprocal of the series stiffness.

[0153] Similarly, the calculation process for the ideal coordinate values ​​x1 and z1 corresponding to the X-axis and Z-axis is similar to the calculation process for the ideal coordinate value y1 of the Y-axis, and will not be repeated here.

[0154] According to the above process, the corresponding ideal position can be calculated for each historical trajectory point of the historical assembly trajectory that has been traversed, thereby obtaining multiple ideal trajectory points. Finally, by using the least squares method, all the ideal trajectory points within a certain period of time (i.e., within the target segment) are fitted into a straight line to obtain the corrected ideal trajectory.

[0155] This embodiment calculates the deformation and ideal point coordinates of each data set by using a series stiffness model and the pose and contact force data fed back by the robot and the six-dimensional force sensor. During the assembly process, the ideal trajectory is fitted in real time and the original reference trajectory is corrected, thereby significantly reducing the contact force and assembly time during the assembly process.

[0156] In one embodiment, after controlling the robot to stop the assembly feed, the above method may further include:

[0157] First, the robot is controlled to bring the parts to be loaded into contact with the reference parts in a preset direction and cause deformation.

[0158] In this step, the preset directions can be the four directions in the robot tool coordinate system: +Y, -Y, +Z, and -Z. The adjustment process is performed separately for the Y-axis and Z-axis.

[0159] Then, the contact force data collected by the six-dimensional force sensor during the contact and deformation process is acquired;

[0160] Finally, based on the contact force data, the assembly gap between the part to be assembled and the reference part is adjusted.

[0161] In a specific implementation, see Figure 6 Taking the adjustment process in the Y-axis direction as an example, based on the contact force data, the assembly gap between the part to be assembled and the reference part is adjusted, specifically including:

[0162] Step 410: After coordinating the Y-axis positive assembly clearance and contact stress, control the robot to slowly move the workpiece to be assembled along the Y-axis positive direction.

[0163] Step 420: Determine whether the contact force data reaches the first set range. In this embodiment, the first set range can be -11N to -9N.

[0164] Step 430: When the contact force data is determined to reach the first set range, determine the position coordinates (x1, y1, z1) of the first key point.

[0165] Step 440: Then control the robot to slowly move the part to be loaded along the negative Y-axis.

[0166] Step 450: Determine whether the contact force data reaches the second set range. In this embodiment, the second set range can be 9N to 11N.

[0167] Step 460: When the contact force data is determined to reach the second set range, determine the position coordinates (x2, y2, z2) of the second key point.

[0168] Step 470: Based on the position coordinates of the first key point and the second key point, determine the midpoint of the line connecting the first key point and the second key point, and control the robot to move the part to be installed to the position of the midpoint of the line.

[0169] Similarly, the adjustment process in the Z-axis direction is basically the same as the adjustment process in the Y-axis direction, and will not be repeated here. In this embodiment, the workpiece to be assembled is brought into contact with the reference workpiece along four preset directions (+Y, -Y, +Z, -Z) in a predetermined order, and small deformations are generated until the contact force data reaches the set range. Then, the gap distribution is calculated and evenly distributed. By using a six-dimensional force sensor and a gap coordination control algorithm, the gap and contact stress are coordinated before the assembly is completed, which can alleviate the problem of uneven contact stress and improve the assembly quality.

[0170] like Figure 7 As shown, the complete implementation flow of the compliant assembly method based on admittance control and ideal trajectory fitting provided in this embodiment is as follows:

[0171] Step 510: Set the robot tool coordinate system.

[0172] Step 520: Set the measurement coordinate system of the six-dimensional force sensor.

[0173] Step 530: Working condition preparation. Move the frame car with the reference parts to the assembly docking station and lock the wheel fixing position. Connect the part to be assembled to the six-dimensional force sensor at the end of the robot through the clamping fixture.

[0174] Step 540: After completing the preliminary preparations corresponding to steps 510 to 530, the series stiffness can be measured. Specifically, a series stiffness model can be constructed based on the assembly object, and its series stiffness can be measured.

[0175] Step 550: Collaborate to pull the part to be installed to the docking preparation position.

[0176] Step 560: Generate assembly reference trajectory.

[0177] Step 570: The compliant controller coordinates the force-position relationship and outputs commands to control the robot's movement for assembly feeding.

[0178] Step 580: Ideal trajectory fitting and correction. The ideal trajectory is fitted and corrected for subsequent feed processes.

[0179] Step 590: Continuously cycle through steps 570 and 580 until the contact force data of both the feed depth and feed direction reach the set threshold, then stop the assembly feed.

[0180] Step 5100: Stress and clearance coordination. Adjust the assembly clearance by contact stress to make the assembly clearance distribution uniform.

[0181] Step 5110: End assembly operation.

[0182] In summary, the compliant assembly method based on admittance control and ideal trajectory fitting provided by the embodiments of the present invention has at least the following advantages compared with traditional assembly schemes:

[0183] First, compared to assembly guidance processes based on vision or optical measurement, this embodiment adopts a human-machine collaboration mode to manually pull the parts to be assembled to the docking preparation position, which eliminates the measurement process, reduces equipment costs, and reduces operation steps. It is suitable for assembly conditions where there is no accurate target pose or measurement conditions.

[0184] Secondly, compared to the manual assembly mode, this embodiment can quantitatively control the contact force data during the assembly process, making it easier to ensure assembly quality. Moreover, the robot has a stronger load-bearing capacity and can be used to assemble larger and heavier objects.

[0185] Third, compared with the traditional compliant assembly strategy, although this embodiment also constructs a compliant controller based on the admittance principle, it is no longer a passive compromise adjustment, but actively fits and corrects the ideal trajectory based on the force position data in the assembly process, thereby greatly reducing the deviation between the actual assembly trajectory and the ideal value, and finally coordinates the assembly gap and contact stress based on the contact force data, thus improving the assembly quality.

[0186] Based on the same general inventive concept, this invention also protects a compliant assembly system based on admittance control and ideal trajectory fitting. The compliant assembly system based on admittance control and ideal trajectory fitting provided by this invention will be described below. The compliant assembly system based on admittance control and ideal trajectory fitting described below can be referred to in correspondence with the compliant assembly method based on admittance control and ideal trajectory fitting described above.

[0187] Figure 8 This is a schematic diagram of the compliant assembly system based on admittance control and ideal trajectory fitting provided in an embodiment of the present invention.

[0188] like Figure 8 As shown, the compliant assembly system based on admittance control and ideal trajectory fitting provided in this embodiment of the invention specifically includes: a control device 600 and a robot 700 equipped with a six-dimensional force sensor.

[0189] It is understandable that the control device 600 can be the robot's own controller or a controller deployed outside the robot.

[0190] like Figure 9 As shown, the control device may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640. The processor 610, communications interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call logic instructions from the memory 630 to execute the compliant assembly method based on admittance control and ideal trajectory fitting provided in the above embodiments.

[0191] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0192] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the compliant assembly method based on admittance control and ideal trajectory fitting provided in the above embodiments.

[0193] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the compliant assembly method based on admittance control and ideal trajectory fitting provided in the above embodiments.

[0194] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0195] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A compliant assembly method based on admittance control and ideal trajectory fitting, characterized in that, include: After moving the reference component to the assembly docking station and installing the component to be assembled onto the robot end effector, the series stiffness of the assembly object including the reference component and the component to be assembled is determined. After the part to be assembled is moved to the docking preparation position through human-machine collaboration, the docking preparation position is used as the trajectory starting point. Based on the pose information of the trajectory starting point, the expected assembly time and the assembly feed distance, the assembly reference trajectory is determined. The robot is controlled to drive the part to be assembled to feed along the assembly reference trajectory; During the feeding process, based on the robot's pose data and the contact force data collected by the six-dimensional force sensor on the robot, the target position of the trajectory point after force-position relationship coordination is determined by the admittance controller, and the assembly reference trajectory is initially corrected based on the target position. Based on the actual position of historical trajectory points during the feeding process, the contact force data, and the series stiffness, an ideal trajectory is fitted and generated, and the subsequent feeding process is iteratively corrected based on the ideal trajectory. The robot is controlled to stop assembly feeding until the feed termination condition is met. The pose information of the trajectory starting point is obtained through the following process: Based on the structure of the component to be assembled, establish a robot tool coordinate system; Based on the robot tool coordinate system, determine the measurement coordinate system corresponding to the six-dimensional force sensor on the robot; Based on the robot tool coordinate system and the measurement coordinate system, the workpiece to be loaded is pulled to the docking preparation position, with the docking preparation position as the starting point of the trajectory. Determine the position coordinates of the trajectory starting point in the robot's base coordinate system; Using the positive direction of the coordinate axis corresponding to the assembly feed direction in the robot tool coordinate system as the trajectory extension direction, the direction vector of the trajectory starting point in the base coordinate system is determined. The position coordinates and the direction vector are used as the pose information of the starting point of the trajectory; The process of determining the assembly reference trajectory based on the pose information of the trajectory starting point, the expected assembly time, and the assembly feed distance includes: The number of trajectory points is determined based on the expected assembly time and the preset control cycle; The single-step feed distance is determined based on the assembly feed distance and the number of trajectory points; Based on the single-step feed distance and the direction vector of the trajectory starting point, the increment of each trajectory point in each direction is determined; Based on the position coordinates of the starting point of the trajectory and the increments of each trajectory point in each direction, the position coordinates of each trajectory point are determined. By combining the position coordinates of all trajectory points, the assembly reference trajectory is obtained; Determining the series stiffness of the assembly object, which includes the reference component and the component to be assembled, includes: Construct a series stiffness model of the assembled object; During the process of direct or indirect contact between the reference component and the component to be installed, the first deformation at the end of the component to be installed, the second deformation at the end of the reference component, the deformation correlation amount, and the elastic force value are measured in the series stiffness model. The series stiffness is calculated based on the first deformation, the second deformation, the deformation correlation amount, and the elastic force value.

2. The compliant assembly method based on admittance control and ideal trajectory fitting according to claim 1, characterized in that, Determining the measurement coordinate system corresponding to the six-dimensional force sensor on the robot based on the robot tool coordinate system includes: In the robot tool coordinate system, the initial position and initial orientation of the measurement origin of the six-dimensional force sensor are calibrated to obtain the calibration results; Based on the calibration results, the initial orientation of the coordinate axes and the position of the measurement origin of the original coordinate system corresponding to the six-dimensional force sensor are adjusted to obtain a measurement coordinate system that matches the robot tool coordinate system.

3. The compliant assembly method based on admittance control and ideal trajectory fitting according to claim 1, characterized in that, The step of determining the target position of the trajectory point after force-position coordination through an admittance controller, based on the robot's pose data and contact force data collected by the six-dimensional force sensor on the robot, includes: Based on the admittance control principle and the robot's pose data, a correspondence is established between the contact force data and the deviation data of the trajectory points; wherein, the deviation data includes the positional deviation between the actual position and the target position of the trajectory point, the velocity deviation between the actual velocity and the target velocity, and the acceleration deviation between the actual acceleration and the target acceleration. The correspondence between the contact force data and the deviation data of the trajectory points is discretized by the admittance controller, and then integrated to obtain the target position of the trajectory points after the force-position relationship is coordinated.

4. The compliant assembly method based on admittance control and ideal trajectory fitting according to claim 1, characterized in that, The process of fitting and generating an ideal trajectory based on the true position of historical trajectory points during the feeding process, the contact force data, and the series stiffness includes: For each historical trajectory point, the contact force data is multiplied by the reciprocal of the series stiffness to obtain intermediate parameters; The ideal position of the historical trajectory point is determined by subtracting the intermediate parameter from the actual position of the historical trajectory point, thus obtaining the ideal trajectory point. The ideal trajectory is generated by fitting all the ideal trajectory points within the target segment.

5. The compliant assembly method based on admittance control and ideal trajectory fitting according to claim 1, characterized in that, After controlling the robot to stop the assembly feed, the method further includes: The robot is controlled to move the parts to be installed to contact the reference parts in a preset direction and cause deformation. The contact force data collected by the six-dimensional force sensor during the contact and deformation process is acquired; Based on the contact force data, the assembly gap between the part to be assembled and the reference part is adjusted.

6. The compliant assembly method based on admittance control and ideal trajectory fitting according to claim 5, characterized in that, The step of adjusting the assembly gap between the part to be assembled and the reference part based on the contact force data includes: Determine whether the contact force data reaches a first preset range; When it is determined that the contact force data has reached the first set range, the position coordinates of the first key point are determined; Determine whether the contact force data reaches the second preset range; When the contact force data is determined to reach the second preset range, the position coordinates of the second key point are determined; Based on the position coordinates of the first key point and the position coordinates of the second key point, determine the midpoint of the line connecting the first key point and the second key point; The robot is controlled to move the component to be installed to the position of the midpoint of the connecting line.

7. A compliant assembly system based on admittance control and ideal trajectory fitting, characterized in that, include: Control equipment and robots equipped with six-dimensional force sensors; The control device includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the program, it implements the compliant assembly method based on admittance control and ideal trajectory fitting as described in any one of claims 1 to 6.

Images