Drift correction method, system, device and medium based on digital twinning

Abstract

The application discloses a drift correction method, system, device and medium based on digital twinning, comprising: acquiring perception data and identity data of a target object; associating the perception data with the identity data, establishing a mapping relationship set between the target object and a digital twin; in response to determining that the mapping relationship set has an abnormal mapping relationship, correcting the abnormal mapping relationship and generating a corrected mapping relationship set; determining a specific dimension or a difference position of a state drift between the target object and the digital twin based on the corrected mapping relationship set; and adjusting the state of the digital twin according to the difference position, so that the digital twin and the target object remain synchronized. The application realizes the continuous and accurate synchronization of the digital twin to the target object by establishing and correcting virtual-real mapping relationships, locating the difference position and adjusting the state of the twin.

Description

Technical Field

[0001] This application belongs to the field of digital twin technology, and in particular relates to a drift correction method, system, device and medium based on digital twin. Background Technology

[0002] With the rapid development of digital twin technology, the application of digital twins in fields such as intelligent manufacturing, smart farming, and autonomous driving simulation continues to expand. To achieve accurate mapping between physical entities and digital twins, related technologies generally adopt the method of sensor data acquisition and identity binding to establish a virtual-physical correspondence, and rely on real-time data to drive the twin's state update to maintain consistency between the virtual and the real.

[0003] However, none of the above solutions adequately consider the anomalies in virtual-real mapping caused by individual occlusion, tag misreading, or sensor parameter drift in various scenarios. Mutual occlusion between individuals can lead to identity confusion in visual tracking. Tagging devices are prone to short-term misreading or simultaneous responses from multiple tags in complex environments, resulting in the same identity being bound to multiple trajectories or multiple trajectories sharing the same identity. Furthermore, changes in sensor installation location or clock drift can cause the deviation between perceived data and the state of the digital twin to accumulate continuously. If these anomalies are not corrected in a timely manner, it will be difficult to maintain continuous and accurate synchronization of the digital twin under long-term conditions. Summary of the Invention

[0004] In view of the above, this application provides a drift correction method, system, device and medium based on digital twin to address some of the deficiencies involved in the background art.

[0005] In a first aspect, embodiments of this application provide a drift correction method based on digital twins, including: Acquire the perception data and identity data of the target object; The perceived data is associated with the identity data to establish a mapping relationship set between the target object and the digital twin; In response to determining that there are abnormal mapping relationships in the mapping relationship set, the abnormal mapping relationships are corrected and a corrected mapping relationship set is generated; Based on the corrected mapping set, the differences between the target object and the digital twin are determined; The state of the digital twin is adjusted according to the differences to keep the digital twin synchronized with the target object.

[0006] In a second aspect, embodiments of this application provide a drift correction system based on digital twins, comprising: The data acquisition module is configured to acquire the perception data and identity data of the target object; The relationship building module is configured to associate the perceived data with the identity data to establish a mapping relationship set between the target object and the digital twin; The difference determination module is configured to: determine the difference locations between the target object and the digital twin based on the corrected mapping relationship set; The state synchronization module is configured to adjust the state of the digital twin according to the differences, so that the digital twin is synchronized with the target object.

[0007] In a third aspect, this application provides a computer device including one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, and the one or more programs include instructions for performing the method described in the first aspect.

[0008] In a fourth aspect, this application provides a non-volatile computer-readable storage medium containing a computer program that, when executed by one or more processors, causes the one or more processors to perform the method described in the first aspect.

[0009] In a fifth aspect, this application provides a computer program product comprising one or more computer programs that, when executed by one or more processors, implement the method described in the first aspect.

[0010] The drift correction method, system, device, and medium based on digital twins provided in this disclosure establish a mapping relationship set by acquiring the perception data and identity identification data of the target object, and corrects the mapping relationship when there are abnormalities. Then, based on the corrected mapping relationship, the difference parts are located and the state of the twin is adjusted, realizing the continuous and accurate synchronization of the state of the digital twin with the state of the target object. This effectively solves the problem of virtual and real state drift caused by individual occlusion, identification loss, or parameter drift in multi-target scenarios. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a drift correction method based on digital twins provided in an embodiment of this application is shown. Figure 2 This illustration shows a schematic flowchart of a drift correction method based on digital twins provided in an embodiment of this application; Figure 3 A schematic diagram of the structure of a drift correction system based on digital twin provided in an embodiment of this application is shown; Figure 4 A more specific device hardware structure diagram provided in an embodiment of this application is shown. Detailed Implementation

[0013] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0014] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0015] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0016] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0017] The term "in response to" as used herein refers to a state in which a corresponding event occurs or a condition is satisfied. It will be understood that the timing of subsequent actions performed in response to such event or condition is not necessarily strongly correlated with the time when the event occurs or the condition is met. For example, in some cases, subsequent actions may be performed immediately upon the occurrence of the event or the fulfillment of the condition; while in others, they may be performed some time after the occurrence of the event or the fulfillment of the condition.

[0018] For clarity, before describing the specific technical solutions of the embodiments of this application, the technical terms involved in the embodiments of this application will be explained first.

[0019] Perception data refers to data reflecting the state of a target object collected by sensors deployed in the target area. In this embodiment, the perception data includes, but is not limited to, at least one of image data, sound data, location data, and environmental data.

[0020] Identity identification data refers to data used to uniquely identify a target object. In this application embodiment, the identity identification data includes electronic tag reading records, and the electronic tag includes at least one of ear tags, ankle bracelets, or wearable devices.

[0021] Mapping Relationship Set: Refers to the set of correspondences between target objects and digital twins. In the embodiments of this application, the mapping relationship set contains a one-to-one mapping relationship between each target object and each digital twin, used to represent the corresponding state between the virtual and the real.

[0022] Abnormal mapping relationship: refers to a mapping relationship between a target object and its digital twin that does not conform to a preset correspondence rule. In the embodiments of this application, the abnormal mapping relationship includes multiple digital twins corresponding to the same target object, multiple target objects corresponding to the same digital twin, or the association confidence between the target object and the digital twin being lower than a preset threshold.

[0023] Difference Part: Refers to the specific dimension or location of state drift between the digital twin and the target object. In the embodiments of this application, the difference part includes at least one of identity association confidence deviation, spatial location deviation, morphological parameter deviation, and time axis deviation.

[0024] State synchronization refers to the process of maintaining state consistency between a digital twin and a target object. In the embodiments of this application, state synchronization includes ensuring that at least one of the spatial location, morphological parameters, and time reference of the digital twin is consistent with the target object in real time.

[0025] Digital twin: refers to a virtual mapping entity of a target object in digital space. In the embodiments of this application, the digital twin is used to represent the real-time state of the target object and is updated as the target object's perception data is updated. The state includes at least one of the target object's spatial location, morphological parameters, and time reference.

[0026] As described in the background section, the relevant technologies have many problems in the scenario of digital twin state synchronization.

[0027] First, current virtual-real mapping methods primarily rely on a one-time binding of sensor-collected identification data with visual tracking trajectories, lacking mechanisms for continuous verification of the mapping relationship and anomaly detection. However, in multi-target scenarios, persistent occlusion between target objects can disrupt the visual tracking link, leading to mismatches between electronic identification data and visual trajectories. Identification devices are prone to short-term readouts or simultaneous responses from multiple tags under conditions of metallic interference or dense targets, resulting in the same identity being bound to multiple trajectories or multiple trajectories sharing the same identity. If this mapping anomaly is not promptly identified and corrected, the virtual-real correspondence will remain incorrect, and the status data of each target object will be incorrectly archived, causing chaos in individual profile data.

[0028] Secondly, while existing sensor calibration schemes based on fixed markers or manual calibration can correct perception deviations caused by changes in sensor installation parameters, they are only triggered when there is a significant shift in the sensor's extrinsic parameters. They lack the ability to continuously monitor and actively identify state drift between the digital twin and the target object, and cannot distinguish whether the deviation originates from identity mapping errors, sensor parameter drift, changes in the target object's own shape, or time reference shifts. Existing solutions cannot implement differentiated calibration strategies based on the source of the deviation, resulting in an excessively large calibration range that may disrupt the correct mapping relationship where no anomalies have occurred.

[0029] Furthermore, existing full-scale reconstruction schemes in related technologies reconstruct the mapping relationship between all target objects and digital twins or load the entire 3D scene when mapping anomalies or state drift are detected. However, such schemes involve a large computational scope and high communication volume. When a small number of target objects experience anomalies, full-scale reconstruction will affect the already confirmed correct mapping relationships, and it is impossible to reduce computational overhead by utilizing local corrections.

[0030] Therefore, how to achieve continuous verification and anomaly detection of the virtual-real mapping relationship under multi-objective and long-term conditions, distinguish the specific sources of state drift and perform differentiated local corrections, and avoid the computational burden caused by anomaly propagation and full reconstruction has become a technical problem that urgently needs to be solved in this field.

[0031] Figure 1 A flowchart illustrating a drift correction method based on digital twins provided in an embodiment of this application is shown. Figure 2 The diagram illustrates the framework flow of a drift correction method based on digital twins provided in an embodiment of this application.

[0032] like Figure 1 and Figure 2 As shown, the drift correction method based on digital twins in this disclosure further includes the following steps: S102: Obtain the perception data and identity data of the target object.

[0033] In some optional embodiments, the target object can be an individual livestock or poultry in a farming setting, including any one of pigs, cattle, sheep, or poultry. The sensing data may include, but is not limited to, visible light images, depth images, infrared thermal images, and audio data. This data is acquired by multimodal sensors deployed within the target area and reflects the spatial location, morphological outline, and behavioral state of the target object. Identification data includes electronic tag reading records, whereby the electronic tag includes at least one of ear tags, ankle bracelets, or wearable devices, used to uniquely identify the target object.

[0034] The aforementioned data constitutes the complete information foundation required to establish a virtual-real mapping relationship. Among them, the perception data is used to generate the motion trajectory and state description of the target object, and the identity identification data is used to provide the identity anchor of the target object. Together, they constitute the data foundation for associating the physical target with the digital twin.

[0035] More specifically, perception data can be acquired in real time from visible light cameras, depth cameras, and infrared thermal imagers deployed in the target area. After being formatted and quality-marked by edge computing nodes, it is stored in a time-series database. Identification data can be collected through electronic tag readers. When a target object enters the identification area, the reader reads the electronic tag and records the reading time and location. Edge computing nodes pull data from each device according to a set period, adding both the device timestamp and the edge node's receiving timestamp to each frame of data, and storing it according to spatial region numbers.

[0036] S104: Associate the perceived data with the identity data to establish a mapping relationship set between the target object and the digital twin.

[0037] In this disclosure, the sensing data and identity data obtained through the above steps reflect the motion state and identity information of the target object, respectively. However, a single data source cannot directly establish a correspondence between the target object and its digital twin: the sensing data provides the continuous trajectory of the target object, but cannot determine which specific identity the trajectory corresponds to; the identity data provides the identity of the target object, but can only reflect the reading results at discrete moments and cannot provide continuous motion information. Therefore, it is necessary to correlate the two types of data in the spatiotemporal dimension to establish a complete set of virtual-real mapping relationships.

[0038] In some alternative embodiments, the association process in this disclosure may be implemented based on spatiotemporal consistency constraints.

[0039] Specifically, the system first performs target detection and continuous tracking on the perceived data to generate motion trajectories for each target object. Then, it executes target detection algorithms (such as deep learning-based target detection networks) on visible light or depth images to obtain the bounding boxes and contour information of each target object. Finally, it uses cross-frame tracking algorithms (such as tracking frameworks based on Kalman filtering and Hungarian matching) to associate the same target in consecutive frames, forming the target trajectory under a single camera. Finally, it generates the complete motion trajectory of the target object by stitching together cross-camera trajectories (based on column topology and reachability time constraints).

[0040] The electronic tag reader parses the identity information of each target object and its corresponding spatiotemporal location from the identity data. Each time a read is successful, the electronic tag reader records the unique identity code of the target object, the time of read, and the read location. This location is determined by the spatial area where the reader is located. All read records are arranged in chronological order to form an identity anchor sequence.

[0041] Determine the association weights between each movement trajectory and each piece of identity information. The association weights are determined based on the degree of temporal overlap and spatial accessibility between the movement trajectory and the identity information. Temporal overlap refers to the proximity of the movement trajectory and the identity reading time on the timeline; a high degree of temporal overlap occurs when the target object appears near the reader's location around the identity reading time. Spatial accessibility refers to the distance between the movement trajectory and the identity reading location; a high degree of spatial accessibility occurs when the trajectory point and the reading location are spatially adjacent. Both factors together constitute the association confidence score.

[0042] Based on association weights, motion trajectories are bound to identity information, and a mapping relationship set between target objects and digital twins is established. For each target object, the identity information with the highest association weight is selected and bound to its motion trajectory to establish a one-to-one mapping relationship. The mapping relationships of all target objects constitute the mapping relationship set. The mapping relationship set contains the correspondence between each target object and each digital twin, and is used to represent the corresponding state between the virtual and real worlds.

[0043] S106: In response to determining that there are abnormal mapping relationships in the mapping relationship set, correct the abnormal mapping relationships and generate a corrected mapping relationship set.

[0044] The mapping relationship set established through the above steps is the optimal matching result based on the current perceived data and identity identification data under the constraint of spatiotemporal consistency. However, in multi-target scenarios, mutual occlusion between target objects can interrupt the visual tracking link, resulting in broken trajectory segments corresponding to the same identity or mismatch between identity and trajectory. Identification devices are prone to short-term readouts or simultaneous responses from multiple tags under conditions of metal interference or dense targets, causing the same identity to correspond to multiple trajectory candidates at a certain moment, or multiple identities to correspond to the same trajectory candidate. If the above anomalies are not identified and corrected in time, the mapping relationship will continue to be incorrect, and the state data of each target object will be incorrectly archived.

[0045] Therefore, this application employs an abnormal mapping relationship detection and local correction mechanism.

[0046] In some optional embodiments, the association confidence of each mapping relation in the mapping relation set is first determined. The association confidence, also known as the aforementioned association weight, reflects the reliability of the match between the motion trajectory and the identity information.

[0047] An abnormal mapping relationship is determined to exist in the mapping relationship set when any of the following situations exist: the association confidence of the mapping relationship between the same target object and multiple digital twins is higher than the preset threshold, indicating that the target object is highly matched with multiple candidate identities at the same time, and a unique correspondence cannot be determined; the association confidence of the mapping relationship between the same digital twin and multiple target objects is higher than the preset threshold, indicating that the virtual identity is highly matched with multiple target objects at the same time, and a unique correspondence cannot be determined; the association confidence of any mapping relationship is lower than the preset threshold, indicating that the matching reliability of the mapping relationship is insufficient to support the correspondence between virtual and real.

[0048] After determining the existence of abnormal mapping relationships, the abnormal target objects and abnormal digital twins involved in these relationships are identified. The mapping relationships between target objects and digital twins not involved in abnormal mapping relationships within the mapping relationship set remain unchanged; only the abnormal parts are processed. The association confidence is recalculated for both abnormal target objects and abnormal digital twins to obtain a corrected confidence score. During recalculation, the spatiotemporal search range is expanded to include more candidate trajectories and identity information for comprehensive matching. The mapping relationships between abnormal target objects and abnormal digital twins are corrected based on the corrected confidence score, generating a corrected mapping relationship set.

[0049] S108: Determine the differences between the target object and the digital twin based on the corrected mapping relationship set.

[0050] The mapping relationship set corrected through the above steps has established a correct correspondence between each target object and its digital twin. However, a correct mapping relationship only means that "who corresponds to whom" is correct; it does not mean that the states of the digital twin and the target object are completely consistent. Sensor installation position shifts due to vibration, causing the spatial position of the target object to be incorrectly represented in the twin; the target object grows or deforms over time, causing the morphological parameters of the twin to differ from the actual shape of the target object; sensor clock drift causes the sensed data to be misaligned on the time axis, affecting the accuracy of state comparison. If these deviations are not located, subsequent correction operations will lack a target.

[0051] Therefore, this application employs a multi-dimensional deviation comparison method to determine the areas of difference. Specifically: In some optional embodiments, the target digital twin corresponding to the target object is first determined based on the corrected mapping set. Each target object has one and only one corresponding digital twin in the corrected mapping set.

[0052] The perceptual data of the target object is compared with the current state data of the target digital twin to obtain the deviation comparison results. The deviation comparison results include identity association confidence deviation, spatial position projection deviation, morphological parameter deviation, and temporal axis alignment deviation. Identity association confidence deviation reflects the reliability of the current mapping relationship; spatial position projection deviation is obtained by projecting the three-dimensional position of the digital twin onto the image plane and comparing it with the actual position of the target object in the perceptual data; morphological parameter deviation is obtained by comparing the body size parameters of the digital twin with the actual body size of the target object; and temporal axis alignment deviation is obtained by comparing the temporal misalignment between the perceptual data and the state data of the digital twin.

[0053] Based on the deviation comparison results, the areas of difference between the target object and the target digital twin are determined. These areas of difference are the specific dimensions that exceed the normal range among the four types of deviations mentioned above, and are used to guide subsequent correction operations.

[0054] S110: Adjust the state of the digital twin according to the differences to keep the digital twin synchronized with the target object.

[0055] The differences identified through the above steps reflect the state drift between the digital twin and the target object in specific dimensions.

[0056] However, the sources of deviation differ for different areas, requiring different correction strategies. If the identity association confidence deviation is too high, it indicates a problem with the mapping relationship itself, and the mapping should be re-established. If the spatial location deviation is too high, it indicates that the sensory data acquisition parameters have drifted, and the acquisition parameters should be recalibrated. If the morphological parameter deviation shows a continuous unidirectional trend, it indicates that the target object has undergone natural growth or deformation, and the twin's morphological parameters should be updated. If the time axis deviation continues to shift, it indicates that the time reference of the corresponding data stream has drifted, and the time calibration parameters should be adjusted. Treating all deviations uniformly without distinguishing their sources will result in an excessively large correction range or an incorrect correction direction, potentially damaging the already correct mapping relationship or introducing new deviations.

[0057] Therefore, this application adopts a source discrimination and differentiation correction mechanism.

[0058] First, before adjusting the state of the digital twin, the source of the discrepancy is determined based on the type of deviation. Specifically, when the identity association confidence deviation is higher than a first preset threshold, the source of the discrepancy is determined to be an identity mapping error; when the spatial location deviation is higher than a second preset threshold, the source of the discrepancy is determined to be a drift in the sensory data acquisition parameters; when the morphological parameter deviation shows a continuous unidirectional change trend, the source of the discrepancy is determined to be a change in the shape of the target object; and when the time axis deviation is continuously offset, the source of the discrepancy is determined to be a drift in the sensory data time reference.

[0059] Then, the corresponding state adjustment is performed according to the identified source of the difference. When the source of the difference is an identity mapping error, the mapping relationship between the target object and the digital twin is re-established, that is, the abnormal correction process in S106 is returned to be executed; when the source of the difference is a drift in the sensing data acquisition parameters, the sensing data acquisition parameters are recalibrated, only the external parameters of this device are updated, and the parameters of other devices remain unchanged; when the source of the difference is a change in the shape of the target object, the shape parameters of the digital twin corresponding to the target object are updated, and the shape parameters of the twins of other target objects remain unchanged; when the source of the difference is a drift in the sensing data time base, the sensing data time calibration parameters are adjusted, only the clock offset and drift rate of this data stream are corrected, and the time parameters of other data streams remain unchanged.

[0060] Through the aforementioned differential correction operations, the state drift between the digital twin and the target object is eliminated one by one, maintaining real-time state synchronization between the digital twin and the target object. After correction is completed, the system continuously acquires a new round of perception data and identity data, repeating the above steps to form a continuous synchronization closed loop.

[0061] Compared with related technologies, this application has the following advantages: First, this application establishes a mapping relationship set between the target object and its digital twin by acquiring the target object's perception data and identity data, and performs local correction when mapping anomalies occur. This achieves continuous verification and anomaly detection of the virtual-real mapping relationship in multi-target scenarios, effectively solving the mapping mismatch problem caused by individual occlusion, identifier misreading, or parameter drift in related technologies. By responding to abnormal mapping relationships and only re-associating the target object and digital twin that have anomalies, while keeping the mapping relationship between the target object and digital twin that have not anomalies unchanged, the correction scope is narrowed from global to local, avoiding the computational burden caused by anomaly propagation and full reconstruction. This fundamentally solves the technical defects of related technologies where a small number of anomalies triggering full reconstruction leads to high computational and communication costs and may destroy correctly mapped relationships.

[0062] Second, this application determines the differences between the target object and the digital twin based on the corrected mapping relationship set. It compares deviations across four dimensions: identity association confidence deviation, spatial location projection deviation, morphological parameter deviation, and time axis alignment deviation. This achieves precise localization of state drift between the digital twin and the target object, effectively solving the problem of related technologies lacking the ability to finely discriminate the source of virtual-real state drift. This mechanism quantifies the specific differences between the digital twin and the target object at the state level through multi-dimensional deviation comparison results, providing a clear target for subsequent correction operations. This prevents correction operations from being blind and overcomes the limitation of related technologies that can only detect anomalies but cannot locate specific deviations.

[0063] Third, this application identifies the source of discrepancies based on the type of deviation and distinguishes between four sources: identity mapping errors, drift in perceptual data acquisition parameters, changes in target object morphology, and drift in perceptual data time base. It then performs corresponding differentiated state adjustments for each source, achieving precise correction of the digital twin's state. This effectively solves the problem in related technologies where a uniform correction method is used due to the inability to distinguish the source of deviation, leading to excessively large correction ranges or incorrect correction directions. This mechanism ensures that when there is an identity mapping error, only the mapping relationship is re-established; when there is a drift in acquisition parameters, only the acquisition parameters are recalibrated; when there is a change in the target object morphology, only the corresponding twin morphology parameters are updated; and when there is a drift in the time base, only the time calibration parameters are adjusted. Each source of deviation has a corresponding correction strategy, significantly improving correction accuracy and efficiency.

[0064] Fourth, this application integrates continuous verification of the mapping relationship set, local correction of abnormal mapping relationships, multi-dimensional positioning of discrepancies, and differentiated adjustment of the sources of discrepancies to form a complete closed-loop framework of "association-verification-positioning-correction". Compared with the passive compensation schemes in related technologies that bind once and then do not verify or rely solely on full reconstruction, this application introduces a continuous verification mechanism from the source of mapping establishment. It actively identifies mapping problems through anomaly judgment, compares positioning state differences through multi-dimensional deviation comparison, and then feeds the judgment results back to the differentiated correction stage. The four stages form a positive feedback loop, which effectively suppresses the cumulative risk of virtual-real state drift in long-term operation in multi-target scenarios. Under the premise of ensuring continuous and accurate synchronization between the digital twin and the target object, it minimizes the correction range and maximizes the correction accuracy.

[0065] In summary, this application provides a drift correction method based on digital twins. Through a collaborative mechanism of continuous verification and anomaly correction of the mapping relationship set, and positioning of multi-dimensional difference parts and adjustment of differentiated states, it minimizes the correction range and maximizes the correction accuracy while ensuring continuous and accurate synchronization between the digital twin and the target object. It effectively overcomes the technical difficulties in related technologies where the virtual and real states drift continuously accumulates and is difficult to accurately correct due to individual occlusion, identifier misreading or parameter drift in multi-target scenarios, and has significant engineering application value.

[0066] Based on the same inventive concept, and corresponding to any of the above embodiments, this application also provides a drift correction system based on digital twins.

[0067] Figure 3 A schematic diagram of a drift correction system based on digital twins provided in an embodiment of this application is shown.

[0068] refer to Figure 3 As shown, the drift correction system based on digital twins includes: The data acquisition module 202 is configured to acquire the perception data and identity data of the target object.

[0069] The relationship building module 204 is configured to associate the perceived data with the identity identification data to establish a mapping relationship set between the target object and the digital twin.

[0070] In some optional embodiments, the relationship building module 204 further performs the following steps: The perceived data is subjected to state detection to generate the motion trajectory of the target object; The identity information of the target object and its corresponding spatiotemporal location are parsed from the identity data. Determine the association weight between the movement trajectory and the identity information; wherein, the association weight is determined based on the degree of temporal overlap and spatial accessibility between the movement trajectory and the identity information; Based on the association weights, the motion trajectory is bound to the identity information, and a mapping relationship set between the target object and the digital twin is established.

[0071] The relationship correction module 206 is configured to: in response to determining that there is an abnormal mapping relationship in the mapping relationship set, correct the abnormal mapping relationship and generate a corrected mapping relationship set.

[0072] In some optional embodiments, the relationship correction module 206 further performs the following steps: Determine the association confidence of each mapping relation in the mapping relation set; The mapping relationship set is determined to contain an abnormal mapping relationship if any of the following conditions exist: The association confidence of the mapping relationship between the same target object and multiple digital twins is higher than the preset threshold; The association confidence of the mapping relationship between the same digital twin and multiple target objects is higher than the preset threshold; The association confidence of any mapping relationship is lower than a preset threshold; Identify the abnormal target objects and abnormal digital twins involved in the aforementioned abnormal mapping relationship; The mapping relationships between target objects and digital twins that do not involve abnormal mapping relationships remain unchanged in the aforementioned mapping relationship set; The correlation confidence is recalculated for the abnormal target object and the abnormal digital twin to obtain the corrected confidence. The mapping relationship between the abnormal target object and the abnormal digital twin is corrected based on the corrected confidence level, and a corrected mapping relationship set is generated.

[0073] The difference determination module 208 is configured to determine the difference between the target object and the digital twin based on the corrected mapping relationship set.

[0074] In some optional embodiments, the difference determination module 208 further performs the following steps: Identify the abnormal target objects and abnormal digital twins involved in the aforementioned abnormal mapping relationship; The mapping relationships between target objects and digital twins that do not involve abnormal mapping relationships remain unchanged in the aforementioned mapping relationship set; The correlation confidence is recalculated for the abnormal target object and the abnormal digital twin to obtain the corrected confidence. The mapping relationship between the abnormal target object and the abnormal digital twin is corrected based on the corrected confidence level, and a corrected mapping relationship set is generated. When the identity association confidence deviation is higher than the first preset threshold, the source of the difference in the difference is determined to be an identity mapping error; When the spatial position deviation is higher than the second preset threshold, the source of the difference in the difference is determined to be the drift of the sensing data acquisition parameters; When the deviation of the morphological parameter shows a continuous unidirectional trend, the source of the difference in the difference is determined to be the morphological change of the target object; When the time axis deviation continues to shift, the source of the difference in the difference location is determined to be the drift of the time reference of the sensing data.

[0075] The state synchronization module 210 is configured to adjust the state of the digital twin according to the differences in the parts, so that the digital twin is synchronized with the target object.

[0076] In some optional embodiments, the state synchronization module 210 further performs the following steps: When the source of the difference is an identity mapping error, the mapping relationship between the target object and the digital twin is re-established; When the source of the difference is the drift of the sensing data acquisition parameters, the sensing data acquisition parameters are recalibrated. When the source of the difference is a change in the shape of the target object, update the shape parameters of the digital twin corresponding to the target object; When the source of the difference is the time reference drift of the sensed data, the time calibration parameter of the sensed data is adjusted.

[0077] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0078] The apparatus of the above embodiments can be used to implement the corresponding digital twin-based drift correction method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0079] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the drift correction method based on digital twins described in any of the above embodiments.

[0080] Figure 4 This illustration shows a more specific hardware structure diagram of a computer device provided in this embodiment. The device may include: a processor 810, a memory 820, an input / output interface 830, a communication interface 840, and a bus 850. The processor 810, memory 820, input / output interface 830, and communication interface 840 are interconnected internally via the bus 850.

[0081] The processor 810 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, to execute relevant programs to implement the digital twin-based drift correction method provided in any of the foregoing embodiments of this specification.

[0082] The memory 820 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 820 can store operating devices and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 820 and is called and executed by the processor 810.

[0083] The input / output interface 830 is used to connect input / output modules to enable information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0084] The communication interface 840 is used to connect the communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0085] Bus 850 includes a pathway for transmitting information between various components of the device, such as processor 810, memory 820, input / output interface 830, and communication interface 840.

[0086] It should be noted that although the above-described device only shows the processor 810, memory 820, input / output interface 830, communication interface 840, and bus 850, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0087] The electronic devices described above are used to implement the corresponding digital twin-based drift correction methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0088] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the drift correction method based on digital twins as described in any of the above embodiments.

[0089] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0090] The aforementioned non-transitory computer-readable storage media can be any available medium or data storage device that a computer can access, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0091] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the drift correction method based on digital twin as described in any of the embodiments in the exemplary method section above, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0092] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0093] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0094] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0095] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A drift correction method based on digital twins, characterized in that, include: Acquire the perception data and identity data of the target object; The perceived data is associated with the identity data to establish a mapping relationship set between the target object and the digital twin; In response to determining that there are abnormal mapping relationships in the mapping relationship set, the abnormal mapping relationships are corrected and a corrected mapping relationship set is generated; Based on the corrected mapping relationship set, the differences between the target object and the digital twin are determined, and the differences are the specific dimensions or locations where there is a state drift between the digital twin and the target object; The state of the digital twin is adjusted according to the differences to keep the digital twin synchronized with the target object.

2. The drift correction method based on digital twins according to claim 1, characterized in that, The step of associating the perceived data with the identity data to establish a mapping relationship set between the target object and the digital twin includes: The perceived data is subjected to state detection to generate the motion trajectory of the target object; The identity information of the target object and its corresponding spatiotemporal location are parsed from the identity data. Determine the association weight between the movement trajectory and the identity information; wherein, the association weight is determined based on the degree of temporal overlap and spatial accessibility between the movement trajectory and the identity information; Based on the association weights, the motion trajectory is bound to the identity information, and a mapping relationship set between the target object and the digital twin is established.

3. The drift correction method based on digital twins according to claim 2, characterized in that, The response to determining that there are abnormal mapping relationships in the mapping relationship set includes: Determine the association confidence of each mapping relation in the mapping relation set; The mapping relationship set is determined to contain an abnormal mapping relationship if any of the following conditions exist: The association confidence of the mapping relationship between the same target object and multiple digital twins is higher than the preset threshold; The association confidence of the mapping relationship between the same digital twin and multiple target objects is higher than the preset threshold; The association confidence of any mapping relationship is lower than a preset threshold.

4. The drift correction method based on digital twins according to claim 3, characterized in that, The step of correcting the abnormal mapping relationship and generating a corrected mapping relationship set includes: Identify the abnormal target objects and abnormal digital twins involved in the aforementioned abnormal mapping relationship; The mapping relationships between target objects and digital twins that do not involve abnormal mapping relationships remain unchanged in the aforementioned mapping relationship set; The correlation confidence is recalculated for the abnormal target object and the abnormal digital twin to obtain the corrected confidence. The mapping relationship between the abnormal target object and the abnormal digital twin is corrected based on the corrected confidence level, and a corrected mapping relationship set is generated.

5. The drift correction method based on digital twins according to claim 4, characterized in that, Determining the differences between the target object and the digital twin based on the corrected mapping set includes: Based on the corrected mapping set, determine the target digital twin corresponding to the target object; The perceptual data of the target object is compared with the current state data of the target digital twin to obtain the deviation comparison result; wherein, the deviation comparison result includes at least identity association confidence deviation, spatial position projection deviation, morphological parameter deviation, and time axis alignment deviation; Based on the deviation comparison results, the differences between the target object and the target digital twin are determined.

6. The drift correction method based on digital twins according to claim 5, characterized in that, Before adjusting the state of the digital twin based on the differences, the method further includes: When the identity association confidence deviation is higher than the first preset threshold, the source of the difference in the difference is determined to be an identity mapping error; When the spatial position deviation is higher than the second preset threshold, the source of the difference in the difference is determined to be the drift of the sensing data acquisition parameters; When the deviation of the morphological parameter shows a continuous unidirectional trend, the source of the difference in the difference is determined to be the morphological change of the target object; When the time axis deviation continues to shift, the source of the difference in the difference location is determined to be the drift of the time reference of the sensing data.

7. The drift correction method based on digital twins according to claim 6, characterized in that, Adjusting the state of the digital twin based on the differences includes: When the source of the difference is an identity mapping error, the mapping relationship between the target object and the digital twin is re-established; When the source of the difference is the drift of the sensing data acquisition parameters, the sensing data acquisition parameters are recalibrated. When the source of the difference is a change in the shape of the target object, update the shape parameters of the digital twin corresponding to the target object; When the source of the difference is the time reference drift of the sensed data, the time calibration parameter of the sensed data is adjusted.

8. A drift correction system based on digital twins, characterized in that, include: The data acquisition module is configured to acquire the perception data and identity data of the target object; The relationship building module is configured to associate the perceived data with the identity data to establish a mapping relationship set between the target object and the digital twin; The relationship correction module is configured to: in response to determining that there are abnormal mapping relationships in the mapping relationship set, correct the abnormal mapping relationships and generate a corrected mapping relationship set; The difference determination module is configured to: determine the difference between the target object and the digital twin based on the corrected mapping relationship set, wherein the difference is the specific dimension or location of the state drift between the digital twin and the target object; The state synchronization module is configured to adjust the state of the digital twin according to the differences, so that the digital twin is synchronized with the target object.

9. A computer device, characterized in that, It includes one or more processors, memory; and one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, and the one or more programs include instructions for performing the method of any one of claims 1 to 7.

10. A non-volatile computer-readable storage medium containing a computer program, characterized in that, When the computer program is executed by one or more processors, the one or more processors perform the method according to any one of claims 1 to 7.

Images