Continuous control system for industrial robot

Abstract

The invention discloses a continuity control system for an industrial robot, and relates to the technical field of industrial automation, and the system comprises a spraying behavior recording module, an interruption recognition module, a recovery trajectory planning module, an inertia recovery control module, a spraying recovery execution module and a coating uniformity evaluation module. By recording a spraying state before interruption and identifying an interruption event, the system constructs an interruption influence area and recovers a spraying track; an inertia recovery section is arranged before recovery execution, and smooth transition between the motion state and the spraying state of the end effector is achieved; and after spraying is completed, the coating thickness of the area before and after interruption and the optical consistency are compared and analyzed, if local abnormity exists, secondary compensation spraying parameters are generated, and fine correction is completed. The spraying continuity can be effectively improved, the quality risk caused by interruption is reduced, and workpiece waves are avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of industrial automation, in particular to a continuity control system for industrial robots. BACKGROUND

[0002] In modern industrial production, industrial robots face the industrial field, have the characteristics of high precision, high speed, high flexibility, etc., and in the field of industrial automation spraying, industrial spraying robots have become the main equipment in the spraying process because of their high repeatability, stable operation, controllable operation quality, etc.

[0003] Ideally, the robot completes continuous spraying operation according to the preset trajectory and parameters. However, in actual application, due to equipment abnormalities, control system fluctuations, material blockage or safety triggering, etc., the spraying task has the possibility of being accidentally interrupted. Such interruption behavior can occur at any position and at any time, bringing great challenges to the subsequent spraying quality control.

[0004] On the one hand, if the spraying is directly continued after the interruption, it is easy to produce uneven coating thickness, obvious lap joint line, color deviation, rough coating edge, etc. in the interruption area and its adjacent position, which can be seen by the naked eye or physical defects, causing uncertainty of spraying quality.

[0005] On the other hand, in order to avoid the risk of inconsistent spraying, some manufacturing processes choose to directly discard the workpiece and re-spray after interruption, although it can guarantee the quality standard, but it causes material waste, energy consumption increase and production cost rise, especially in high value products or large size workpieces, the influence is more obvious.

[0006] Therefore, when facing the interruption event in the spraying process, the prior art often has difficulty in achieving effective balance between "quality control" and "resource conservation". The lack of recovery control mechanism after spraying interruption has become one of the bottleneck problems commonly existing in current spraying automation production. SUMMARY

[0007] The purpose of the present application is to provide a continuity control system for industrial robots, which has the advantages of effectively avoiding uneven coating thickness or surface defects caused by mutations and improving the overall quality consistency of spraying.

[0008] The above technical purpose of the present application is realized by the following technical scheme:

[0009] A continuity control system for industrial robots, comprising:

[0010] a spraying behavior recording module, configured to collect state parameters for representing spraying behavior in real time during the robot performing the spraying task, the state parameters including but not limited to trajectory information, pose information and spraying output parameters of the end effector, and to construct a time-series dataset for spraying state restoration and thickness calculation;

[0011] an interruption identification module, configured to detect an interruption event during the robot performing the spraying task, determine an interruption time point and an interruption position, and demarcate an interruption influence area according to the interruption time point and the interruption position;

[0012] a recovery trajectory planning module, the recovery trajectory planning module including a trajectory planning unit and a coating state unit, the trajectory planning unit being configured to generate a recovery spraying trajectory for resuming the task in combination with the spraying task plan and the interruption influence area output by the interruption identification module, the recovery spraying trajectory having a starting point at a boundary of the interruption influence area away from the sprayed area; and the coating state unit being configured to calculate theoretical coating thicknesses of positions in the interruption influence area based on the time-series dataset, identify sub-areas with missing spraying or over-thick coating as compensation spraying areas;

[0013] an inertial recovery control module, configured to construct an inertial recovery segment before executing the recovery spraying trajectory, the inertial recovery segment starting from inside the completed spraying area and ending at the starting point of the recovery spraying trajectory, the inertial recovery control module including:

[0014] a motion inertia simulation unit, configured to make the running state of the end effector in the inertial recovery segment smoothly transition to the running state before the interruption according to the state parameters;

[0015] a paint deposition response inertia simulation unit, configured to perform a compensatory spraying operation on the compensation spraying areas when the end effector runs into the interruption influence area, and adjust the spraying parameters to transition to the spraying state before the interruption according to the state parameters before reaching the end point of the inertial recovery segment;

[0016] a spraying recovery execution module, configured to control the end effector to complete the inertial recovery segment and complete the spraying task of the unsprayed area according to the recovery spraying trajectory after completing the inertial recovery segment;

[0017] a coating uniformity evaluation module, configured to analyze the coating thickness and optical consistency in the recovery spraying trajectory, the interruption influence area and the completed spraying area before the interruption after completing the spraying recovery task, and generate secondary compensation spraying parameters and control the robot to make local corrections if local deviations exceeding a preset threshold are detected.

[0018] Further provided is that the spraying behavior recording module includes:

[0019] The state parameter acquisition unit is configured to acquire trajectory information, attitude information and spraying output parameters of the end effector in real time during spraying operation of the robot, wherein the spraying output parameters include spraying flow, spraying angle and spraying opening state.

[0020] The data time sequence organization unit is configured to construct the state parameters into a time sequence data set according to time sequence of acquisition, and perform time synchronization and missing compensation processing on different types of parameters.

[0021] The behavior record management unit is configured to store the constructed time sequence data set of spraying behavior into a preset spraying history database.

[0022] Further, the interruption identification module comprises:

[0023] The interruption event detection unit is configured to monitor control state and output state of the end effector of the robot in real time during spraying operation, and if interruption of spraying trajectory instruction or spraying output abnormal stop event is detected, an interruption event identifier is generated, and an interruption time point and a spatial position of the end effector at the interruption moment are recorded.

[0024] The interruption inertia estimation unit is configured to call state parameters of the end effector within a preset time window before interruption based on the time sequence data set stored in the spraying behavior record module, calculate spraying inertia continuous influence time and spatial delay distance after the interruption event occurs.

[0025] The interruption influence area determination unit constructs a boundary of the interruption influence area according to the interruption position, spraying inertia continuous influence time and spatial delay distance.

[0026] Further, the step of constructing the boundary of the interruption influence area according to the interruption position, spraying inertia continuous influence time and spatial delay distance comprises the following steps:

[0027] According to the spraying inertia continuous influence time and spatial delay distance output by the interruption inertia estimation unit, the spraying action area at the interruption moment and the delay spraying action range possibly deposited due to response lag of paint flight are determined as the interruption response action area.

[0028] According to the state parameters of the end effector before interruption and a preset verification time length, a minimum verification distance of the interruption position in a direction away from the completed spraying area is determined, and the boundary of the interruption response action area is expanded by the minimum verification distance to form a spatial buffer area.

[0029] The interruption influence area is composed of the interruption response action area and the spatial buffer area with the interruption position as a starting point.

[0030] Further, the trajectory planning unit comprises:

[0031] a task plan mapping subunit configured to determine a corresponding progress position of the interruption affected area in a preset spraying task plan according to the interruption affected area output by the interruption identification module;

[0032] a start point determination subunit configured to determine a start point of the resuming spraying trajectory according to the progress position and a coverage range of the interruption affected area on the spraying path, the start point being located at a side boundary of the interruption affected area away from the completed spraying area;

[0033] a trajectory generation subunit configured to generate the resuming spraying trajectory according to the start point and the spraying task plan after the progress position, the resuming spraying trajectory satisfying a continuity requirement of the spraying task plan in spatial position and spraying parameter.

[0034] Further settings: the coating state unit specifically comprises:

[0035] a theoretical coating thickness calculation subunit configured to calculate the theoretical coating thickness of each position in the interruption affected area according to the time series data set recorded in the spraying behavior recording module, in combination with the coating angle and distance of each spatial position in the interruption affected area;

[0036] a thickness difference determination subunit configured to compare the theoretical coating thickness with an expected coating thickness in the spraying task plan, and identify a missed spraying sub-area with a lower expected thickness and an overlapping excessively thick sub-area with a higher expected thickness;

[0037] a compensation area marking subunit configured to mark the missed spraying sub-area and the overlapping excessively thick sub-area identified above with position coordinates and area information, and output to the inertia resuming control module.

[0038] Further settings: the coating state unit further comprises an image-assisted identification subunit, the image-assisted identification subunit being configured to acquire image information of a surface of the interruption affected area through an image sensor, calculate the actual coating thickness of each spatial position in the interruption affected area based on brightness distribution, color saturation or reflection intensity of the image, and correct the theoretical coating thickness output by the theoretical coating thickness calculation subunit to obtain a new theoretical coating thickness.

[0039] Further settings: the motion inertia simulation unit comprises:

[0040] an inertia state prediction subunit configured to construct an acceleration-velocity-position correlation model in a specific time period before interruption based on end effector state parameters in the spraying behavior recording module;

[0041] A transition control subunit is configured to determine a control strategy in the inertia recovery segment according to the pre-interruption acceleration-velocity-position correlation model, and the control strategy specifically includes:

[0042] A start preparation segment starts at the beginning of the inertia recovery segment and ends at the beginning of the interruption affected area, and in the end of the segment, the end effector is controlled according to the dynamic velocity curve generated by the acceleration-velocity-position correlation model to simulate the state before the interruption;

[0043] A compensation transition segment starts at the beginning of the interruption affected area and ends at the beginning of the state fitting segment, and the beginning of the state fitting segment is dynamically determined according to whether the area proportion of the over-thick sub-area in the interruption affected area is lower than a preset threshold; in the compensation transition segment, the movement speed of the end effector is adjusted according to the compensation spraying requirements in different positions in the interruption affected area;

[0044] A state fitting segment starts at the end of the compensation transition segment and ends at the end of the interruption affected area, and gradually makes the movement state of the end effector conform to the acceleration-velocity-position correlation model according to the fitting function.

[0045] Further provided is that the paint deposition response inertia simulation unit includes:

[0046] A spraying requirement perception subunit is configured to receive the compensation spraying area information output by the coating state unit, including the spatial position and compensation type of the missed spraying sub-area and the over-thick sub-area;

[0047] A spraying parameter regulation subunit is configured to execute a differential spraying control strategy in the inertia recovery segment according to the current position of the end effector controlled by the movement inertia simulation unit and the type of the segment, and the spraying control strategy includes:

[0048] In the compensation transition segment, the spraying parameters are adjusted according to the compensation spraying type corresponding to the current position, the spraying flow is increased and the spraying angle is reduced for the missed spraying sub-area to enhance the deposition density, and the spraying flow is reduced and the spraying angle is increased for the over-thick sub-area to reduce the adhesion thickness;

[0049] In the state fitting segment, the spraying parameter regulation subunit calls the pre-interruption spraying output parameters stored in the spraying behavior record module as target values according to the real-time position and speed information of the end effector in the inertia recovery segment, and gradually corrects the spraying output parameters by using an interpolation fitting or segmented control strategy, so that the spraying output parameters are consistent with the pre-interruption spraying state at the end of the state fitting segment.

[0050] In summary, the present application has the following beneficial effects:

[0051] The time sequence dataset of the whole process before interruption is established by the spraying behavior recording module to provide data support for trajectory recovery after interruption, spraying parameter restoration and thickness calculation, and to ensure that the subsequent recovery operation has traceable and quantifiable basis, thereby providing a "data basis" for the system. The interruption event is identified by the interruption identification module, the spraying inertia delay caused by the interruption is estimated, and the spraying path output by the recovery trajectory planning module is combined to realize complete coverage of the interruption affected area, effectively avoiding missing or repeated spraying. The inertia recovery control module smoothes the end motion state through the "motion inertia simulation unit", and adjusts and controls the spraying parameters through the "coating deposition response inertia simulation unit", so that the motion trajectory and the spraying behavior are synchronized in the inertia recovery segment, realizing the double matching of "shape" and "quantity", and ensuring the uniformity of the spraying transition segment.

[0052] The coating state unit identifies the missing or over-thick sub-area and marks the compensation area, and realizes the differential spraying control strategy in the area in cooperation with the spraying parameter adjustment and control sub-unit, and further realizes local accurate compensation through linkage control of the real-time speed and position of the end, thereby significantly improving the coating thickness consistency and material utilization efficiency. The coating uniformity evaluation module analyzes the coating thickness and optical properties after the spraying recovery is completed, compares with the completed area before interruption, identifies whether there is a perceptible deviation in the recovery segment, and if the deviation exceeds the tolerance threshold, triggers secondary repair to realize quality assurance closed loop.

[0053] The modules in the system are not simply connected in series, but form an information closed loop and parameter linkage at the functional level: the spraying behavior data support identification and planning, the identification results drive the recovery path generation, the path and state identification results jointly drive the control unit execution, and the execution effect is fed back by the evaluation module to realize a spraying interruption recovery system with self-sensing, self-adjusting and self-repairing capabilities. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 is the overall structure flow chart of the embodiment. DETAILED DESCRIPTION

[0055] The application will be further described in detail below with reference to the accompanying drawings.

[0056] Embodiment:

[0057] In industrial spraying production process, spraying robots usually execute continuous spraying tasks at constant speed and fixed posture to ensure uniform deposition of coating on the workpiece surface. However, in actual operation, due to electrical failure, safety interlocking, process intervention and other factors, the spraying process may be interrupted unexpectedly. Once the interruption occurs, the spray gun stops moving or stops spraying, at this time, due to the spraying inertia and the robot's own motion inertia, it may still produce uneven coating deposition on the workpiece surface, thereby affecting the spraying quality.

[0058] In addition, if the spraying recovery phase directly restarts the spraying operation from the interruption point or its adjacent area, it is easy to cause dynamic deviation between the spraying trajectory and the state before interruption, resulting in uneven spraying coverage or excessive thickness in the overlapping area, and even obvious color difference, thickness mutation or joint boundary. Therefore, how to reasonably determine the affected area after spraying interruption, and smoothly transition the robot spraying state through the control strategy to realize continuous coating quality is a key technical problem that needs to be solved for industrial spraying control system.

[0059] As shown in Figure 1 The present application proposes a continuity control system for industrial robots, which includes a spraying behavior recording module, an interruption identification module, a recovery trajectory planning module, an inertia recovery control module, a spraying recovery execution module, and a coating uniformity evaluation module, etc. Through dynamic modeling and transition control of the state before and after interruption, the spraying operation can still maintain continuity with the original operation state after interruption recovery, avoiding abnormal coating quality.

[0060] Specifically, during the execution of the spraying operation, the system collects spraying parameters and motion states through the spraying behavior recording module to form a time series data set that can be used for state recovery and coating evaluation; when a spraying interruption event occurs, the interruption identification module locates the interruption time and corresponding position in real time, and delimits the interruption affected area based on the spraying dynamics influence; before the recovery operation, the system re-plans the recovery spraying trajectory based on the interruption information and the original spraying task, and evaluates the coating coverage in the interruption area to identify the area that needs to be compensated for spraying.

[0061] In order to make the recovery spraying state closer to the dynamic process before interruption, the system introduces an inertia recovery section, and performs robot motion inertia simulation and paint deposition inertia compensation operations in this section respectively, so that the spraying state realizes gradual restoration from static to dynamic, and finally smoothly connects to the recovery spraying trajectory, maintaining the consistency and appearance continuity of the coating. After the completion of the recovery spraying task, the system can also analyze the coating quality of the key area, and output compensation parameters for secondary correction operation if necessary.

[0062] In an embodiment of the present application, the spraying behavior recording module is used to collect, organize and store key parameters representing the state of the spraying process in real time during the execution of the spraying task by the spraying robot, so as to provide data support for interruption recovery trajectory planning, inertia recovery control and coating state calculation. The module includes a state parameter acquisition unit, a data time sequence organization unit and a behavior recording management unit.

[0063] The state parameter acquisition unit is configured to acquire state parameters related to the robot end effector in real time during the spraying process, including but not limited to trajectory information, attitude information and spraying output parameters. The trajectory information includes the spatial position, linear velocity and angular velocity of the end effector, which is used to fully characterize the dynamic motion state of the end effector during the spraying process. The attitude information can be the Euler angle or the quaternion attitude representation of the spray gun along three coordinate axes, which is used to reflect the change of the spray gun orientation over time. The spraying output parameters include the spraying flow (the amount of spraying per unit time), the spraying angle (the spraying fan angle) and the spraying opening state (on / off). The above data can be provided in real time by the robot control system feedback, the spray gun control interface and the sensor equipment, and the sampling frequency can be set to 50 Hz to 200 Hz according to the control accuracy requirement.

[0064] The data time sequence organization unit is configured to construct the acquired various state parameters into a spraying behavior time sequence data set according to a unified time axis. Since various parameters can come from different subsystems, there is a sampling delay or data asynchronization problem. This unit adds a unified timestamp to all state parameters and performs synchronization processing and missing compensation by interpolation, alignment and other methods to ensure the consistency and time accuracy of the data.

[0065] The behavior record management unit is configured to store the constructed spraying behavior time sequence data set into a preset spraying history database. The database records the dynamic change process of the spraying operation in a structured data manner and supports indexing management according to the task number, workpiece ID or time period. The database not only provides support for spraying process reconstruction and recovery control, but also provides basic data support for coating thickness estimation, coating continuity analysis and quality traceability and other applications.

[0066] Through the design of the spraying behavior record module, the dynamic behavior characteristics of the end effector in the spraying process can be fully captured and reconstructed, which provides key parameter support for the subsequent modules to realize motion inertia recovery and spraying effect continuity control, and improves the recoverability and spraying consistency guarantee capability of the system in the case of unplanned interruption.

[0067] In one specific embodiment of the application, the interruption identification module is configured to automatically identify an interruption event during the execution of a spraying task by an industrial spraying robot, and reasonably determine the spraying influence area caused by the interruption event on the basis of considering the motion inertia of the robot and the response inertia of the paint flight, so as to provide a complete and accurate spatial range reference for the planning of the subsequent recovery spraying trajectory and the reconstruction of the spraying state. The interruption identification module includes the following three functional units:

[0068] The interrupt event detection unit is responsible for real-time monitoring of the possible unplanned interruption during the spraying robot operation. This unit specifically monitors the following parameters: whether the spraying trajectory instruction is normally issued and executed; whether the spraying state of the spray gun is abnormal, such as zero flow, sudden change of spray angle, sudden change of open state to closed, etc.; whether the controller feedback signal indicates that the motion instruction is paused or interrupted. When any of the above abnormalities is detected, the system determines that the current time T c is the spraying interruption time point, and records the position P c of the end effector at this time, which is considered as the interruption position. c , y c , z c .

[0069] The interrupt inertia estimation unit considers the continuity feature of the spraying operation. After the interruption event occurs, even if the robot end effector stops moving, there may still be two aspects of inertia influence: 1. inertia of the robot body motion; 2. inertia lag effect of the spraying material flight deposition. In order to identify the possible delay impact on the spraying effect after the interruption, this unit calls the historical time series data stored in the spraying behavior recording module with T c as the end point and time length Δt (such as 0.5 seconds). This data set contains the trajectory, pose, speed and spraying output parameters of the end effector. Based on the extracted state parameters, the system calculates the following two key indicators:

[0070] Spraying inertia continuous influence time t d : represents the average length of time that the paint continues to fly and deposit due to particle kinetic energy inertia after the spray gun is turned off. This parameter is obtained by offline experiment or modeling simulation, and is related to the working distance between the robot-controlled spray gun and the workpiece, the paint ejection speed and the paint type.

[0071] Spatial delay distance d f : represents the spatial range covered by the spraying effect within the above t d time, which is jointly affected by the inertia of the robot spray gun and the inertia of the paint particles. In this embodiment, in order to reduce the complexity and computational burden of the system, an empirical approximation estimation of the spraying inertia deposition area is used as a simplified method. If the end effector speed before interruption is v, then:

[0072] df=v·td

[0073] This area is considered as the delay response area where spraying deposition may still occur.

[0074] If the precision is to be improved, physical parameters such as spraying angle, particle flight speed, actual air pressure, etc. are included in the model to construct a two-section jet deposition range model.

[0075] The interruption influence area is composed of an interruption response action area and a spatial buffer area, and the construction process is as follows:

[0076] In combination with the current attitude of the spraying head and the delay response characteristics of the paint spraying, the system first determines the interruption response action area, which covers a range extending from the interruption moment position to the spraying direction by a length, which is the spatial delay distance df, for representing the area possibly affected by the paint deposition under the action of inertia. The paint may still be deposited in this range, and thus it is a potential overlap and over-thickness area or an uneven coverage area.

[0077] To ensure that the end effector has completed the transition of the spraying state before formally entering the recovery spraying trajectory, the system sets an additional verification length along the spraying direction after the interruption response action area to form a spatial buffer area. The length of the area is the minimum verification distance D verify , D verify which is calculated from the average speed of the end effector before interruption and the verification time length preset by the system.

[0078] The spatial buffer area is not used for further adjustment of the spraying parameters, but is used as a state confirmation area for verifying whether the end effector has reached the spraying state before interruption (including trajectory speed, attitude angle, spraying flow, etc.). If the system judges that the parameters in the area have stabilized and meet the pre-interruption standard, it can seamlessly connect to the execution process of the recovery spraying trajectory. Finally, the system concatenates the interruption response action area and the spatial buffer area in space to form the interruption influence area, which is used to guide the execution control of the subsequent compensation spraying and inertia recovery process.

[0079] In the embodiment, the recovery trajectory planning module includes a trajectory planning unit and a coating state unit, which cooperate to realize the spraying trajectory reconstruction for the interruption influence area and the identification of the coating compensation area, and provide accurate input for the subsequent inertia recovery control and spraying recovery execution.

[0080] The trajectory planning unit specifically includes a task plan mapping subunit, a starting point determination subunit, and a trajectory generation subunit.

[0081] Task plan mapping subunit: this subunit is used to map the interruption affected area to the progress dimension of the preset spraying task plan. Specifically, in combination with the spatial boundary and timing information of the interruption affected area output by the interruption identification module, the corresponding spatial segment is searched in the spraying task plan path set, and the serial number position or task node number of the segment in the entire spraying process, i.e. the progress position, is marked. For example, for a rectangular workpiece, if the interruption affected area falls within the eighth parallel spraying track range, the corresponding spraying progress is marked as task node T8. The setting of the task plan mapping subunit can logically correspond the position of the actual interruption to the progress node of the preset spraying task, so that the generation of the recovered spraying track can strictly continue the unfinished task process, avoiding repeated or missed spraying, thereby improving the continuity and accuracy of the spraying path.

[0082] Starting point determination subunit: after the above mapping is completed, the system identifies the starting and ending boundaries of the interruption affected area on the track according to the spatial range of the interruption affected area, in combination with the spraying track direction and span corresponding to the progress position. The boundary of the interruption affected area away from the completed spraying area is taken as the starting point of the recovered spraying track, avoiding direct continuation from the interruption point leading to re-spraying or missed spraying.

[0083] Track generation subunit: this unit constructs a recovered spraying track according to the starting point and subsequent unfinished spraying track information in the task plan. The recovered spraying track is consistent with the spraying task plan in terms of track continuity, spraying parameters (speed, angle, spraying width), etc., ensuring that the spraying path is smoothly connected to the original plan without visible defects.

[0084] The coating state unit includes a theoretical coating thickness calculation subunit, a thickness difference determination subunit, and a compensation region marking subunit. The core goal is to identify the missed spraying and over-thick overlapping sub-regions in the interruption affected area caused by factors such as inertia delay, spraying stop, and parameter change, and to mark them to guide subsequent compensation spraying.

[0085] Theoretical coating thickness calculation subunit: to accurately calculate the theoretical coating thickness of each spatial position in the interruption affected area, the state parameters recorded in the spraying behavior recording module are combined to design the following calculation method:

[0086] In the sampling, the trajectory, speed, attitude, and spraying parameters of the end effector in a time window before the interruption are constructed to form a spraying behavior point set P = {(x i , y i , z i , v i , θ i , φ i , q i )}, where: (x i , y i , zi ) is the spatial coordinate of the spraying point, v i is the terminal velocity, θ i , φ i is the spraying gun attitude angle, q i is the spraying flow. For any spatial point P in the interrupted influence area, based on the Gaussian coverage model or the spraying energy diffusion model, the superimposed contribution of the spraying influence is defined as:

[0087]

[0088] wherein: d i is the spatial distance between the spraying point i and the to-be-estimated point P, α i : the included angle between the spraying direction and the vector; K i : the coating deposition intensity proportional to the spraying flow q i ; σ d , σ α : the weight factors controlling the distance attenuation and the angle attenuation.

[0089] The model fuses the influences of the spatial position, the direction consistency and the spraying flow, and is more physically reasonable in modeling the spraying coverage quality, and is applicable to the estimation scene under the non-ideal deposition state caused by inertial disturbance.

[0090] The thickness difference determination sub-unit carries out difference analysis on the theoretical coating thickness H(P) calculated by the above formula and the target thickness H ref (P)

[0091]

[0092] ΔH(P)=H(P)-H ref (P)

[0093] If ΔH(P)<-δ1, it is determined as a missed spraying area;

[0094] If ΔH(P)>δ2, it is determined as an overlap area with too thick coating;

[0095] wherein δ1, δ2 are preset difference tolerance thresholds.

[0096] ​The compensation area labeling subunit clusters and divides all points meeting the above conditions, extracts the boundary coordinates and area information of the sub-area formed, and generates compensation spraying area description data for calling by the inertia recovery control module, including: compensation area type (omission / over-thickness); area spatial boundary; area center point; recommended compensation strategy label (such as increasing flow / reducing speed through / reducing spray width, etc.). The setting of the theoretical coating thickness calculation subunit enables the system to estimate the theoretical coating thickness of each area based on historical spraying parameters and spatial geometric relationship in the absence of direct thickness measurement. This method not only has non-contact and high adaptability, but also can improve the early identification ability of coating defects.

[0097] Further, to improve the accuracy of coating state determination in the interruption affected area, in the present embodiment, the coating state unit further comprises an image-assisted identification subunit. The image-assisted identification subunit is provided with an image sensor for image acquisition of the spraying result on the surface of the interruption affected area before the spraying robot completes the inertia recovery segment and is about to enter the recovery spraying track. The image sensor can adopt an industrial camera or a multispectral imaging module, and the acquisition content includes the image brightness, color saturation, and reflectivity of the target area surface.

[0098] The image-assisted identification subunit pre-processes the collected image data, including image denoising, geometric correction, and color normalization processing, and then extracts the brightness L, saturation S, and reflectivity R of each image area corresponding to a spatial position, and estimates the actual coating thickness H a (X, Y, Z) based on an empirical model or a machine learning model. The calculation process can be represented as:

[0099]

[0100] where f(.) represents an optical-thickness mapping function obtained by spraying experiment calibration or training, the unit normal vector at the three-dimensional point (used to correct the angle influence); the mapping relationship needs to be established through image-three-dimensional reconstruction or structured light, etc. The actual coating thickness H a (X, Y, Z) estimated by the above estimation is fused with the theoretical value H(P) output by the theoretical coating thickness calculation subunit to form a new theoretical coating thickness H m (x, y, z) to improve the reliability of identifying omitted spraying and overlapped thick areas. The fusion method can adopt a weighted average method or a model self-correction mechanism based on error feedback, such as the expression when adopting the weighted average method:

[0101] H m (X, Y, Z) = a (X, Y, Z) · H a(X, Y, Z) + (1 - a(X, Y, Z)) · H p (X, Y, Z)

[0102] wherein a e [0, 1] represents the confidence weight of image thickness estimation, which can be adaptively adjusted based on image quality indicators.

[0103] Through the above setting, the image-assisted recognition subunit can correct the theoretical model calculation in combination with the actual optical detection results, especially in the case of large theoretical calculation error caused by disturbance in the spraying process, effectively improving the accuracy and consistency of the compensation spraying area recognition, thereby enhancing the reliable recovery capability of the system in complex spraying scenarios.

[0104] The inertia recovery control module is used to control the end effector of the industrial spraying robot to run in a specific path (i.e. inertia recovery section) after the spraying interruption occurs and before the formal resumption of spraying work, to realize the transition adjustment of the spraying trajectory and spraying parameters by simulating the spraying inertia effect, and to avoid obvious thickness mutation or optical difference at the coating joint.

[0105] The inertia recovery control module includes a motion inertia simulation unit and a paint deposition response inertia simulation unit, which are respectively responsible for the coordinated control of the motion state and the spraying output parameters of the end effector. The goal is to ensure that the motion state and the spraying state of the end effector are consistent with those before the interruption before entering the recovery spraying trajectory, thereby improving the continuity and consistency of the coating.

[0106] The motion inertia simulation unit includes:

[0107] Inertia state prediction subunit: This subunit is used to construct an acceleration-velocity-position three-dimensional inertia correlation model based on the time series data of the trajectory speed, acceleration and position of the end effector before the interruption recorded in the spraying behavior recording module. This model can describe the dynamic behavior characteristics of the end effector in the state before the interruption, providing a target reference for the subsequent control strategy. The modeling basis of this subunit is the state parameters collected by the spraying behavior recording module, including: the trajectory position vector p(t) = [x(t), y(t), z(t)] of the end effector, the corresponding running speed vector v(t), and the corresponding running acceleration vector a(t). This subunit first extracts the continuous state parameters in a time window (e.g. AT = 1-3 seconds) before the interruption from the spraying behavior recording module, and constructs the following three-dimensional time series matrix:

[0108]

[0109] wherein t N is the interruption time point, t1 = t N- ΔT. Next, the inertia correlation model of acceleration-velocity-position is constructed by fitting modeling method:

[0110] The function model is fitted for each dimension (x, y, z), for example, using cubic spline fitting, B-spline or first-order difference model, to obtain:

[0111] a x (t) = f x (v x (t), x(t)), a y (t) = f y (v y (t), y(t)), a z (t) = f z (v z (t), z(t))

[0112] That is, each acceleration component is expressed as a function of velocity and position, which is used to describe the dynamic inertia characteristics of the actuator in the actual working state.

[0113] The above three-dimensional relationship is combined to construct a whole inertia model M, which is used to represent:

[0114]

[0115] The inertia correlation model M does not depend on the kinematics theory model assumption, but is directly constructed based on the actual working state data, which can truly reflect the inertia behavior of the actuator under specific working conditions. After construction, the model will be used as a reference template for control strategy in the starting preparation section and state fitting section, which is used to: set the speed target of the end effector; derive the expected speed or acceleration at each position; realize the interpolation fitting matching of the spraying state parameters. Through the design of this subunit, each adjustment behavior in the control process has the "inertia state before interruption" as the target reference, avoiding the fragmentation or distortion of the control strategy, effectively improving the continuity of the recovery section and the stability of the spraying quality.

[0116] Transition control subunit: According to the target state constructed by the inertia correlation model, the inertia recovery section is dynamically divided, and the segmented control strategy is set. The inertia recovery section is divided into three continuous stages:

[0117] Starting preparation section: this stage starts at the starting point of the inertia recovery section and ends at the starting point of the interruption influence area, mainly covering the transition section from the completed spraying area to the unsprayed area.

[0118] The control target is to make the end effector reach a speed state coordinated with the pre-interruption spraying state before entering the compensation transition section, avoiding abnormal coating track or mechanical shaking caused by sudden acceleration or deceleration.

[0119] The speed trajectory of this phase is derived according to the acceleration-speed-position three-dimensional inertial correlation model constructed by the inertial state prediction subunit, and is combined with the current position and the adaptive adjustment of the expected fitting point distance to generate a dynamic speed curve as the control input of this section.

[0120] The compensation transition section starts at the beginning of the interruption affected area and ends at the beginning of the state fitting section, and the end position is dynamically determined according to whether the area proportion of the over-thick lap sub-area in the interruption affected area is lower than the first preset threshold θ1. Due to the dual influence of the mechanical inertia of the end effector and the residual paint spraying inertia at the beginning of the interruption, an over-thick lap sub-area is often formed near the beginning of the interruption affected area, and a missed spraying sub-area is more likely to occur in the direction away from the interruption point. Therefore, this section needs to dynamically adjust the running speed v c (x) of the end effector according to the compensation spraying type of different spatial positions to improve the compensation spraying effect. First, the coating state unit outputs the spatial position x of each sub-area in the interruption affected area and its corresponding compensation type label ("missed spraying" or "over-thick lap"). Based on this, a compensation demand mapping function η(x) is constructed:

[0121]

[0122] In order to realize targeted regulation, the basic running speed is set as v0, and the speed regulation function is introduced:

[0123] V c (x) = V0·(1-β·η(x))

[0124] Where: v c (x) is the running speed of the end effector at position x, v0 is the reference speed set for the compensation transition section (which can be taken as 0.8-0.9 of the speed before interruption); η(x) is the compensation demand label corresponding to the position; β ∈ (0, 0.5): speed adjustment coefficient. According to the control logic:

[0125] When η(x) = +1 (missed spraying): the running speed is reduced, i.e. v c (x) < V0, to increase the coating time per unit area and improve the deposition thickness;

[0126] When η(x) = -1 (over-thick lap): the running speed is increased, i.e. v c (x) > v0, to reduce the spraying time per unit area and reduce further accumulation;

[0127] When η(x) = 0: keep the reference speed.

[0128] The controller calculates v c(x(t)) and sends speed commands to the end effector drive module. Simultaneously, combining historical control feedback, a sliding window filter or PID controller is used to suppress oscillations, ensuring continuous speed variation at the end effector and avoiding severe jitter. Furthermore, to ensure smooth transitions at the boundaries of the compensation region, a buffer function Y(x) can be introduced, such as a Gaussian smoothing factor or a sigmoid function, adjusting the speed regulation function to:

[0129] v c (x)=V0·(1-β·η(x)·γ(x))

[0130] Among them, γ(x) changes dynamically according to the distance from the location to the boundary of the compensation region to avoid sudden velocity changes.

[0131] During the compensation transition section, although the spraying strategy is controlled by the coating deposition response inertial simulation unit, this control module needs to be linked with the speed control strategy of this section to enable the coordinated operation of "position-speed-spraying parameters" to achieve: automatically increase the spraying flow rate / density when the operating speed is low; reduce the spraying flow rate or increase the spraying angle when the operating speed is high; and dynamically adjust the spraying parameters in combination with the operating status of the end effector.

[0132] State alignment segment: This segment begins at the end of the compensation transition segment and ends at the end of the interruption-affected area. The control objective of this segment is to ensure that the actual motion state (velocity, acceleration) of the end effector is consistent with the inertial state model constructed before the interruption, thereby ensuring the continuity of the recovered trajectory.

[0133] First, using the "acceleration-velocity-position three-dimensional inertial correlation model" constructed in the inertial state prediction sub-unit, the time segment T before the interruption is selected. pre (e.g., status data within 1-2 seconds before the interruption):

[0134]

[0135] s i Here, vi is the reference position, and a is the speed at which the end effector operates at that position. i The corresponding acceleration is N; N is the number of interpolation points, determined based on the sampling accuracy of the state-fitting segment division.

[0136] Then, construct the target velocity function v target (s) and the target acceleration function a target (s), using spline interpolation or higher-order polynomial fitting:

[0137]

[0138] Where β k is the fitting coefficient, and m is the fitting order (usually between 2 and 4).

[0139] In the state matching segment, the system acquires the current operating state of the end effector at a fixed sampling period Δt: real-time acquisition of end position s t , velocity v t ; based on the current position s t , the target velocity is calculated:

[0140]

[0141] The current speed deviation is obtained:

[0142]

[0143] In order to make the end effector operating state gradually match the reference state, the controller corrects the speed based on the feedback mechanism:

[0144] The proportional-integral control strategy (PI) is used:

[0145]

[0146] Where: K p is the proportional gain, K i is the integral gain, is the historical cumulative speed error.

[0147] Since the state matching segment still needs to complete the spraying task, and the speed of the end effector is constantly adjusted, in order to ensure that the spraying thickness is consistent before interruption, the paint deposition response inertia simulation unit needs to be guided to adjust the spraying parameters Q t (spraying flow) and a t (angle of spray width), to achieve constant spraying amount per unit path length.

[0148] The end conditions of the state matching segment include: the current sampling point position reaches the end of the interruption affected area, and the current speed deviation meets the preset matching accuracy threshold:

[0149] |Δv t |<ε v , |Δa t |<ε a

[0150] At this time, it is considered that the end effector has completed the motion state matching, and can switch to the normal operation mode of resuming the spraying trajectory.

[0151] This segmented control strategy avoids the sudden change of the spraying path before and after the interruption point in mechanical behavior, which helps the gradual matching of subsequent spraying parameters.

[0152] The paint deposition response inertia simulation unit includes:

[0153] Spraying demand perception subunit: used to receive the compensation spraying area information output by the coating state unit, the area including the missed spraying sub-area and the overlap over-thick sub-area, respectively attached with its coordinate position in space, area size and compensation type, as the input basis of the differentiated spraying control strategy.

[0154] Spraying parameter regulation subunit: according to the end effector position and the corresponding inertia segment type output by the motion inertia simulation unit, dynamically implement the following control strategy:

[0155] In the compensation transition section: according to the compensation spraying type corresponding to the current position, adjust the parameters: if the current area is a missed spraying sub-area, increase the spraying flow Q and reduce the spraying angle α, to improve the deposition density per unit area; if it is an overlap over-thick sub-area, reduce Q and increase α, to slow down the accumulation of adhesion thickness. In order to effectively complete the compensation work of the missed spraying sub-area and the overlap over-thick sub-area in the compensation transition section of the inertia recovery section, the spraying parameter regulation subunit not only adjusts the spraying parameters according to the compensation spraying type (i.e. missed or over-thick) of the current position, but also further combines the real-time running speed v(x) of the end effector controlled by the motion inertia simulation unit to perform the linkage adjustment operation of the spraying parameters. Specifically, the coating deposition effect per unit area is jointly affected by the spraying output parameters (such as spraying flow Q and spraying angle α) and the running speed v(x) of the end effector. If the speed factor is ignored, the adjustment relationship between the spraying flow and the adhesion thickness may be biased, resulting in insufficient or excessive compensation spraying effect. Therefore, in the compensation transition section, the spraying parameter regulation subunit adopts the following linkage regulation model to optimize the parameters:

[0156]

[0157] Where: Q(x) represents the actual spraying flow at the current position x; Q0 represents the standard spraying flow recorded in the spraying operation before interruption; v(x) is the running speed of the end effector at the current position, which is provided by the motion inertia simulation unit in real time, δ m (x) is an indication function of whether the current position belongs to the missed spraying sub-area (1 if it belongs, otherwise 0); δ o (x) is an indication function of whether the current position belongs to the overlap over-thick sub-area, λ1 and λ2 are empirical weight factors, respectively representing the control sensitivity of deposition strengthening and weakening.

[0158] The above control strategy achieves the following goals: when the current position is a missed spraying sub-area (δ m (x) = 1), the system tends to increase the deposition per unit area. If the running speed is fast (v(x) is large) at this time, the spraying flow Q(x) is proportionally increased to ensure the compensation coating thickness; when the current position is an overlap over-thick sub-area (δ o(x) = 1), the system needs to reduce the deposition amount per unit area. If the running speed is slow (v(x) is small), the spray flow Q(x) is further reduced to prevent thickness accumulation; if the current position does not belong to any compensation spray area (δ m (x) = 0, δ o (x) = 0), the spray flow Q(x) returns to the standard spray state Q0.

[0159] In addition, the adjustment of the spray angle a(x) also refers to the above-mentioned logic, and the spray width can be moderately reduced for the missing area to enhance the concentration, and the spray width can be expanded for the over-thick area to reduce the density per unit area, but the specific function expression can be further fitted according to the device spray width response model. In this embodiment, by linking the spray flow and the running speed of the end effector, the spray deposition effect is adjusted adaptively in space during dynamic operation, thereby effectively completing thickness recovery and excess suppression in the compensation transition section, and enhancing the robustness and spray continuity of the system.

[0160] In the state matching section: the end effector runs smoothly in this stage, and this subunit gradually adjusts the current spray parameters to the state before interruption according to the real-time position and speed information. Since the running state of the end effector in this stage gradually transitions to the state before interruption, the spray parameter control subunit needs to meet two goals: ensure that the spray quality is consistent with the state before interruption throughout the state matching section; and make the spray output parameters smoothly transition to the state before interruption at the end of the state matching section.

[0161] The control formula in this embodiment can be expressed as:

[0162]

[0163]

[0164] Wherein, Q(t), a(t) are the spray flow and the spray width angle at the current time t; Q0, a0 are the spray flow and the spray width angle before interruption. v(t) is the real-time speed of the end effector at time t, v0 is the reference speed recorded before interruption, Q comp (t), a comp (t) are temporary compensation spray parameters calculated at the current position and the current speed to maintain the consistency of the coating deposition, which can be obtained by coating thickness back modeling.

[0165] This way realizes the gradual approaching mechanism of the spray parameters under the speed driving, so that: when the initial speed is low, the consistent spray effect is mainly maintained by the compensation parameters; when the speed gradually approaches the speed before interruption, the spray parameters gradually approach the parameters before interruption; and finally the parameter consistency goal is achieved at the end of the section.

[0166] The spraying recovery execution module is responsible for the recovery and continuation of the subsequent spraying task after the spraying interruption event is identified and the inertial recovery control is completed, to ensure the continuity of the overall spraying process and the consistency of the coating. The control flow of this module can be divided into the following two stages:

[0167] 1. Inertial recovery segment execution control stage

[0168] In this stage, the spraying recovery execution module receives the end effector speed control instructions and spraying parameter adjustment instructions output by the inertial recovery control module, controls the robot end effector to move along the planned inertial recovery segment trajectory, and performs compensatory spraying operations as needed. Specifically, it includes:

[0169] Trajectory tracking control: combined with the inertial recovery segment path information output by the recovery trajectory planning module, a position closed-loop control strategy is adopted to adjust the spatial position of the end effector in real time, ensuring trajectory accuracy;

[0170] Spraying parameter linkage control: in the compensatory transition segment and state matching segment within the inertial recovery segment, the system calls the spraying parameter adjustment strategy set in the paint deposition response inertia simulation unit according to the current position and speed of the end effector, dynamically adjusts the spraying flow and spraying angle, and ensures uniform paint deposition and achieves compensation effect;

[0171] End point state verification: at the end point of the inertial recovery segment (i.e. the starting point of the recovery spraying trajectory), the module will judge whether the end effector speed and spraying output parameters meet the requirements of the state before interruption, and if the error is below the preset threshold, it will enter the next stage to ensure that the state matching reaches the recovery requirements.

[0172] 2. Recovery spraying trajectory execution stage

[0173] After completing the inertial recovery segment, the spraying recovery execution module immediately switches to the control task of the recovery spraying trajectory, and continues to execute the remaining spraying task according to the spraying path and parameter plan generated by the recovery trajectory planning module. The specific control logic is as follows:

[0174] Path continuity management: the module starts from the starting point of the recovery spraying trajectory and continuously moves the end effector along the recovery path, ensuring that the spatial path does not jump or break, and achieving smooth continuation of the spraying operation;

[0175] Spraying parameter inheritance and calibration: the spraying parameters (such as spraying flow, angle, etc.) corresponding to the recovery spraying trajectory in the spraying behavior recording module are called as target parameters, and the current spraying output state is monitored in real time. If there is an environmental disturbance or actual running deviation, fine tuning is performed based on feedback control to ensure coating quality;

[0176] Terminal segment check: when approaching the end of the spraying task, the module can also be optionally equipped with a terminal segment quality evaluation mechanism to determine whether there is missing or coating deviation in the terminal area by comparing the theoretical thickness with the actual image recognition data, and triggering a micro compensation strategy if necessary.

[0177] Through the above control process, the spraying recovery execution module can realize seamless spraying continuation from the breakpoint, improve coating uniformity and spraying process reliability, and avoid coating overlap, omission or performance abnormalities caused by interruption.

[0178] The coating uniformity evaluation module is used to compare and analyze the coating quality consistency between the recovery spraying area (including the recovery spraying track and the interruption affected area) and the area that has been sprayed before interruption after the recovery spraying task is completed, to ensure that the interruption of spraying does not have a significant impact on the overall effect of the workpiece. The module includes:

[0179] A coating thickness consistency analysis unit, which obtains: the thickness data of the area that has been sprayed before interruption as a reference benchmark; the actual thickness data of the recovery spraying track and the interruption affected area as a comparison object; through spatial registration, the same structure or feature area in the above area is analyzed for thickness mean value or local difference value:

[0180] ΔT(x, y, z) = |T recovered (x, y, z) - T before (x', y', z')|

[0181] T recovered (x, y, z) is the actual measured thickness of the point of the recovery spraying area, T before (x', y', z') is the actual thickness of the corresponding area before interruption, (x', y', z') is the corresponding point obtained through position transformation or registration, if ΔT > δ T , it is considered that there is a thickness deviation.

[0182] An optical consistency analysis unit, which compares the surface image feature values of the recovery spraying area and the area before interruption, including brightness L, saturation S, and reflectivity R, reflecting the differences in visual effects such as surface color, smoothness, and reflectivity uniformity.

[0183] Δ opt (x, y) = w L ·|L rec -L before |+w s ·|S rec -S before |+w R ·|R rec -R before |

[0184] If Δ opt > δ0, mark as visual deviation area, w L , w S , w R are image feature weight coefficients, which are set according to process requirements. The image acquisition mode can be based on RGB image or structured light scanning device.

[0185] To realize the accurate comparison of the above two modules, the module further includes a region registration subunit: according to the spraying path data and the pose information, the spraying area after recovery and the area before interruption are registered in space; all points exceeding the thickness deviation threshold or the optical deviation threshold are clustered to form a “to-be-compensated sub-region set”.

[0186] If the to-be-compensated sub-region set is not empty, the system automatically generates a compensation spraying trajectory and spraying parameters; the compensation strategy is set according to the deviation type: for insufficient thickness, the spraying flow is increased and the angle is reduced; for color deviation, the spraying material or the coating thickness is corrected; the spraying recovery execution module performs fine local correction spraying according to the compensation trajectory, and closes the loop to repair the differences caused by interruption.

[0187] The time sequence data set of the whole process before interruption is established through the spraying behavior recording module, which provides data support for the trajectory recovery after interruption, the spraying parameter restoration and the thickness calculation, ensures that the subsequent recovery operation has traceable and quantifiable basis, and provides “data basis” for the system. The interruption event is identified by the interruption identification module, the spraying inertia delay caused by interruption is estimated, and the spraying path output by the recovery trajectory planning module is combined to realize complete coverage of the interruption affected area, effectively avoiding omission or repeated spraying. The inertia recovery control module smoothes the end motion state through the “motion inertia simulation unit”, and adjusts and controls the spraying parameters through the “coating deposition response inertia simulation unit”, so that the motion trajectory and the spraying behavior are synchronized in the inertia recovery segment, realizing the double matching of “shape” and “quantity”, and ensuring the uniformity of the spraying transition segment.

[0188] The missing or over-thick sub-regions are identified and the compensation areas are marked by the coating state unit, and the differential spraying control strategy in the region is realized in cooperation with the spraying parameter adjustment and control sub-unit, further combined with the real-time speed and position of the end to realize local accurate compensation, significantly improve the coating thickness consistency and material utilization efficiency. The coating uniformity evaluation module analyzes the coating thickness and optical properties after the spraying recovery is completed, identifies whether there is a perceptible deviation in the recovery segment by comparing with the completed area before interruption, and if the deviation exceeds the tolerance threshold, triggers secondary repair to realize quality assurance closed loop.

[0189] The modules of the system are not simply connected in series, but form an information closed loop and parameter linkage at the functional level: the spraying behavior data support identification and planning, the identification result drives the recovery path generation, the path and the state identification result jointly drive the control unit to execute, and the execution effect is fed back by the evaluation module to form a closed loop, so as to realize a spraying interruption recovery system with self-sensing, self-adjusting and self-repairing capabilities.

[0190] The above-described embodiments do not constitute a limitation on the protection scope of the technical solutions. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the above-described embodiments shall be included in the protection scope of the technical solutions.

Claims

1. A continuous control system for industrial robots, characterized in that, include: The spraying behavior recording module is used to collect state parameters that characterize the spraying behavior in real time during the robot's spraying operation. The state parameters include, but are not limited to, the trajectory information, attitude information and spraying output parameters of the end effector, and construct a time-series dataset for spraying state reconstruction and thickness estimation. The interruption identification module is used to detect interruption events during the robot's spraying operation, determine the interruption time and location, and delineate the area affected by the interruption accordingly. The recovery trajectory planning module includes a trajectory planning unit and a coating status unit. The trajectory planning unit is used to combine the spraying task plan and the interruption influence area output by the interruption identification module to generate a recovery spraying trajectory for recovery operation. Its starting point is located at the boundary of the interruption influence area on the side away from the already sprayed area. The coating state unit is used to calculate the theoretical coating thickness at each location within the interruption-affected area based on the time-series dataset, and to identify sub-regions with missed coating or excessive overlap as compensation coating areas. An inertial recovery control module is used to construct an inertial recovery segment before resuming the spraying trajectory. The inertial recovery segment begins within the already sprayed area and terminates at the starting point of the resumed spraying trajectory. The inertial recovery control module includes: The motion inertia simulation unit is used to smoothly transition the operating state of the end effector during the inertia recovery phase to the operating state before the interruption, based on the state parameters. The coating deposition response inertial simulation unit is used to perform compensatory spraying operation on the compensatory spraying area when the end effector runs into the interruption influence area, and adjust the spraying parameters according to the state parameters before reaching the end of the inertial recovery segment to transition it to the spraying state before the interruption. The spraying recovery execution module is used to control the end effector to complete the inertial recovery segment and then complete the spraying task of the unsprayed area according to the recovery spraying trajectory after the inertial recovery segment is completed; The coating uniformity assessment module is used to analyze the coating thickness and optical consistency of the restored spraying trajectory, the area affected by the interruption, and the area that was sprayed before the interruption after the spraying recovery operation is completed. If a local deviation exceeding a preset threshold is detected, secondary compensation spraying parameters are generated and the robot is controlled to make local corrections.

2. The continuous control system for industrial robots according to claim 1, characterized in that, The spraying behavior recording module includes: The status parameter acquisition unit is used to acquire the trajectory information, attitude information and spraying output parameters of the end effector in real time during the robot's spraying operation. The spraying output parameters include spraying flow rate, spraying angle and spraying on status. The data time-series organization unit is used to construct the state parameters into a time-series dataset according to the acquisition time sequence, and to perform time synchronization and missing compensation processing on different types of parameters. The behavior record management unit is used to store the completed time-series dataset of spraying behavior into a preset spraying history database.

3. A continuous control system for industrial robots according to claim 1, characterized in that, The interruption identification module includes: The interruption event detection unit is used to monitor the control status and output status of the robot end effector in real time during the spraying operation. If the interruption of the spraying trajectory command or the abnormal stop of the spraying output is detected, an interruption event identifier is generated and the interruption time point and the spatial position of the end effector at the moment of interruption are recorded. The interruption inertia estimation unit is used to calculate the duration and spatial delay distance of the spraying inertia based on the time-series dataset stored in the spraying behavior recording module, after the interruption event occurs, by calling the state parameters of the end effector within the preset time window before the interruption. The interruption-affected area determination unit constructs the boundary of the interruption-affected area based on the interruption location, the duration of the spraying inertia effect, and the spatial delay distance.

4. A continuous control system for an industrial robot according to claim 3, characterized in that, The process of constructing the boundary of the interruption-affected area based on the interruption location, the duration of the spraying inertia effect, and the spatial delay distance specifically includes the following steps: Based on the spraying inertia duration and spatial delay distance output by the interruption inertia estimation unit, the spraying action area at the moment of interruption and the delayed spraying action range that may be deposited due to the lag in paint flight response are determined as the interruption response action area. Based on the state parameters of the end effector before the interruption and the preset verification time, the minimum verification distance of the interruption location in the direction away from the completed spraying area is determined, and the boundary of the interruption response area is expanded with the minimum verification distance to form a spatial buffer area. The interruption-affected area starts from the interruption location and is composed of an interruption response area and a spatial buffer area.

5. A continuous control system for an industrial robot according to claim 1, characterized in that, The trajectory planning unit includes: The task planning mapping subunit is used to determine the corresponding progress position of the interruption-affected area in the preset spraying task plan based on the interruption-affected area output by the interruption identification module. The starting point determination subunit is used to determine the starting point for resuming the spraying trajectory based on the progress position and the coverage of the interrupted area on the spraying path. The starting point is located on the side boundary of the interrupted area that is far away from the completed spraying area. The trajectory generation subunit is used to generate a restored spraying trajectory based on the starting point and the spraying task plan after the progress position. The restored spraying trajectory meets the continuity requirements of the spraying task plan in terms of spatial position and spraying parameters.

6. A continuous control system for an industrial robot according to claim 5, characterized in that, The coating state unit specifically includes: The theoretical coating thickness calculation subunit is used to calculate the theoretical coating thickness at each location within the interruption influence area based on the time-series dataset recorded in the spraying behavior recording module, combined with the coating angle and distance at each spatial location within the interruption influence area. The thickness difference determination subunit is used to compare the theoretical coating thickness with the expected coating thickness in the spraying task plan, and identify the missed spraying sub-areas that are lower than the expected thickness and the overlapping sub-areas that are higher than the expected thickness. The compensation area labeling sub-unit is used to label the identified missed spraying sub-regions and overly thick overlapping sub-regions with position coordinates and area information, and output them to the inertial recovery control module.

7. A continuous control system for an industrial robot according to claim 6, characterized in that, The coating state unit also includes an image-assisted recognition subunit, which is used to collect image information of the surface of the interruption-affected area through an image sensor, and calculate the actual coating thickness at each spatial location within the interruption-affected area based on the brightness distribution, color saturation, or reflectivity of the image. The image-assisted recognition subunit is also used to correct the actual coating thickness and the theoretical coating thickness output by the theoretical coating thickness calculation subunit to obtain a new theoretical coating thickness.

8. A continuous control system for an industrial robot according to claim 1, characterized in that, The motion inertial simulation unit includes: The inertial state prediction subunit is used to construct an acceleration-velocity-position correlation model for a specific time period before the interruption, based on the end effector state parameters in the spraying behavior recording module. The transition control subunit is used to determine the control strategy during the inertial recovery segment based on the pre-interruption acceleration-velocity-position correlation model. The control strategy specifically includes: The initial preparation phase begins at the start of the inertial recovery phase and ends at the start of the interruption-affected area. During this phase, the end effector generates a dynamic velocity curve based on the acceleration-velocity-position correlation model to simulate the state before the interruption. The compensation transition section starts at the beginning of the interruption-affected area and ends at the beginning of the state-fitting section. The starting point of the state-fitting section is dynamically determined based on whether the area ratio of the overlapped thick sub-region in the interruption-affected area is lower than a preset threshold. In the compensation transition section, the movement speed of the end effector is adjusted according to the compensation spraying requirements at different positions in the interruption-affected area. The state fitting segment begins at the end of the compensation transition segment and ends at the end of the interruption influence area. The motion state of the end effector is gradually made to conform to the acceleration-velocity-position correlation model according to the fitting function.

9. A continuous control system for an industrial robot according to claim 8, characterized in that, The coating deposition response inertial simulation unit includes: The spraying demand sensing subunit is used to receive the compensation spraying area information output by the coating status unit, including the spatial location and compensation type of the missed spraying sub-area and the overlapping overly thick sub-area. The spraying parameter control subunit is used to execute a differentiated spraying control strategy in the inertial recovery segment based on the current position of the end effector under the control of the motion inertial simulation unit and the segment type. The spraying control strategy includes: Within the compensation transition section, the spraying parameters are adjusted according to the compensation spraying type corresponding to the current position. For the missed spraying sub-areas, the spraying flow rate is increased and the spraying angle is decreased to enhance the deposition density. For the overly thick overlapping sub-areas, the spraying flow rate is decreased and the spraying angle is increased to reduce the adhesion thickness. Within the state-fitting segment, the spraying parameter control subunit uses the real-time position and velocity information of the end effector within the inertial recovery segment to call the spraying output parameters stored in the spraying behavior recording module before the interruption as the target value. It then uses interpolation fitting or segmented control strategies to gradually correct the spraying output parameters so that the spraying output parameters are consistent with the spraying state before the interruption at the end of the state-fitting segment.

Images