A Robotic Flexible Assembly Method and System Based on Assembly State Recognition

Abstract

This invention belongs to the field of robot program control technology, specifically relating to a robot flexible assembly method and system based on assembly state cognition. The method includes: acquiring force and torque signals and filtering them to obtain smooth force and torque vectors; calculating the coordinates of the equivalent contact point based on the wrench decomposition principle; determining the contact surface normal vector based on the force vector sequence and the principle of minimizing projection variance, and decomposing the force vector into normal support force and friction force; obtaining the instability index of the contact state based on the degree of change between the equivalent contact point position and the contact surface normal vector; and executing an adaptive avoidance strategy to eliminate the risk of jamming when the instability index exceeds a preset threshold. This invention solves the technical problems of unclear contact state cognition and difficulty in identifying early signs of jamming in robot flexible assembly, improving the stability and success rate of the assembly process.

Description

Technical Field

[0001] This invention relates to the field of robot program control technology. More specifically, this invention relates to a robot flexible assembly method and system based on assembly state recognition. Background Technology

[0002] In automated production, robotic flexible assembly is a key technology for achieving intelligent manufacturing. Its core lies in equipping robots with force and torque sensors, giving them a human-like sense of touch, enabling them to perceive and adapt to changes in contact force caused by uncertainties such as part tolerances and positioning errors. When performing tasks such as precision fitting, the robot dynamically adjusts its movement posture by monitoring the interaction forces and torques between its end effector and the environment in real time, thus preventing parts from jamming or being damaged, and completing the assembly.

[0003] However, existing technologies still face systemic challenges in understanding and applying the high-dimensional, coupled data output by force and torque sensors. The original force and torque signals are essentially abstract numerical streams, and traditional methods struggle to accurately deduce the complete physical picture of contact events from them, such as the specific location of the contact and the direction of the normal to the contact surface. Moreover, due to a lack of in-depth understanding of the contact state, existing technologies cannot effectively assess the stability of the assembly process, especially for the critical stage of the evolution from stable single-point contact to unstable multi-point contact, where a reliable identification mechanism is lacking.

[0004] To address these issues, some industry research attempts to analyze force and torque data using algorithmic models. For example, based on rigid body dynamics principles, the position of the equivalent point of application can be calculated from the force and torque data. However, this model typically relies on a single-point contact assumption. In complex scenarios where the contact state transitions from single-point to multi-point, the calculation results can fluctuate drastically or even fail, making it difficult to stably reflect the true contact geometry. Furthermore, some methods attempt to decouple normal force and friction through force projection variance analysis. The basic assumption is that the normal force remains stable during sliding, while the friction changes with the direction of motion. However, when the contact surface itself changes rapidly or enters a jammed state, this assumption no longer holds, causing the algorithm to fail to accurately separate the normal force and friction, thus affecting subsequent control strategies. Therefore, robots still largely rely on passive force threshold judgments for trial-and-error correction, making it difficult to achieve high-success-rate precision assembly under complex working conditions. Summary of the Invention

[0005] To address the technical problem that robots struggle to achieve high success rates in precision assembly under complex working conditions, this invention provides solutions in several aspects.

[0006] In a first aspect, the present invention provides a robot flexible assembly method based on assembly state cognition, comprising: during the robot's assembly task, acquiring and smoothing the force vector and torque vector generated when the robot's end effector contacts the workpiece; based on the wrench decomposition principle, converting the smoothed force vector and torque vector into a spatial equivalent force system, and acquiring the coordinate vectors of the equivalent points of action of the contact force and torque sensed by the sensor; based on the force vector sequence collected in a continuous control cycle, determining the normal vector of the contact surface between the robot's end effector and the workpiece based on the principle of minimizing projection variance, and decomposing the force vector into a normal support force perpendicular to the contact surface and a frictional force along the contact surface; acquiring the instability index of the contact state based on the changes in the equivalent contact point position and the changes in the normal vector of the contact surface in adjacent control cycles; preset a threshold for the instability index, and in response to the instability index being less than or equal to the threshold, the controller executes compliant assembly control, and in response to the instability index being greater than the threshold, the controller executes an adaptive avoidance strategy to ensure the stability of the assembly process.

[0007] This invention achieves precise understanding of the physical geometry of the contact state by analyzing the collected force and torque signals into equivalent contact point positions and contact surface normal vectors. The stability of the contact state is assessed by calculating the rate of change of the equivalent contact points and normal vectors, resulting in an instability index that characterizes the transition from stable sliding to unstable multi-point contact. This index serves as an effective precursor to jamming risk, enabling the system to identify unstable trends in the assembly process before high-force jamming occurs. When the index exceeds a preset threshold, the system can proactively execute adaptive avoidance strategies such as normal retraction and lateral jitter to break the forming jamming conditions and restore stability. This closed-loop control based on precise state understanding and precursor recognition solves the technical problem of traditional methods struggling to effectively assess stability and avoid jamming under complex working conditions, thereby improving the stability and success rate of robotic flexible assembly.

[0008] Preferably, the step of obtaining the force vector and torque vector generated when the robot end effector contacts the workpiece includes: in each control cycle, the system collects the force vector and torque vector generated when the robot end effector contacts the workpiece, and uses a moving average filtering method to smooth the data to obtain the smoothed force vector and torque vector generated when the robot end effector contacts the workpiece.

[0009] This invention uses a moving average filter to process the original signal, providing a stable and reliable data foundation for subsequent accurate analysis of the contact state. By averaging the data from the most recent control cycle, the interference of the inherent high-frequency noise of the sensor on the force and torque signals can be suppressed, ensuring the stability and accuracy of subsequent calculations and avoiding model solution failures or violent oscillations in results caused by data jumps.

[0010] Preferably, the coordinate vector of the equivalent point of application of the contact force and torque sensed by the sensor satisfies the expression: ;in, Represents the coordinate vector of the equivalent contact point; Represents a force vector; Represents the torque vector; This represents the dot product operation of vectors. This represents the magnitude of the vector.

[0011] This invention is based on the wrench decomposition principle in rigid body dynamics, providing a calculation method for real-time analysis of the equivalent contact point position. It can transform the abstract, high-dimensional force and torque signals output by the sensor into an intuitive and concrete spatial coordinate in real time, and concretize the complex force information into a geometric position. This allows the system to clearly recognize the specific location where the contact occurs, laying the foundation for subsequent evaluation of the stability of the contact state. It is a key step in realizing the transformation from abstract force perception to concrete state cognition.

[0012] Preferably, the normal vector of the contact surface between the robot end effector and the workpiece satisfies the expression: ;in, Represents the normal vector of the contact surface; Represents the candidate normal vector; This indicates the search for the candidate normal vector that minimizes the value of the expression within the parentheses. ; Indicates variance; It is a time window At any time within; Indicates the time window Internal acquisition at any time The force vector; Indicates the current moment; This indicates the length of the time window used to calculate the variance. This represents the dot product operation of vectors.

[0013] Preferably, the force vector is decomposed into a normal supporting force perpendicular to the contact surface and a frictional force along the contact surface, including: ; ;in, This represents the normal force vector between the robot's end effector and the workpiece. Represents the normal vector of the contact surface; This represents the frictional force vector between the robot's end effector and the workpiece. Represents a force vector; This represents the dot product operation of vectors.

[0014] Preferably, the instability index of the contact state satisfies the expression: ;in, It is a moment; Indicates in The instability index at any given moment; These represent the equivalent contact point coordinate vectors analyzed at the current moment and in the previous control cycle, respectively. These represent the contact surface normal vectors resolved at the current moment and in the previous control cycle, respectively; Indicates the length of a control cycle; This represents the first hyperparameter used to balance position changes; This represents the second hyperparameter used to balance attitude changes; This represents the dot product operation of vectors. This represents the magnitude of the vector.

[0015] The instability index of this invention integrates the rate of change of the position of the equivalent contact point and the rate of change of the normal vector of the contact surface, thereby enabling real-time and objective assessment of the stability of the assembly process. Since the physical precursor of jamming lies in the transition of the contact geometry from a continuous and smooth evolution to a drastic and discontinuous jump, this index is highly sensitive to the trend of deviating from the stable state. Therefore, it can serve as a reliable precursor indicator for identifying jamming risk. By calculating this index, the system can provide early warning and identify potential assembly failure risks before the contact force increases sharply and complete jamming occurs, providing a key decision-making basis for implementing proactive avoidance strategies.

[0016] Preferably, the threshold of the preset instability index includes: the system running for a long time under various normal operating conditions, collecting a large number of historical instability indices, arranging the historical instability indices from smallest to largest, and setting... Quantiles are threshold values.

[0017] This invention uses statistical learning methods to determine the threshold of the instability index, which improves the reliability and seriousness of the avoidance strategy triggering, reduces false alarms, and ensures that the threshold is data-driven and objective. This allows the system to initiate avoidance actions only when an abnormal event with a very high probability occurs, thereby effectively distinguishing normal signal fluctuations from true jamming precursors, reducing system misjudgments and unnecessary interruptions caused by normal noise, and ensuring the smoothness and high efficiency of the entire assembly process.

[0018] Preferably, the first hyperparameter is set to 1.0.

[0019] Preferably, the second hyperparameter is set to 0.5 m / rad.

[0020] Secondly, the present invention provides a robot flexible assembly system based on assembly state recognition, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned robot flexible assembly method based on assembly state recognition is implemented.

[0021] By adopting the above technical solution, a computer program is generated from the above-mentioned robot flexible assembly method based on assembly state cognition and stored in the memory so that it can be loaded and executed by the processor. In this way, a terminal device can be made based on the memory and the processor for convenient use.

[0022] The beneficial effects of this invention are as follows:

[0023] This invention preprocesses the original signal using moving average filtering, providing a stable and reliable data foundation for subsequent accurate calculations. It resolves force and torque signals into the spatial coordinates of an equivalent contact point, concretizing complex force information into a geometric location. Based on the principle of minimizing projection variance, it determines the unit normal vector of the contact surface and further decomposes the total contact force into physically meaningful normal support force and friction force. This series of steps together constructs a complete physical picture of the location, attitude information, and force properties of the contact event, solving the problem that traditional methods struggle to invert contact information from high-dimensional coupled data.

[0024] This invention establishes a mechanism that can provide early warning of jamming risks, transforming the system from passive force threshold judgment to proactive risk prediction. It obtains the instability index of the current contact state stability and derives it by calculating the equivalent contact point position and the rate of change of the contact surface normal vector in real time. This allows it to sensitively capture the physical precursors of jamming during the assembly process. To ensure the reliability of the prediction, historical data under normal operating conditions is collected to set the trigger threshold of the instability index, thereby effectively avoiding false alarms caused by fluctuations in normal operating conditions.

[0025] This invention integrates precise state perception and risk prediction capabilities into the robot control loop, forming a highly efficient adaptive assembly strategy. When the contact state is stable, the controller can use the decoupled normal force and friction information to perform fine control, thereby ensuring a smooth and efficient assembly process. Once the system identifies a sign of impending jamming through the instability index, it will immediately pause the current task and actively execute adaptive avoidance strategies, including normal retraction, lateral jitter, or rotational relaxation. After the instability index falls below the threshold, the robot can continue to perform the assembly task, enabling the robot to autonomously cope with uncertainties in the assembly process and significantly improving the success rate and stability of flexible assembly under complex working conditions. Attached Figure Description

[0026] Figure 1This is a flowchart illustrating a robot flexible assembly method based on assembly state recognition in this invention;

[0027] Figure 2 This is a schematic diagram illustrating the comparison between the instability index and the threshold in this invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present 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 the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] This invention discloses a robot flexible assembly method based on assembly state recognition, referring to... Figure 1 This includes steps S1-S5:

[0031] S1. During the robot's assembly task, acquire and smooth the force vector and torque vector generated when the robot's end effector contacts the workpiece.

[0032] Specifically, before performing the assembly task, the robot system establishes a tool coordinate system at the tool center point; force and torque sensors are installed between the robot wrist flange and the end effector; during assembly, the force and torque sensors collect force and torque information generated when the robot end effector comes into contact with the workpiece environment in real time at a preset control cycle. The control cycle length in this embodiment is [length missing]. The implementers can choose the length of the control cycle according to the actual situation.

[0033] Furthermore, in each control cycle, the system acquires the force vector generated when the robot's end effector comes into contact with the workpiece. With torque vector The data is smoothed using a moving average filter method, and the size of the moving window is... Preferred ,when Less than When the filtering effect is not obvious, it cannot effectively suppress noise; when Greater than While the signal is smoother, it introduces a significant delay, affecting the system's real-time response capability. This embodiment selects... A good balance can be achieved between noise reduction effect and system real-time performance, and the implementer can choose the size of the sliding window according to the actual situation. The force vector and torque vector generated when the smoothed robot end effector comes into contact with the workpiece are simply referred to as force vector and torque vector.

[0034] S2. Based on the wrench decomposition principle, the smoothed force vector and torque vector are converted into a spatial equivalent force system to obtain the coordinate vector of the equivalent point of action of the contact force and torque sensed by the sensor.

[0035] It should be noted that, after obtaining the force vector and torque vector, to address the technical problem of robots struggling to achieve high-success-rate precision assembly under complex working conditions, this invention first identifies the contact position between the robot and the contact surface. Based on the wrench decomposition principle in rigid body dynamics, any spatial force system can be equivalently represented as a force acting on the line of action of the equivalent force in space, and a torque parallel to that force. The goal of this invention is to deduce a representative point of action in this equivalent force system, i.e., the equivalent contact point.

[0036] Specifically, the coordinate vector of the equivalent point of application of the contact force and torque sensed by the sensor satisfies the expression:

[0037]

[0038] in, Represents the coordinate vector of the equivalent contact point; Represents a force vector; Represents the torque vector; This represents the dot product operation of vectors. This represents the magnitude of the vector.

[0039] In the formula, The vector of the point closest to the origin of the coordinate system on the line of action of the equivalent force was calculated. . Calculate a force vector The displacement vector in the direction, the magnitude of which is determined by the torque. In force The projection in the direction determines the unique location of the equivalent contact point, and the sum of the two vectors ultimately confirms the location of the point of contact.

[0040] When the contact state between the robot's end effector and the environment changes, the force measured by the sensor... and torque This will change accordingly, and the present invention can convert these dynamically changing force information into an intuitive change in geometric position in space in real time. For example, when the side of the pin contacts the edge of the hole, the generated torque... This will cause the calculated equivalent contact point to be located on the side of the pin; when the tip of the pin contacts the bottom of the hole, the torque... When the force is close to zero, the calculated equivalent contact point will be located on the axis of the pin, and the system transforms the abstract perception of force into a concrete understanding of the contact position.

[0041] It should be added that a small threshold for the magnitude of the force vector is set, such as 0.01 Newtons. When the magnitude of the force vector is greater than the preset threshold, the equivalent contact point expression is used for calculation; when its magnitude is less than or equal to the threshold, it is considered that there is no effective contact. At this time, the equivalent contact point is meaningless, and the robot continues to move according to the original assembly program path.

[0042] S3. Based on the force vector sequence collected during the continuous control cycle, determine the normal vector of the contact surface between the robot end and the workpiece according to the principle of minimizing the projection variance, and decompose the force vector into a normal support force perpendicular to the contact surface and a frictional force along the contact surface.

[0043] It should be noted that after resolving the coordinate vectors of the equivalent contact points, the system also needs to understand the nature of the contact, i.e., how the robot should adjust its posture. The force generated when the robot's end effector contacts the workpiece is the vector sum of the normal force vector perpendicular to the contact surface and the frictional force vector tangential to the contact surface. This invention separates the normal force vector and the frictional force vector from the mixed force vector.

[0044] Specifically, the expression for the normal vector of the contact surface between the robot end effector and the workpiece is as follows:

[0045]

[0046] in, Represents the normal vector of the contact surface; Represents the candidate normal vector; This indicates the search for the candidate normal vector that minimizes the value of the expression within the parentheses. ; Indicates variance; It is a time window At any time within; Indicates the time window Internal acquisition at any time The force vector; Indicates the current moment; This indicates the length of the time window used to calculate the variance. This represents the dot product operation of vectors.

[0047] The system iterates through all possible unit directions, ranging from 1 degree to 360 degrees, in increments of 1 degree. And for each Calculation in the past Force vector sequence within time t. Projection value in this direction variance Find the direction that minimizes variance. As the most reliable contact surface normal vector currently available The reason is that, in the process of tiny sliding assembly, as long as the contact does not change abruptly, such as sliding from one surface to another, the normal direction of the contact surface is relatively stable; while the magnitude and direction of the friction force will change significantly with the change of the robot's motion trend. Therefore, the projection of the force vector in the real normal direction should have minimal fluctuation.

[0048] Furthermore, once the normal vector of the contact surface is determined, the normal force and frictional force between the robot's end effector and the workpiece can be decoupled through vector projection. Specifically, ; ;in, This represents the normal force vector between the robot's end effector and the workpiece. Represents the normal vector of the contact surface; This represents the frictional force vector between the robot's end effector and the workpiece. Represents a force vector; This represents the dot product operation of vectors.

[0049] In this embodiment, the preferred time window length is... to If the time window length is too short, for example less than... If the data sample is too small, the variance calculation is easily affected by random noise, leading to... The estimation is unstable; if the time window length is too long, for example, greater than... If this happens, the system's response to actual changes in the contact surface will become sluggish. In this embodiment, the time window length is selected as... It can achieve a good balance between estimating stability and dynamic responsiveness, and implementers can choose other time window lengths according to actual needs.

[0050] S4. Obtain the instability index of the contact state based on the changes in the equivalent contact point position and the changes in the contact surface normal vector within adjacent control cycles.

[0051] It should be noted that, based on obtaining the equivalent contact point and the normal vector of the contact surface, this invention aims to assess the stability of the current contact state and provide early warning of the impending jamming risk. The physical basis for this is that the evolution of the equivalent contact point and the normal vector of the contact surface in a stable single-point sliding contact process should be continuous and smooth; conversely, when the assembly process transitions from stable single-point contact to unstable multi-point contact, the equivalent contact point and the normal vector of the contact surface will undergo drastic and discontinuous jumps.

[0052] Specifically, in order to assess the degree of this transition, the present invention obtains an instability index of the contact state, which satisfies the expression:

[0053]

[0054] in, It is a moment; Indicates in The instability index at any given moment; These represent the equivalent contact point coordinate vectors analyzed at the current moment and in the previous control cycle, respectively. These represent the contact surface normal vectors resolved at the current moment and in the previous control cycle, respectively; Indicates the length of a control cycle; This represents the first hyperparameter used to balance position changes; This represents the second hyperparameter used to balance attitude changes; This represents the dot product operation of vectors. This represents the magnitude of the vector.

[0055] in, This is a weighted value of the moving speed of the equivalent contact point. During stable sliding, its value matches the robot's nominal motion speed; when the equivalent contact point slides from one plane to another, this value will surge instantaneously.

[0056] in, This is a weighted value of the rotational angular velocity of the normal vector of the contact surface. When moving along a smooth surface, its value changes slowly; when the equivalent contact point encounters a sharp corner or forms a new equivalent contact point, causing a drastic change in the equivalent normal, this value will surge instantaneously.

[0057] The first hyperparameter range is [0.7, 1.3], used to balance the weight of the equivalent contact point position change rate. If the first hyperparameter is set too small, the system will be insensitive to sudden changes in contact point position, which may lead to the inability to timely identify the precursors of jamming caused by drastic position changes. If the first hyperparameter is set too high, the system may overreact to normal jitter or smooth movement of the contact position, causing normal assembly to be misjudged as unstable, thereby frequently triggering avoidance strategies and reducing assembly efficiency. In this embodiment, the first hyperparameter is set to... The second hyperparameter range is [0.2m / rad, 0.8m / rad]. If the second hyperparameter is set too low, the system will become insensitive to drastic changes in contact posture, causing the system to fail to identify such severe jamming risks, thus missing the opportunity to execute adaptive avoidance strategies and reducing the assembly success rate. If the second hyperparameter is set too high, the system will overreact to any small changes in the contact surface normal vector, including the normal continuous changes in the normal vector when the robot moves along a smooth surface. As a result, the system will continuously misjudge and frequently trigger unnecessary avoidance actions, seriously affecting the smoothness and efficiency of assembly. The second hyperparameter in this embodiment is set to... =0.5m / rad. In other embodiments, the implementer can select the values ​​of the first hyperparameter and the second hyperparameter according to the actual situation.

[0058] Instability Index It integrates the degree of drastic change in contact geometry and attitude, becoming an indicator that is highly sensitive to any deviation from a stable and continuous evolutionary trend, and is the core index for recognizing precursors to catastrophic events.

[0059] S5. A threshold for the instability index is preset. When the instability index is less than or equal to the threshold, the controller executes compliant assembly control. When the instability index is greater than the threshold, the controller executes an adaptive avoidance strategy to ensure the stability of the assembly process.

[0060] Specifically, the system operates under various normal conditions for extended periods, collecting a large number of historical instability indices, which are then arranged from smallest to largest. quantiles are thresholds This ensures that the system only triggers the avoidance strategy when an abnormal event with a very high probability occurs, thus avoiding false alarms caused by normal noise.

[0061] Furthermore, the system monitors in real time during the assembly process. The value is then used to execute adaptive control logic:

[0062] In response to The system determines that the current contact state is stable. At this point, the robot controller can use the decoupled force information to perform fine control. For example, a force controller can be implemented to fine-tune the robot's posture so that the normal force... Maintain at a level such as A smaller expected value is needed to achieve a smooth fit between the robot and the contact surface; this can be achieved based on friction. The magnitude of the value is used to feedforward compensation of the driving force of the robot along the direction of motion to overcome sliding resistance.

[0063] In response to The system determines that a significant anomaly requiring immediate intervention has occurred, indicating a potential jamming precursor. At this point, the system immediately suspends the current propulsion or insertion action and actively executes a pre-set avoidance strategy aimed at breaking the emerging self-locking or wedge-like conditions.

[0064] For example, the avoidance strategy may include one or more combinations: the controller instructs the robot to move along the currently estimated contact surface normal vector. Make a small backward movement in the opposite direction, for example, move... This action reduces contact pressure and releases stress; while pausing the spindle advance, it applies a force perpendicular to the spindle direction, such as... High frequency, such as A slight translational vibration. This action helps to loosen the equivalent contact point that may have become slightly stuck; by applying a small rotational torque about the main axis, it attempts to change the contact angle through slight rotation, thereby breaking the wedging condition.

[0065] Furthermore, after executing one or more avoidance actions, the system will re-enter state-aware mode to re-analyze the current contact state and instability index. ,if If the temperature falls back below the threshold, it indicates that the jamming risk has been eliminated, and the system will attempt to continue the assembly task using a corrected attitude or path. If If the error rate remains high, the system may try another circumvention strategy, or after multiple failed attempts, it may eventually stop the task and issue a request for manual intervention.

[0066] An exemplary diagram comparing the instability index with a threshold is shown below. Figure 2 As shown, Figure 2 The blue curve represents the change of the instability index of the contact state over time. The red dashed line represents the threshold. The red dots are unstable points where the instability index exceeds the threshold. When the instability index exceeds the threshold, a preset avoidance strategy is executed, and the instability index decreases until it is less than the threshold.

[0067] Thus, a robot flexible assembly method based on assembly state cognition has been realized.

[0068] This invention also discloses a robot flexible assembly system based on assembly state recognition, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a robot flexible assembly method based on assembly state recognition according to the present invention is implemented.

[0069] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A robot flexible assembly method based on assembly state awareness, characterized by, The method comprises: acquiring and smoothing force vectors and torque vectors generated when a robot end effector contacts a workpiece during execution of an assembly task by the robot; based on a wrench decomposition principle, converting the smoothed force vectors and torque vectors into a spatial equivalent force system to obtain a coordinate vector of an equivalent action point of contact force and torque sensed by a sensor; based on a projection variance minimum principle, determining a contact surface normal vector of the robot end effector and the workpiece according to a force vector sequence collected in a continuous control period, and decomposing the force vector into a normal support force perpendicular to the contact surface and a friction force along the contact surface; based on a change in equivalent contact point position and a change in the contact surface normal vector in adjacent control periods, obtaining an instability index of a contact state, satisfying: ； is the time instant; denotes the instability index at the time instant; , denote the current time instant and the equivalent contact point coordinate vector resolved at the previous control period, respectively; , denote the current time instant and the contact surface normal vector resolved at the previous control period, respectively; denotes the length of a control period; is a first hyperparameter for balancing position variations; is a second hyperparameter for balancing attitude variations; denotes the dot product operation of vectors, denotes the vector norm. a preset threshold value of the instability index, in response to the instability index being less than or equal to the threshold value, a controller executing compliant assembly control, and in response to the instability index being greater than the threshold value, the controller executing an adaptive avoidance strategy to ensure stability of the assembly process.

2. The robot flexible assembly method based on assembly state awareness according to claim 1, characterized in that, The method further comprises: in each control period, the system acquires force vectors and torque vectors generated when the robot end effector contacts the workpiece, and uses a moving average filtering method to smooth the data to obtain smoothed force vectors and torque vectors generated when the robot end effector contacts the workpiece. 3.The robot flexible assembly method based on assembly state awareness according to claim 1, wherein, The coordinate vector of the equivalent action point of the contact force and torque sensed by the sensor satisfies the expression: ； wherein, represents a coordinate vector of the equivalent contact point; represents a force vector; represents a moment vector; represents a dot product operation of vectors, represents a vector norm.

4. The robot flexible assembly method based on assembly state awareness according to claim 1, characterized in that, The contact surface normal vector of the robot end effector and the workpiece satisfies the expression: ； wherein, denotes a contact surface normal vector; denotes a candidate normal vector; denotes finding a candidate normal vector that minimizes the value of the expression within the parentheses ; denotes a variance; is any time within a time window ; denotes a force vector acquired at any time within a time window ; denotes a current time; denotes a time window length for calculating variance, denotes a dot product operation of vectors.

5. The robot flexible assembly method based on assembly state awareness according to claim 1, characterized in that, The method further comprises: ； ； wherein, represents a normal force vector of the robot end effector in contact with the workpiece; represents a contact surface normal vector; represents a friction force vector of the robot end effector in contact with the workpiece; represents a force vector; represents a dot product operation of vectors.

6. The robot flexible assembly method based on assembly state awareness according to claim 1, wherein, The method further comprises: The acquisition system collects a large number of historical instability indexes under various normal working conditions for a long time, arranges the historical instability indexes from small to large, sets The quantile is a threshold value.

7. The robot flexible assembly method based on assembly state awareness according to claim 1, characterized in that, The first hyperparameter is set to 1.

0.

8. The robot flexible assembly method based on assembly state awareness according to claim 1, wherein, The second hyperparameter is set to 0.5 m / rad.

9. A robot flexible assembly system based on assembly state awareness, characterized by, The method comprises: a processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a robot compliant assembly method based on assembly state cognition according to any one of claims 1-8.

Images