A mechanical arm clamping and path planning system based on digital twinning

Abstract

This invention discloses a robotic arm clamping and path planning system based on digital twins, including a digital twin control state construction module, a control association generation module, a clamping change determination module, a local failure control domain determination module, a collaborative incremental planning module, a digital twin joint verification module, and a centralized execution control module. Based on changes in workpiece clamping posture, the system determines the affected path segments and associated control ranges, collaboratively corrects the local robotic arm path and associated equipment control conditions, and verifies the boundary states through digital twin simulation. After successful verification, control data is issued and the control process is updated based on feedback from the physical execution state. This invention can accurately limit the actual impact range of clamping state changes on a given manufacturing process, reduce unnecessary overall replanning, and improve the coordination of multi-device operations.

Description

Technical Field

[0001] This invention relates to the field of industrial automatic control technology, and in particular to a robotic arm clamping and path planning system based on digital twins. Technical Background

[0002] With the development of intelligent manufacturing and flexible production, robotic arms have been widely used in manufacturing processes such as processing, assembly, welding and inspection, and are often used together with fixtures, positioning devices, processing equipment and testing equipment to form collaborative work units.

[0003] Existing manufacturing systems typically generate the robotic arm's motion trajectory in advance when the workpiece is in a predetermined clamping state, and organize each task to operate sequentially according to a set cycle or equipment status. Some systems also combine sensor detection, simulation models, or digital mapping to monitor the status of the robotic arm and workpiece, so as to adjust subsequent control processes in case of operational abnormalities. However, in continuous production, the workpiece may experience slight positional or posture deviations due to changes in clamping load, differences in contact surfaces, local slippage, or station adjustments. Such deviations usually only affect the part of the operation process directly related to the current spatial relationship of the workpiece, but existing control methods often struggle to accurately determine the actual impact range of this state change on subsequent robotic arm movements and other equipment movements. This can easily lead to the replanning of control processes that could have continued to be executed, or the continued use of the original equipment movement relationships after local adjustments, resulting in additional calculations, control waiting times, and mismatches in operation coordination.

[0004] Therefore, the existing multi-equipment collaborative manufacturing control method is still not accurate enough in judging the actual impact range of changes in the workpiece clamping state, which can easily cause unnecessary control adjustments and fluctuations in operation. There is still room for further improvement. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a robotic arm clamping and path planning system based on digital twins to solve or alleviate the technical problems existing in the prior art.

[0006] The technical solution of this invention is implemented as follows: a robotic arm clamping and path planning system based on digital twins includes a digital twin control state construction module, which is used to acquire the workpiece reference pose, the robotic arm reference path, the operating status of at least one associated production equipment and the sequence of equipment actions and release conditions, establish a digital twin control state under a unified coordinate reference, and divide the robotic arm reference path into multiple path segments.

[0007] The control association generation module is used to generate state-dependent control association records for each path segment, recording clamping state references, workpiece relative target pose and allowable range, preceding equipment state conditions, associated equipment control segments and subsequent release conditions.

[0008] The clamping change determination module is used to determine the clamping pose increment based on the current workpiece pose and the workpiece reference pose.

[0009] The local failure control domain determination module is used to transform the target execution pose of the uncorrected path segment to be executed to the current workpiece coordinate system based on the current workpiece pose, determine the path segment that exceeds the corresponding allowable range as the initial failure path segment, and include the subsequent control segment with the initial failure path segment completion state as the entry condition into the local failure control domain according to the state-dependent control association record.

[0010] The collaborative incremental planning module is used to correct the robot arm path within the local failure control domain based on the clamping pose increment, and update the release conditions of the associated equipment control segment with the completion status of the corrected path segment.

[0011] The digital twin joint verification module is used to deduce the boundary output state of the corrected local failure control domain, verify the robot arm motion constraints and equipment control timing, and compare it with the entry conditions of the next uncorrected control segment. If the timing control association is not met, the local failure control domain is iteratively expanded and the collaborative incremental planning is re-executed.

[0012] The centralized execution control module is used to send control data to the robotic arm controller and the associated production equipment controller after the joint verification is passed, and to update the digital twin control status according to the physical execution status.

[0013] Compared with the prior art, the present invention has the following advantages:

[0014] This invention determines the affected robotic arm path based on changes in the workpiece clamping state and limits the actual control range that needs to be adjusted in conjunction with subsequent control relationships. This avoids replanning the entire manufacturing task due to local pose changes and helps reduce unnecessary control data reconstruction.

[0015] This invention synchronously adjusts the control relationship of related production equipment during the local path correction process, so that the changes in the robotic arm path and the actions of related equipment can maintain corresponding connection, reducing the problem of asynchronous execution caused by continuing to use the original control relationship after the local path is adjusted.

[0016] This invention uses digital twins to jointly simulate the modified control process and determines whether the scope of influence should continue to expand based on whether the subsequent control segments still meet the execution conditions. This allows the influence boundary of the clamping state change to be judged according to the actual control state, improving the rationality of determining the local adjustment range.

[0017] After the control data is issued, this invention continues to collect the actual operating status of the robotic arm, workpiece, and related production equipment, and compares it with the corresponding predicted status. Based on the degree of deviation, it selects to continue execution, partially replan, or expand the adjustment range, which is beneficial to maintaining the continuity and control coordination of the multi-equipment manufacturing process. Attached Figure Description

[0018] The accompanying drawings are provided to further understand the present invention and form part of the specification. They are used together with the embodiments of the present invention to explain the invention and do not constitute a limitation thereof.

[0019] Figure 1 This is a schematic diagram of the overall system structure of a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0020] Figure 2 This is a system flowchart of a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0021] Figure 3 This is a schematic diagram of a state-dependent control association record of a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0022] Figure 4 This is a schematic diagram of a multi-device collaborative operation scenario for a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0023] Figure 5 This is a schematic diagram illustrating the clamping pose change and initial failure path segment determination of a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0024] Figure 6 This is a schematic diagram illustrating the formation and iterative expansion of the local failure control domain in a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0025] Figure 7 This is a schematic diagram of collaborative incremental path planning and associated equipment release condition update of a robotic arm clamping and path planning system based on digital twins according to the present invention.

[0026] Figure 8 This is a schematic diagram of the digital twin joint verification and entity execution feedback closed loop of a robotic arm clamping and path planning system based on digital twins according to the present invention. Detailed Implementation

[0027] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0028] Example 1

[0029] See Figures 1-4 This invention provides a robotic arm clamping and path planning system based on digital twins, including a digital twin control state construction module, a control association generation module, a clamping change determination module, a local failure control domain determination module, a collaborative incremental planning module, a digital twin joint verification module, and a centralized execution control module.

[0030] The digital twin control state construction module acquires the workpiece reference pose, robotic arm reference path, operating status of at least one associated production equipment, and the sequence of equipment actions and release conditions corresponding to the current manufacturing task. It converts the data corresponding to different control objects to a unified coordinate reference, establishes the digital twin control state, and divides the robotic arm reference path into multiple path segments that can be released and executed independently.

[0031] The control association generation module generates state-dependent control association records for each path segment, which are used to record the correspondence between the path segment and the workpiece clamping state, the workpiece relative to the target pose, the pose allowable range, the state conditions of the preceding equipment, the associated equipment control segment, and the subsequent release conditions.

[0032] The clamping change determination module acquires the current workpiece pose and determines the clamping pose increment based on the current workpiece pose and the workpiece reference pose. The local failure control domain determination module keeps the target execution pose of the uncorrected path segment to be executed unchanged, transforms it to the current workpiece coordinate system, and determines the initial failure path segment based on the comparison between the current workpiece relative pose and the corresponding workpiece relative target pose and the pose allowable range. Then, it establishes the local failure control domain based on the state-dependent control association record.

[0033] The collaborative incremental planning module only corrects the robot path and associated equipment control segments in the local failure control domain; the digital twin joint verification module extrapolates the corrected local failure control domain, obtains the boundary output state, and determines whether to expand the local failure control domain based on whether the boundary output state can meet the entry conditions of the next uncorrected control segment.

[0034] After the joint verification is passed, the centralized execution control module sends control data to the robotic arm controller and the associated production equipment controller according to the control sequence determined by the control association record. It also updates the digital twin control state based on the actual execution state of the robotic arm, the current workpiece pose, and the actual operating state of the associated production equipment. This forms a continuous control process of clamping state change detection, local failure determination, collaborative incremental planning, joint verification, centralized execution, and entity state feedback.

[0035] Example 2

[0036] This embodiment, based on the above embodiment, describes the process of establishing digital twin control states and state-dependent control association records.

[0037] The digital twin control state construction module first acquires the workpiece reference pose, the robotic arm reference path, the operating status of associated production equipment, and the sequence of equipment actions and release conditions. Spatial state data from different sources are uniformly converted to a pre-established unified coordinate reference, so that the workpiece pose, the robotic arm path target pose, and the subsequent entity execution state can be compared under the same coordinate relationship.

[0038] In one specific implementation, the task coordinate system corresponding to the current manufacturing task is used as the unified coordinate reference. The workpiece reference pose is denoted as:

[0039] ;

[0040] in, Indicates a unified coordinate datum. Indicates the workpiece coordinate system. Used to describe the position and orientation of a workpiece relative to a unified coordinate datum under a reference clamping state.

[0041] The robotic arm's reference path is divided into multiple path segments according to the control phases that can independently determine the entry and completion states. For path segments The target's execution pose under a unified coordinate reference is denoted as:

[0042] ;

[0043] in, This indicates the coordinate system of the robotic arm's end effector.

[0044] Based on the workpiece reference pose and the target execution pose corresponding to the path segment, determine the workpiece relative to the target pose in the reference clamping state for that path segment:

[0045] .

[0046] The workpiece's relative target pose is used to represent a path segment. When executing the reference path of the robotic arm, the position and orientation relationship that the end effector of the robotic arm should maintain relative to the workpiece.

[0047] Set the corresponding pose allowable range for each path segment. The allowable pose range includes at least the allowable range of position components and the allowable range of attitude components, which is used to determine whether the target execution pose of the original path segment still satisfies the workpiece relative execution constraints corresponding to that path segment after the workpiece clamping state changes.

[0048] The path segments and associated equipment control segments are all managed uniformly as control segments. The control association generation module establishes state-dependent control association records for each control segment according to the sequence of equipment actions corresponding to the manufacturing task.

[0049] For any control segment Corresponding state-dependent control association records At least including:

[0050] Control section identifier;

[0051] Clamping status reference;

[0052] Workpiece pose relative to target;

[0053] Permissible pose range;

[0054] Pre-control section identifier;

[0055] Pre-processor status conditions;

[0056] Subsequent control section identifier;

[0057] And the conditions for subsequent release.

[0058] Among them, the clamping status reference item is used to indicate whether the execution of the current control segment is affected by the current workpiece clamping status; the preceding control segment identifier is used to indicate the previous control segment that the current control segment directly depends on; the preceding equipment status condition is used to record the operating status that the associated production equipment needs to meet when the current control segment can enter the execution state; the subsequent control segment identifier is used to determine the next control segment that needs to be checked after the current control segment is completed; and the subsequent release condition is used to record the correspondence between the completion status of the current control segment and the entry condition of the subsequent control segment.

[0059] In this embodiment, the entry conditions for a control segment are determined based on the completion status of the preceding control segment, the status conditions of the preceding equipment, and the workpiece status or robotic arm status actually referenced by the control segment in the state-dependent control association record corresponding to that control segment. For a control segment that does not reference a certain status item, the corresponding status item does not participate in the entry condition judgment of that control segment. When performing the entry condition judgment, the current status of each participating judgment is read from the digital twin control status and compared with the corresponding target status, completion status, or allowable range in the state-dependent control association record; when all participating status items meet the corresponding requirements, it is determined that the entry condition of the control segment is satisfied; if any participating status item does not meet the corresponding requirements, it is determined that the entry condition of the control segment is not satisfied.

[0060] For example, state-dependent control association records can record:

[0061] Path segment The preceding equipment status condition is associated production equipment In a permitted state;

[0062] Path segment The subsequent control section after completion is the associated device control section. ;

[0063] Entry condition reference The completed state.

[0064] Thus, a control data structure is formed in the digital twin control state, with control segments as nodes, the control sequence determined by the identifiers of the preceding and subsequent control segments, and the state transmission relationship determined by the subsequent release conditions.

[0065] The aforementioned state-dependent control association records serve as common data inputs for subsequent local failure control domain determination, collaborative incremental planning, digital twin joint verification, and centralized execution control.

[0066] Example 3

[0067] This embodiment is based on all the above embodiments and describes the process of determining the clamping pose increment, the current workpiece relative pose, and the initial failure path segment.

[0068] Before the path segment to be executed is released, the clamping change determination module acquires the current workpiece pose and converts the current workpiece pose to the unified coordinate reference. The current workpiece pose is denoted as:

[0069] .

[0070] Current workpiece pose relative to workpiece reference pose Perform a relative transformation to obtain the clamping pose increment:

[0071] .

[0072] The clamping pose increment includes position change and attitude change. Position change represents the displacement of the workpiece along the three coordinate directions of a unified coordinate reference, while attitude change represents the rotational change of the current workpiece attitude relative to the reference workpiece attitude.

[0073] After determining the clamping pose increment, the current workpiece pose and the clamping pose increment are written into the digital twin control state.

[0074] During the initial failure path segment determination process, the target execution pose of the uncorrected path segment to be executed remains unchanged. That is, the original robot arm path is not modified based on the clamping pose increment, but the actual relative positional relationship formed by the original robot arm path under the current workpiece state is used as the basis for failure judgment.

[0075] For any uncorrected path segment to be executed Read its target execution pose under a unified coordinate reference. And utilize the current workpiece pose Transform it to the current workpiece coordinate system to obtain the current workpiece relative pose:

[0076] .

[0077] The current workpiece relative pose refers to the position and orientation relationship between the target execution pose and the current actual workpiece, while the robot arm is still executing the target execution pose according to the original target execution pose.

[0078] Read path segment The workpiece relative to the target pose in the corresponding state-dependent control association record and positional allowable range and according to and Determine the relative pose deviation between the two.

[0079] Specifically, position deviations in the three coordinate directions are obtained for the position component, and corresponding attitude deviations are obtained for the attitude component. Each position deviation and attitude deviation is then compared with the allowable pose range. The corresponding allowed values ​​are compared.

[0080] When all position and attitude components are within the allowable range of their respective poses, it indicates that the original target's execution pose still satisfies the path segment requirement relative to the current workpiece. The execution constraints mean that this path segment is not considered as the initial failure path segment corresponding to the current clamping state change.

[0081] When any position component or attitude component exceeds the allowable range of the corresponding pose, path segment Pi is determined as the initial failed path segment.

[0082] For a path segment containing multiple target execution poses, the above transformation and comparison can be performed on each target execution pose in sequence; if at least one target execution pose that needs to satisfy the workpiece relative execution constraint exceeds the pose allowable range, the corresponding path segment will be determined as the initial failure path segment.

[0083] After determining the initial failure path segment, the local failure control domain determination module reads the subsequent control segment identifier and subsequent release conditions from its corresponding state-dependent control association record.

[0084] When the entry condition of a subsequent control segment is based on the completion status of the initial failure path segment, the subsequent control segment is added to the local failure control domain.

[0085] The initial local failure control domain is formed by the initial failure path segment and subsequent control segments that directly depend on its completion state. .

[0086] Therefore, the inputs in this embodiment are the workpiece reference pose, the current workpiece pose, the target execution pose of the uncorrected path segment to be executed, and the corresponding state-dependent control association record; after coordinate transformation, relative pose comparison and control dependency query, the outputs the clamping pose increment, the initial failure path segment and the initial local failure control domain, and provide them to the collaborative incremental planning module.

[0087] Example 4

[0088] This embodiment, based on all the above embodiments, describes the process of performing cooperative incremental planning on the robotic arm path and associated equipment control segment in the local failure control domain.

[0089] The inputs to the collaborative incremental planning module include at least the current workpiece pose, the clamping pose increment, the local failure control domain, the workpiece relative to the target pose for each failure path segment in the local failure control domain, and state-dependent control association records.

[0090] First, the robot arm path is corrected for the failure path segment in the local failure control domain.

[0091] For the failed path segment Read the workpiece relative to the target pose in the reference clamping state:

[0092] .

[0093] Combined with the current workpiece pose The corrected target pose that the path segment should achieve under a unified coordinate reference should be redefined:

[0094] .

[0095] According to the aforementioned definition of clamping pose increment, the current workpiece pose and the workpiece reference pose satisfy the following:

[0096]

[0097] Therefore, the corrected target pose can also be expressed as:

[0098]

[0099] Based on the workpiece's reference pose, the workpiece state is updated to the current clamping state using the clamping pose increment. Then, combined with the workpiece's relative target pose corresponding to the failed path segment, the corrected target pose for that path segment is determined. Thus, the clamping pose increment directly serves as the data input for correcting the target pose of the local failed path segment, and the corrected target pose further participates in subsequent transformations of the robot arm's working coordinate system and inverse kinematics solutions.

[0100] Through the above processing, the corrected target pose is made to satisfy the original workpiece-to-target pose relationship specified in the path segment again with respect to the current workpiece.

[0101] The target pose is then further transformed to the robot arm's working coordinate system. Let the robot arm's working coordinate system be... Relative to a unified coordinate datum The position is The corrected target pose in the corresponding robotic arm working coordinate system is:

[0102] .

[0103] by Using the input of the inverse kinematics solution and combined with the current joint state of the robotic arm, the inverse kinematics solution is performed to obtain the candidate joint positions that can reach the corrected target pose.

[0104] When there are multiple inverse kinematic solutions that satisfy the joint reachability range for the same corrected target pose, a candidate solution that can maintain a small joint state change with the previous trajectory point is selected from the candidate solutions, and candidate joint trajectories are formed according to the execution order of each corrected target pose.

[0105] Then, the joint angles, velocities, and accelerations of the candidate joint trajectories are verified sequentially.

[0106] Let the joint positions of two adjacent trajectory points in the candidate joint trajectory be respectively and The control cycle is Then the corresponding joint velocity can be determined based on the change in joint position between adjacent trajectory points; and the corresponding joint acceleration can be determined based on the change in velocity of adjacent joints.

[0107] The joint angle, joint velocity, and joint acceleration corresponding to each trajectory point are compared with the allowable range of the corresponding joint of the robotic arm.

[0108] Trajectory points that satisfy all motion constraints are directly retained.

[0109] When multiple trajectory points that do not meet the constraints of joint angle, velocity, or acceleration appear consecutively, the consecutive interval is determined as the trajectory interval to be regenerated.

[0110] Using the previous and next valid trajectory points of the trajectory interval to be regenerated as boundaries, intermediate trajectory points are regenerated between them, so that the regenerated trajectory interval is continuous with the uncorrected path segments before and after it in terms of position and velocity.

[0111] In one specific implementation, cubic Hermite interpolation can be used to generate intermediate trajectory points within the trajectory interval to be regenerated. Let the joint positions and joint velocities corresponding to the starting valid trajectory points of the trajectory interval to be regenerated be respectively... and The joint position and joint velocity corresponding to the termination valid trajectory point are respectively and The planning time length of the trajectory interval is At any time within the interval Normalization The corresponding joint position can then be determined based on the starting position, ending position, starting velocity, and ending velocity, so that... The generated trajectory has the same position and velocity as the initial valid trajectory point, and makes... The generated trajectory and the position and velocity of the terminated valid trajectory point are consistent. The interpolated trajectory is discretely sampled according to the robotic arm control cycle to obtain multiple intermediate trajectory points within the trajectory interval to be regenerated. Joint angle, velocity, and acceleration are then verified for each intermediate trajectory point. If trajectory points still do not meet the motion constraints, the planning time length of the trajectory interval is adjusted or the corresponding intermediate trajectory points are regenerated until a corrected trajectory interval that meets the motion constraints is obtained.

[0112] After regenerating intermediate trajectory points, inverse kinematics reachability and joint angle, velocity and acceleration checks are performed again until the trajectory interval meets the corresponding motion constraints.

[0113] After completing the robotic arm path correction, the collaborative incremental planning module continues to process the associated equipment control segments in the local failure control domain.

[0114] For any associated device control segment, read the subsequent release condition in its state-dependent control association record and determine whether the release condition references the completion state of a certain original robotic arm path segment in the local failure control domain.

[0115] If the corresponding robotic arm path segment has been corrected, the completion status of the original robotic arm path segment referenced in the subsequent release condition will be updated to the completion status of the corresponding corrected path segment.

[0116] For example, the original control relationship is:

[0117] Once the completion status is reached, the associated device control section is allowed. implement.

[0118] when Corrected path segments are formed through collaborative incremental planning. In the future, The entry requirements have been updated to:

[0119] Allowed after reaching the corrected completion state implement.

[0120] The completion status of the corrected path segment includes at least the corresponding corrected path segment completion identifier; when the subsequent associated equipment control segment still requires the robotic arm to reach a specific spatial position, the completion status also includes the state that the robotic arm end effector has reached the corrected target pose.

[0121] For robotic arm path segments and associated equipment control segments located outside the local failure control domain, the collaborative incremental planning module does not modify their original control data or release conditions.

[0122] After the above processing, the collaborative incremental planning module outputs:

[0123] Corrected local robotic arm path data;

[0124] Updated release conditions for associated device control section;

[0125] And reference information for external control data that has not been modified for local failure control.

[0126] The above output serves as the input to the digital twin joint verification module.

[0127] Example 5

[0128] This embodiment, based on all the above embodiments, describes the digital twin joint verification of the local failure control domain and the iterative expansion process of the local failure control domain.

[0129] The digital twin joint verification module receives the corrected robotic arm path data and the updated release conditions of the associated equipment control segment output by the collaborative incremental planning module, and reads the current local failure control domain, the completion status of its previous control segment, and the entry conditions of each next uncorrected control segment connected to the local failure control domain.

[0130] First, load the above data into the digital twin control state.

[0131] The state of completion of the previous control segment in the local failure control domain is taken as the starting state of the current simulation process. Each control segment in the local failure control domain is simulated in turn according to the control order determined by the state-dependent control association record.

[0132] For each robotic arm path segment, the corresponding corrected candidate joint trajectory is read, and the position, velocity, and acceleration of the robotic arm joints in the digital twin control state are updated according to the execution order of the trajectory points.

[0133] Each item is checked to determine whether the corresponding joint angle is within the allowable range, whether the joint speed exceeds the allowable speed, and whether the joint acceleration exceeds the allowable acceleration.

[0134] If any motion constraint is not satisfied, the corresponding robotic arm path segment is marked as a control segment that fails the joint verification, and the control segment is returned to the collaborative incremental planning module for correction.

[0135] For the associated device control segment, read the preceding device state conditions and subsequent release conditions from its state-dependent control association record before deriving the control segment.

[0136] First, check whether the completion status of the front-end control segment and the operating status of the associated production equipment in the digital twin control state meet the corresponding front-end equipment status conditions.

[0137] When the conditions are met, the corresponding associated equipment control segment is allowed to enter the simulation, and the operating status of the associated production equipment in the digital twin control state is updated according to the corresponding control data of the associated equipment control segment.

[0138] Each associated device control segment records a waiting state, an execution state, and a completion state in the digital twin control state. When the corresponding control segment has not yet met its entry conditions, the associated device control segment remains in the waiting state. When the completion state of the preceding control segment and the state conditions of the preceding device both meet the requirements of the corresponding state-dependent control association record, the associated device control segment is updated from the waiting state to the execution state, and the digital twin operating state of the associated production equipment is updated according to the control data corresponding to the associated device control segment. When the simulated state reaches the pre-corresponding completion state of the associated device control segment, the associated device control segment is updated to the completion state, and the completion state is written into the digital twin control state as data input for judging the entry conditions of its subsequent control segments. Thus, during the digital twin simulation, the associated device control segment sequentially completes state entry, state execution, and state output according to its actual control dependencies.

[0139] If the conditions are not met, the corresponding associated device control segment is recorded as a control segment that fails the joint verification, and the process is returned to the collaborative incremental planning module to redetermine the corresponding control data or release conditions.

[0140] Once all control segments within the local failure control domain can be simulated according to state-dependent control association records, the simulated state at each control moment is recorded according to a unified control time base during the digital twin joint verification process. For any control moment... The states of the robotic arm, workpiece, and associated production equipment, derived from the corrected robotic arm path data, associated equipment control data, and state-dependent control association records, are determined as the digital twin predicted states corresponding to the control moment, and are denoted as:

[0141]

[0142] The digital twin predicted state includes at least the state items actually referenced by the current control segment's entry condition, completion condition, or subsequent release condition. As each control segment within the local failure control domain is sequentially deduced, the digital twin predicted state corresponding to each control moment is saved according to a unified time base and used as a comparison benchmark for the actual execution state at the same control moment during the physical manufacturing process, further determining the boundary output state of the local failure control domain.

[0143] For each next uncorrected control segment connected to the local failure control domain, the completion state of the previous control segment obtained by digital twin deduction is read, and the robot arm state, workpiece state, associated production equipment state and the completion state of the previous control segment related to the entry condition of the next uncorrected control segment are determined as the corresponding boundary output state.

[0144] Read the entry conditions from the state-dependent control association record corresponding to the next uncorrected control segment, and compare the boundary output state with the entry conditions item by item.

[0145] When the boundary output state satisfies all the entry conditions of the next uncorrected control segment, it is determined that the state influence formed by the change in clamping state and the previous local control correction has re-satisfied the execution conditions of the original control sequence at the connection position. The next uncorrected control segment maintains the original control data and is not added to the local failure control domain.

[0146] If any entry condition is not met, the corresponding next uncorrected control segment will be added to the local failure control domain.

[0147] Then, the identifier of the subsequent control segment in the state-dependent control association record corresponding to the newly added control segment is read, and the subsequent control segments that have a direct control dependency relationship with the newly added control segment and require further inspection of the entry conditions are added to the control set to be inspected.

[0148] The expanded local failure control domain is re-output to the collaborative incremental planning module, which then regenerates the corresponding control data for the newly added robotic arm path segments and associated equipment control segments entering the local failure control domain.

[0149] After completing the new collaborative incremental planning, perform digital twin joint verification again.

[0150] This process repeats in a loop:

[0151] Collaborative incremental planning;

[0152] Local failure control domain derivation;

[0153] Verification of robotic arm motion constraints and equipment control conditions;

[0154] Boundary output state determined;

[0155] Comparison of entry conditions for the next uncorrected control segment;

[0156] And the extension of the local failure control domain,

[0157] This continues until the entry conditions for each of the next uncorrected control segments connected to the current local failure control domain are satisfied by the corresponding boundary output states.

[0158] For example, in a manufacturing task, the robotic arm path segment and the associated equipment control segment form a control dependency sequence P3→M2→P4→M3→P5 according to state-dependent control association records. Among them, the associated equipment control segment M2 takes the completion state of path segment P3 as the entry condition, path segment P4 takes the completion state of M2 as the prerequisite condition, associated equipment control segment M3 takes the completion state of P4 as the entry condition, and path segment P5 takes the completion state of M3 as the prerequisite condition.

[0159] After the current workpiece clamping state changes, if P3 is determined to be the initial failure path segment based on the comparison result between the current workpiece relative pose and the pose allowable range, then P3 is first added to the local failure control domain; since the entry condition of M2 references the completion state of P3, M2 is also added to the local failure control domain.

[0160] After the collaborative incremental planning module completes the path correction of P3 and updates the release conditions of M2, the digital twin joint verification module sequentially deduces the corrected P3 and M2, and obtains the boundary output state after M2 is completed.

[0161] When the boundary output state satisfies the entry condition in the original state-dependent control association record of P4, P4 continues to use the original control data, and the local failure control domain stops expanding before P4.

[0162] When the boundary output state cannot satisfy the entry condition of P4, P4 and M3, which has a direct control dependency relationship with P4, are added to the local failure control domain. Cooperative incremental programming and digital twin joint verification are then re-executed for the expanded local failure control domain. After the derivation of P4 and M3 is completed, the boundary output state corresponding to M3 is compared with the entry condition of P5. If the boundary output state can satisfy the entry condition of P5, then P5 retains its original control data, and the local failure control domain stops expanding before P5; otherwise, the local failure control domain continues to expand backward according to the state-dependent control association record.

[0163] Therefore, the local failure control domain corresponding to the same workpiece clamping state change can include only P3 and M2, or it can be further extended to P4 and M3 based on the boundary output state obtained by digital twin deduction. Its actual control boundary is determined by whether the modified preceding control state can re-satisfy the entry condition of the next unmodified control segment.

[0164] The final output includes the local failure control domain verified by digital twin, the corresponding corrected robotic arm path data, and the associated equipment control data, which are then provided to the centralized execution control module.

[0165] Example 6

[0166] This embodiment, based on all the above embodiments, describes the centralized execution control and entity state feedback process after the local failure control domain is jointly verified by digital twins.

[0167] The centralized execution control module receives locally failed control domains that have passed joint verification, corrected robotic arm path data, associated equipment control data, and state-dependent control association records.

[0168] The control sequence of each control segment within the local failure control domain is determined based on the preceding control segment identifier, subsequent control segment identifier, and subsequent release condition in the state-dependent control association record.

[0169] When the current control segment is the robotic arm path segment, the corresponding corrected robotic arm path control data will be sent to the robotic arm controller.

[0170] When the current control segment is the associated equipment control segment, the completion status of the corresponding upstream control segment and the status conditions of the upstream equipment are checked first. After the corresponding conditions are met, the equipment control data corresponding to the control segment is sent to the associated production equipment controller.

[0171] During the entity control process, the centralized execution control module collects data respectively:

[0172] The actual execution state of the robotic arm;

[0173] Current workpiece pose;

[0174] And the actual operating status of related production equipment.

[0175] The actual execution state of the robotic arm includes at least the current joint state of the robotic arm, the current path segment identifier, and the current path segment execution state.

[0176] The actual operating status of the associated production equipment includes at least the current operating status of the corresponding production equipment and the execution status of the control segment of the associated equipment.

[0177] The status data of different controlled objects each carry the corresponding acquisition time.

[0178] To ensure that different source states can correspond one-to-one with the predicted states in the digital twin control state, the centralized execution control module aligns the execution states of each entity according to a unified time base.

[0179] Specifically, the state time is determined according to a preset unified control cycle. The actual execution state of the robotic arm, the current workpiece pose, and the actual operating state of the associated production equipment collected within the corresponding control cycle are mapped to the same state moment.

[0180] For the same controlled object having multiple sampling states within a unified control cycle, the distance from the state time can be selected. The most recent valid state; for continuous pose data, the state at the corresponding time can also be determined by using two valid sampling results adjacent to the state time.

[0181] After completing the time alignment, the state time will be... The actual execution state of the corresponding robotic arm, the current workpiece pose, and the actual operating state of the associated production equipment constitute the entity execution state at that moment, and is denoted as:

[0182]

[0183] The entity execution state Write the digital twin control state and read the digital twin predicted state corresponding to the same state moment saved during the digital twin joint verification phase. The corresponding state components of the two states are compared according to the actual state items referenced in the current control segment. The comparison result is used to determine whether the completion conditions of the current control segment, the execution conditions within the current local failure control domain, and the entry conditions of the next uncorrected control segment are still satisfied.

[0184] Perform the following processing based on the comparison results.

[0185] In the first case, the entity's execution state satisfies the completion conditions corresponding to the current control segment.

[0186] Set the current control segment to the completed state, and read the subsequent control segment identifier and subsequent release conditions from its state-dependent control association record.

[0187] When the status conditions of the preceding equipment corresponding to the next control segment are met simultaneously, the next control segment is released.

[0188] In the second case, there is a deviation between the entity's execution state and the digital twin's predicted state, and this deviation only causes a change in the execution conditions within the current local failure control domain, while the entry conditions for the next uncorrected control segment outside the current local failure control domain can still be met.

[0189] At this point, the boundary of the current local failure control domain remains unchanged.

[0190] Reacquire the current workpiece pose, redetermine the clamping pose increment based on the current workpiece pose and the workpiece reference pose, and re-input the current local failure control domain into the collaborative incremental planning module.

[0191] The collaborative incremental planning module regenerates the robotic arm path and associated equipment control data in the current local failure control domain, and then performs digital twin joint verification again.

[0192] In the third case, the entity's execution state has rendered the boundary output state corresponding to the current local failure control domain unable to meet the entry conditions for the next uncorrected control segment.

[0193] At this point, the corresponding next uncorrected control segment is added to the local failure control domain, and the subsequent control segment with a direct control dependency relationship with it is determined according to its state-dependent control association record.

[0194] Cooperative incremental planning and digital twin joint verification are performed again for the expanded local failure control domain.

[0195] If the expanded local failure control domain can regenerate a boundary output state that satisfies the entry conditions of the subsequent uncorrected control segment, then the manufacturing task will continue to be executed according to the new local failure control domain.

[0196] Only when the local failure control domain is iteratively expanded and there are still situations where the robot arm motion constraints cannot be met, the state conditions of the upstream equipment of the associated equipment control section cannot be met, or the entry conditions of the next uncorrected control section cannot be restored, the centralized execution control module outputs a manufacturing task-level global replanning trigger signal.

[0197] Therefore, the output of entity state feedback does not uniformly trigger a complete replanning of the manufacturing task, but rather forms separate results based on the actual impact range of the state deviation:

[0198] The result of the next control segment release;

[0199] Current results of local failure control domain replanning;

[0200] Results of local failure control domain expansion;

[0201] Alternatively, it can trigger a task-level global replanning result.

[0202] The feedback results are then re-entered into the digital twin control state, the local failure control domain determination process, or the collaborative incremental planning process, enabling the system to maintain closed-loop updates during the execution of manufacturing tasks.

[0203] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in the present invention, and these should all be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A robotic arm clamping and path planning system based on digital twins, characterized in that, include: The digital twin control state construction module is used to acquire the workpiece reference pose, the robot arm reference path, the operating status of at least one associated production equipment, and the sequence of equipment actions and release conditions. It establishes the digital twin control state under a unified coordinate reference and divides the robot arm reference path into multiple path segments. The control association generation module is used to generate state-dependent control association records for each path segment, recording clamping state references, workpiece relative target pose and allowable range, preceding equipment state conditions, associated equipment control segments and subsequent release conditions. The clamping change determination module is used to determine the clamping pose increment based on the current workpiece pose and the workpiece reference pose. The local failure control domain determination module is used to transform the target execution pose of the uncorrected path segment to be executed to the current workpiece coordinate system based on the current workpiece pose, determine the path segment that exceeds the corresponding allowable range as the initial failure path segment, and include the subsequent control segment with the initial failure path segment completion state as the entry condition into the local failure control domain according to the state-dependent control association record. The collaborative incremental planning module is used to correct the robot arm path within the local failure control domain based on the clamping pose increment, and update the release conditions of the associated equipment control segment with the completion status of the corrected path segment. The digital twin joint verification module is used to deduce the boundary output state of the corrected local failure control domain, verify the robot arm motion constraints and equipment control timing, and compare it with the entry conditions of the next uncorrected control segment. If the timing control association is not met, the local failure control domain is iteratively expanded and the collaborative incremental planning is re-executed. The centralized execution control module is used to send control data to the robotic arm controller and the associated production equipment controller after the joint verification is passed, and to update the digital twin control status according to the physical execution status.

2. The robotic arm clamping and path planning system based on digital twin according to claim 1, characterized in that, Both the path segment and the associated device control segment are considered as control segments; The control association generation module stores the control segment identifier, clamping status reference, workpiece relative target pose, pose allowable range, previous control segment identifier, previous equipment status conditions, subsequent control segment identifier and subsequent release conditions in the form of fields according to the sequence of equipment actions. The subsequent release conditions include the correspondence between the current control segment completion status and the subsequent control segment entry conditions.

3. The robotic arm clamping and path planning system based on digital twin according to claim 1, characterized in that, The clamping change determination module converts the current workpiece pose and the workpiece reference pose to the unified coordinate reference, and obtains the clamping pose increment based on the relative transformation between the two. The clamping pose increment includes the position change and the attitude change. During the initial failure path segment determination process, the execution pose of the uncorrected path segment target to be executed remains unchanged.

4. The robotic arm clamping and path planning system based on digital twin according to claim 3, characterized in that, The local failure control domain determination module transforms the target execution pose of the uncorrected path segment to be executed under a unified coordinate reference to the current workpiece coordinate system, obtains the current workpiece relative pose, and compares it with the workpiece relative target pose and pose allowable range in the corresponding control association record. When the position component or attitude component exceeds the allowable pose range, the corresponding path segment is determined as the initial failure path segment, and the subsequent control segment with the completion status of the initial failure path segment as a prerequisite is added to the local failure control domain.

5. The robotic arm clamping and path planning system based on digital twin according to claim 4, characterized in that, After completing a collaborative incremental planning in the local failure control domain, the digital twin joint verification module deduces the local failure control domain according to the control association record. For each next uncorrected control segment connected to the local failure control domain, the completion state of its previous control segment, obtained through deduction, is determined as the corresponding boundary output state, and compared with the entry condition of the next uncorrected control segment. If any entry condition is not met, the corresponding next uncorrected control segment and subsequent control segments that have direct control dependencies with it are added to the local failure control domain, and the collaborative incremental planning and digital twin joint verification are performed again until the entry conditions of each next uncorrected control segment are met.

6. The robotic arm clamping and path planning system based on digital twin according to claim 5, characterized in that, The collaborative incremental planning module determines the corrected target pose based on the current workpiece pose and the workpiece relative to the target pose in the failure path segment within the local failure control domain. The corrected target pose is then transformed to the working coordinate system of the robotic arm and inverse kinematics is solved to obtain the candidate joint trajectory. The joint angles, velocities, and accelerations of the candidate joint trajectories are verified sequentially. For continuous trajectory intervals that do not meet any motion constraints, intermediate trajectory points are regenerated to ensure that the regenerated trajectory intervals are continuous with the preceding and following uncorrected path segments in terms of position and velocity.

7. The robotic arm clamping and path planning system based on digital twin according to claim 6, characterized in that, The collaborative incremental planning module reads the completion status of the original robotic arm path segment referenced in the subsequent release conditions of the associated device control segment in the local failure control domain, and updates the completion status to the completion status of the corresponding corrected path segment, so that the associated device control segment is triggered by the completion status of the corrected path segment. The robotic arm path segment and associated equipment control segment located outside the local failure control domain retain the original control data and release conditions.

8. The robotic arm clamping and path planning system based on digital twin according to claim 5, characterized in that, The digital twin joint verification module loads the local failure control domain, the completion status of its previous control segment, and the entry conditions of each next uncorrected control segment into the digital twin control state, and sequentially deduces the control segments in the local failure control domain according to the control association record. Verify the joint angle, speed, and acceleration constraints of the robotic arm path, as well as the state conditions of the preceding equipment and the subsequent release conditions of the associated equipment control segment; If any condition is not met, the corresponding control segment will be returned to the collaborative incremental planning module for correction.

9. A robotic arm clamping and path planning system based on digital twins according to claim 8, characterized in that, After the centralized execution control module performs joint verification using digital twins in the local failure control domain, it sends control data to the robotic arm controller and the associated production equipment controller respectively according to the control sequence determined by the control association record. It also collects the actual execution status of the robotic arm, the current workpiece pose, and the actual operating status of the associated production equipment, aligns the collected status of different controlled objects according to a unified time reference, and writes them into the digital twin control status.

10. A robotic arm clamping and path planning system based on digital twins according to claim 9, characterized in that, The centralized execution control module compares the aligned entity execution state with the corresponding predicted state in the digital twin control state; The next control segment is released when the entity's execution state meets the completion conditions of the current control segment. When the execution state of the entity causes a change in the execution conditions within the local failure control domain, the clamping pose increment is redefined and the local failure control domain is replanned and verified. When the entity's execution state causes the entry condition for the next uncorrected control segment to be unsatisfied, the local failure control domain is iteratively expanded, and a manufacturing task-level global replanning is triggered only if there are still unsatisfied joint verification items after iterative expansion.

Images