CN112949174A - Digital twin construction method and device, terminal and storage medium - Google Patents
Digital twin construction method and device, terminal and storage medium Download PDFInfo
- Publication number
- CN112949174A CN112949174A CN202110218736.6A CN202110218736A CN112949174A CN 112949174 A CN112949174 A CN 112949174A CN 202110218736 A CN202110218736 A CN 202110218736A CN 112949174 A CN112949174 A CN 112949174A
- Authority
- CN
- China
- Prior art keywords
- current moment
- probability
- state
- sensor
- observation data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
- G06K7/10019—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers.
- G06K7/10079—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions
- G06K7/10089—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions the interrogation device using at least one directional antenna or directional interrogation field to resolve the collision
- G06K7/10099—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves resolving collision on the communication channels between simultaneously or concurrently interrogated record carriers. the collision being resolved in the spatial domain, e.g. temporary shields for blindfolding the interrogator in specific directions the interrogation device using at least one directional antenna or directional interrogation field to resolve the collision the directional field being used for pinpointing the location of the record carrier, e.g. for finding or locating an RFID tag amongst a plurality of RFID tags, each RFID tag being associated with an object, e.g. for physically locating the RFID tagged object in a warehouse
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Toxicology (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computer Networks & Wireless Communication (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method, a device, a terminal and a storage medium for constructing a digital twin, wherein the method comprises the following steps: acquiring observation data of an observation object acquired by at least one sensor, wherein the at least one sensor comprises one sensor or a plurality of sensors of the same type or a plurality of sensors of different types; based on Bayes principle, calculating posterior probability of current time according to observation data, and obtaining state of observed object at current time by utilizing posterior probability; predicting an event of the observed object at the current moment according to the state and the observation data and a preset rule; digital twins in the construction of observations are based on states and events. By the mode, the method can construct real-time, high-robustness and accurate digital twins and improve the precision of the digital twins.
Description
Technical Field
The invention relates to the technical field of three-dimensional construction, in particular to a method, a device, a terminal and a storage medium for constructing a digital twin.
Background
The digital twin is a simulation process integrating multidisciplinary, multi-physical quantity, multi-scale and multi-probability by fully utilizing data such as a physical model, sensor updating, operation history and the like and completing mapping in a virtual space so as to reflect the full life cycle process of corresponding entity equipment, and the digital twin is applied in numerous fields, and is applied in many fields such as product design, product manufacturing, medical analysis, engineering construction and the like at present.
The existing digital twin construction technology depends on the data quality of the sensor unilaterally, and the sensor data is inaccurate, unreliable and incapable of providing direct information in reality, so that the accurate digital twin cannot be constructed only by relying on simple sampling and rule matching in the prior art. Taking a scene of indoor positioning based on RFID reading as an example, the RFID reading has the problems of misreading, missing reading and the like.
Disclosure of Invention
The invention provides a method, a device, a terminal and a storage medium for constructing a digital twin, which aim to solve the problems of low precision and large error of the conventional digital twin construction.
In order to solve the technical problems, the invention adopts a technical scheme that: provided is a construction method of a digital twin, comprising: acquiring observation data of an observation object acquired by at least one sensor, wherein the at least one sensor comprises one sensor or a plurality of sensors of the same type or a plurality of sensors of different types; based on Bayes principle, calculating posterior probability of current time according to observation data, and obtaining state of observed object at current time by utilizing posterior probability; predicting an event of the observed object at the current moment according to the state and the observation data and a preset rule; digital twins in the construction of observations are based on states and events.
As a further improvement of the present invention, when the number of sensors is one, based on the bayesian principle, the posterior probability of the current time is calculated according to the observation data, including: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment, which are obtained in advance; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the current moment.
As a further improvement of the present invention, when the number of sensors is two or more and the same type of sensor, the posterior probability of the current time is calculated from the observation data based on the bayesian principle, including: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the sensor of the current moment.
As a further improvement of the present invention, when the number of the sensors is two or more and different types of sensors, if the first observation data of the main sensor and the second observation data of the auxiliary sensor are data of an observed object collected from different angles, based on the bayesian principle, the method of calculating the posterior probability of the current time based on the observation data includes: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
As a further improvement of the present invention, when the number of the sensors is two or more and different types of sensors, if the second observation data of the auxiliary sensor is a correction of the first observation data of the main sensor, the posterior probability of the current time is calculated according to the observation data based on the bayesian principle, including: confirming the corrected pose parameters of the main sensor in the first observation data and the reading of the main sensor for the observed object; acquiring a predicted pose parameter, a predicted state of an observed object and a predicted probability of the main sensor at the current moment, wherein the predicted pose parameter, the predicted state and the predicted probability are obtained based on the pose parameter of the main sensor at the previous moment and the state of the observed object; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
As a further improvement of the invention, when the events occurring at adjacent moments of the observed object are independent from each other, the event of the observed object at the current moment is obtained by prediction according to the state and the observed data and according to a preset rule, and the packetComprises the following steps: based on a Bayesian model, a first function alpha (t) and a second function beta (t) are defined according to the relationship among observation data, states and events: α (t) ═ p (x)t,lt,ut|at,rt)；β(t)＝p(xt,lt,ut|at-1,rt-1) (ii) a Wherein t represents time, xtRepresents a state, utRepresents an event,/tIndicating the pose of the sensor itself, rtIndicating the reading of the sensor for the observation, atA reading representing a sensor tag placed at a known fixed location; utilizing the first function and the second function to carry out mutual iteration to obtain an event posterior probability for deducing the state and the event based on the observation data time sequence; and obtaining the event of the observed object at the current moment according to the posterior probability of the event.
As a further improvement of the invention, when events occurring at adjacent moments of the observed object are correlated with each other, the event of the observed object at the current moment is predicted and obtained according to the state and the observation data and a preset rule, and the method comprises the following steps: based on the hidden Markov model, a third function γ (t) is defined according to the state and event at time 0-t:
wherein t represents time, x0:tRepresents a state, utRepresenting an event; iterating the third function to obtain the probability of event inference; and obtaining the event of the observed object at the current moment according to the probability deduced for the event.
In order to solve the technical problem, the invention adopts another technical scheme that: provided is a digital twinning construction apparatus, including: the acquisition module is used for acquiring observation data of an observed object acquired by at least one sensor; the state prediction module is used for calculating the posterior probability of the current moment according to the observation data based on the Bayesian principle and obtaining the state of the observed object at the current moment by utilizing the posterior probability; the event prediction module is used for predicting an event of the observed object at the current moment according to the state and the observation data and a preset rule; and the construction module is used for constructing the digital twins of the observed objects according to the states and the events.
In order to solve the technical problems, the invention adopts another technical scheme that: there is provided a terminal comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions which, when executed by the processor, cause the processor to perform the steps of the method of constructing a digital twin of any of the above.
In order to solve the technical problems, the invention adopts another technical scheme that: there is provided a storage medium storing a program file capable of realizing the digital twin construction method of any one of the above.
The invention has the beneficial effects that: according to the method for constructing the digital twin, after the observation data of at least one sensor is obtained, the observation data of the at least one sensor is fused based on the Bayes principle, the state of an observed object at the current moment is predicted according to the posterior probability at the current moment, then the event of the observed object is deduced, the digital twin of the observed object is constructed according to the state and the event at the current moment, the correlation among the time correlations of the reading of the sensor is fully explored, the state of the observed object and the event causing the state change are subjected to data fusion and reduction, the real-time, high-robustness and accurate digital twin is constructed, and a digital basis is provided for subsequent planning, control and scheduling. Moreover, the construction method of the digital twins is suitable for all scenes needing to fuse the observation data of the sensor and construct the digital twins, and is wide in applicability.
Drawings
FIG. 1 is a schematic flow diagram of a method of construction of a digital twin according to an embodiment of the invention;
FIG. 2 is a functional block diagram of a digital twinning building apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a terminal according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first", "second" and "third" in the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of the feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise. All directional indicators (such as up, down, left, right, front, and rear … …) in the embodiments of the present invention are only used to explain the relative positional relationship between the components, the movement, and the like in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indicator is changed accordingly. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
Fig. 1 is a schematic flow chart of a digital twinning construction method according to an embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in fig. 1 if the results are substantially the same. As shown in fig. 1, the method comprises the steps of:
step S101: observation data for an observation acquired by at least one sensor is acquired, the at least one sensor including one sensor or a plurality of sensors of the same type or a plurality of sensors of different types.
In step S101, one or more sensors may be provided, and the types of the sensors may be the same or different, for example, the position parameter of the moving object, i.e., the observed object, and the position parameter of the moving object, i.e., the observed data, are obtained indoors through the RFID sensor. In some scenarios, more than one sensor may also be included, for example, by both an RFID sensor and a camera to observe the motion of an object.
Step S102: based on Bayes principle, the posterior probability of the current moment is calculated according to the observation data, and the state of the observed object at the current moment is obtained by utilizing the posterior probability.
Further, the data fusion can be divided into time sequence data fusion of a single sensor, time sequence \ mixed data fusion of a plurality of sensors of a single type sensor, and time sequence \ mixed data fusion of a plurality of sensors of a plurality of types of sensors. The specific data fusion process is as follows:
when the number of the sensors is one, based on the Bayesian principle, the posterior probability of the current moment is calculated according to the observation data, and the method comprises the following steps:
1.1, acquiring a prediction state and a prediction probability of the current time based on the state prediction of the previous time, which are obtained in advance.
Specifically, after the state at the previous time is obtained, the state at the next time is predicted by using the state at the previous time, and the prediction probability of each prediction state is obtained, and the prediction probability is recorded as: p (x)t|xt-1)。
And 1.2, calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment.
Specifically, the calculation process of the prior probability is a prediction stage, and the calculation formula of the prior probability is as follows:
wherein, t is the time of day,
is the prior probability of the current time, p (x)t-1|z0,....,t-1) Is the posterior probability of the last moment, ztIs the observed data of the sensor.
And 1.3, calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and a pre-acquired likelihood function of the current moment.
Specifically, the calculation process of the posterior probability is an updating stage, and after the prior probability is obtained, the calculation formula of the updated posterior probability is as follows:
wherein, p (x)t|z0,....,t) For posterior probability, η is the normalization coefficient, p (z)t|xt) Is to assume the state as xtA likelihood function of the sensor readings is observed for the situation, which is fit to each specific sensor type and model, and the experimental data. For example, in a scenario where the indoor positioning of an object is performed by an RFID sensor, state xtThe actual position of the object (attached with an RFID tag), the observation data z of the sensortLikelihood function p (z) for whether RFID tag is read given RFID sensor poset|xt) The likelihood function value is larger in the RFID radio frequency wave beam coverage range and smaller outside the radio frequency wave beam range. In this embodiment, the radio frequency coverage of the sensor is not limited.
In the embodiment, the posterior probability is obtained based on the observation data of all the sensors at the time t and before through iteration, and even if the RFID sensor has large deviation in self positioning, the method can still perform indoor centimeter-level accurate positioning on the observed object within a certain numerical value.
Secondly, when the number of the sensors is two or more and the same type of sensors, based on the Bayesian principle, the posterior probability of the current moment is calculated according to the observation data, including:
and 2.1, acquiring the prediction state and the prediction probability of the current moment based on the state prediction of the previous moment.
And 2.2, calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment.
And 2.3, calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the sensor of the current moment.
Specifically, in this embodiment, the prediction stage and the update stage of the state are:
wherein the content of the first and second substances,
is observed data of different sensors, wherein
Are used as the different sensors for the different types of sensors,
is the likelihood function of the kth sensor.
Thirdly, when the number of the sensors is two or more and different types of sensors, if the first observation data of the main sensor and the second observation data of the auxiliary sensor are the data of the observed object collected from different angles, based on the bayesian principle, calculating the posterior probability of the current moment according to the observation data, including:
and 3.1, acquiring the prediction state and the prediction probability of the current moment based on the state prediction of the previous moment.
And 3.2, calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment.
And 3.3, calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
Specifically, when the number of sensors is two or more and different types of sensors, the first observation data of the master sensor is taken as ztRecording the second observation data of the auxiliary sensor as atFor example, when atThe observation object is identified and tracked through monitoring video, ztThe radio frequency signal of an observed object is detected by an RFID sensor, the first observation data of the main sensor and the second observation data of the auxiliary sensor are data of the observed object collected from different angles, and at the moment, the prediction stage and the updating stage of the state are as follows:
wherein, p (a)t|xt) Is the likelihood function of the secondary sensor.
Fourthly, when the number of the sensors is two or more and different types of sensors, if the second observation data of the auxiliary sensor is the correction of the first observation data of the main sensor, based on the Bayesian principle, the posterior probability of the current moment is calculated according to the observation data, including:
and 4.1, confirming the corrected pose parameters of the main sensor in the first observation data and the reading of the main sensor for the observed object.
And 4.2, acquiring the predicted pose parameter, the predicted state of the observed object and the predicted probability of the main sensor at the current moment, which are obtained based on the pose parameter of the main sensor at the previous moment and the state prediction of the observed object.
And 4.3, calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment.
And 4.4, calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
In particular, if the second observation data atIs to the primary sensor first observation data ztCorrection of, e.g. zt＝(lt，rt)，ltIs the pose of the RFID sensor itself, rtIs the reading of the primary sensor for the observed object, atIs a reading of a reference RFID tag placed at a known fixed location, at which time atActually play a role in zt＝(lt，rt) Self pose l intThe correcting action is not directly applied to the state x of the observed object to be estimatedtAt this time, the prediction stage and the update stage of the state are respectively as follows:
wherein the content of the first and second substances,
is a likelihood function of the observed reading given the pose of the sensor and the position of the observed object, p (a)t|lt) Is a likelihood function of readings for reference tags of known fixed positions given the pose of the sensor.
The embodiment performs hybrid fusion between time and different types on the observation data of different types of sensors, and fully utilizes the acquired information to deduce the state.
In some embodiments, after obtaining the posterior probability of the observed object, a corresponding state with the highest probability value can be selected as the observed state at the current time (i.e., the maximum posterior criterion), or an expected value based on the posterior probability can be used as the estimated value of the current observed state.
Step S103: and predicting the event of the observed object at the current moment according to the state and the observation data and a preset rule.
Further, the events occurring at adjacent times of observation may be independent of each other or may be correlated with each other. Specifically, the method comprises the following steps:
when the events occurring at adjacent moments of the observed object are independent from each other, the event of the observed object at the current moment is predicted according to the state and the observation data and a preset rule, and the method comprises the following steps:
1.1, defining a first function alpha (t) and a second function beta (t) according to the relation among observation data, states and events based on a Bayesian model: α (t) ═ p (x)t，lt，ut|at，rt)；β(t)＝p(xt，lt，ut|at-1，rt-1) (ii) a Wherein t represents time, xtRepresents a state, utRepresents an event,/tIndicating the pose of the sensor itself, rtIndicating the reading of the sensor for the observation, atIndicating readings of sensor tags placed at known fixed locations.
1.2, mutually iterating the first function and the second function to obtain the event posterior probability for deducing the state and the event based on the observation data time sequence.
And 1.3, obtaining the event of the observed object at the current moment according to the posterior probability of the event.
Specifically, by defining a first function α (t) and a second function β (t):
α(t)＝p(xt，lt，ut|at，rt)；
β(t)＝p(xt，lt，ut|at-1，rt-1)；
iterating α (t) and β (t):
α(t)＝p(xt，lt，ut|at，rt)＝p(xt，lt，ut|at，rt，at-1，rt-1)＝
ηp(at，rt|xt，lt，ut，at-1，rt-1)p(xt，lt，ut|at-1，rt-1)＝
ηp(al|lt)p(rt|lt，xt)β(t)。
through mutual iteration of alpha (t) and beta (t), the posterior probability of deducing states and events based on the observation data time sequence is obtained.
When the events occurring at the adjacent moments of the observed objects are correlated, predicting the event of the observed object at the current moment according to the state and the observation data and a preset rule, wherein the event comprises the following steps:
2.1, based on the hidden Markov model, defining a third function gamma (t) according to the state and the event at the time 0-t:
wherein t represents time, x0：tRepresents a state, utRepresenting an event.
And 2.2, iterating the third function to obtain the probability of event inference.
And 2.3, obtaining the event of the observed object at the current moment according to the probability deduced for the event.
In this embodiment, global inference of all states and events in a time period from 0 to t is performed based on all observation readings at time from 0 to t, specifically, a viterbi decoding algorithm may be adopted, and also, incremental update of the states and events may be performed at time from specific t, where the incremental learning method is as follows:
definition of
Iterate γ (t):
the inference for event u (t) is:
step S104: digital twins in the construction of observations are based on states and events.
Specifically, after the state and the event of the observed object are obtained, the motion state of the observed object is restored in real time in a pre-established three-dimensional scene, and the digital twin of the observed object is established.
According to the construction method of the digital twin, after the observation data of at least one sensor is obtained, the observation data of the at least one sensor is fused based on the Bayes principle, the state of an observed object at the current moment is predicted according to the posterior probability of the current moment, then the event of the observed object is deduced, the digital twin of the observed object is constructed according to the state and the event of the current moment, the correlation among the time correlations of the readings of the sensor is fully explored, the state of the observed object and the event causing the state change are subjected to data fusion and reduction, the real-time, high-robustness and accurate digital twin is constructed, and a digital basis is provided for subsequent planning, control and scheduling. Moreover, the construction method of the digital twins is suitable for all scenes needing to fuse the observation data of the sensor and construct the digital twins, and is wide in applicability.
Fig. 2 is a functional block diagram of a digital twinning building apparatus according to an embodiment of the present invention. As shown in fig. 2, the apparatus 20 includes an obtaining module 21, a state predicting module 22, an event predicting module 23, and a constructing module 24.
An obtaining module 21, configured to obtain observation data of an observation object acquired by at least one sensor;
the state prediction module 22 is used for calculating the posterior probability of the current moment according to the observation data based on the Bayesian principle, and obtaining the state of the observed object at the current moment by using the posterior probability;
the event prediction module 23 is configured to predict an event of the observed object at the current moment according to the state and the observation data and a preset rule;
a construction module 24 for constructing a digital twin of the observation based on the state and the event.
Alternatively, when the number of sensors is one, the state prediction module 22 may calculate the posterior probability of the current time according to the observation data based on the bayesian principle, and the operation of: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment, which are obtained in advance; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the current moment.
Alternatively, when the number of the sensors is two or more and the same type of sensor, the operation of the state prediction module 22 calculating the posterior probability of the current time from the observation data based on the bayesian principle may be further: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the sensor of the current moment.
Optionally, when the number of the sensors is two or more and different types of sensors, if the first observation data of the main sensor and the second observation data of the auxiliary sensor are data of the observed object collected from different angles, the operation of calculating the posterior probability of the current time according to the observation data by the state prediction module 22 based on the bayesian principle may further be: acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
Optionally, when the number of the sensors is two or more and different types of sensors, if the second observation data of the auxiliary sensor is a correction of the first observation data of the main sensor, the state prediction module 22 may calculate the posterior probability of the current time according to the observation data based on the bayesian principle, and the operation of calculating the posterior probability of the current time according to the observation data may further be: confirming the corrected pose parameters of the main sensor in the first observation data and the reading of the main sensor for the observed object; acquiring a predicted pose parameter, a predicted state of an observed object and a predicted probability of the main sensor at the current moment, wherein the predicted pose parameter, the predicted state and the predicted probability are obtained based on the pose parameter of the main sensor at the previous moment and the state of the observed object; calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment; and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
Optionally, when the events occurring at the adjacent moments of the observed object are independent from each other, the operation of predicting the event of the observed object at the current moment by the event prediction module 23 according to the state and the observed data and according to the preset rule may further be: based on a Bayesian model, a first function alpha (t) and a second function beta (t) are defined according to the relationship among observation data, states and events: α (t) ═ p (x)t,lt,ut|at,rt)；β(t)＝p(xt,lt,ut|at-1,rt-1) (ii) a Wherein t represents time, xtRepresents a state, utRepresents an event,/tIndicating the pose of the sensor itself, rtIndicating sensorReading for an observed object, atA reading representing a sensor tag placed at a known fixed location; utilizing the first function and the second function to carry out mutual iteration to obtain an event posterior probability for deducing the state and the event based on the observation data time sequence; and obtaining the event of the observed object at the current moment according to the posterior probability of the event.
Optionally, when the events occurring at the adjacent moments of the observed object are related to each other, the operation of the event prediction module 23 predicting the event of the observed object at the current moment according to the state and the observation data and the preset rule may further be: based on the hidden Markov model, a third function γ (t) is defined according to the state and event at time 0-t:
wherein t represents time, x0:tRepresents a state, utRepresenting an event; iterating the third function to obtain the probability of event inference; and obtaining the event of the observed object at the current moment according to the probability deduced for the event.
For other details of the technical solution for implementing each module in the digital twin construction apparatus in the above embodiment, reference may be made to the description of the digital twin construction method in the above embodiment, and details are not repeated here.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device-like embodiments, since they are substantially similar to the method embodiments, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiments
Referring to fig. 3, fig. 3 is a schematic structural diagram of a terminal according to an embodiment of the present invention. As shown in fig. 3, the terminal 30 includes a processor 31 and a memory 32 coupled to the processor 31.
The memory 32 stores program instructions that, when executed by the processor 31, cause the processor 31 to execute the steps of the construction method of the digital twin in the above-described embodiment.
The processor 31 may also be referred to as a CPU (Central processing unit). The processor 31 may be an integrated circuit chip having signal processing capabilities. The processor 31 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Referring to fig. 4, fig. 4 is a schematic structural diagram of a storage medium according to an embodiment of the invention. The storage medium of the embodiment of the present invention stores a program file 41 capable of implementing all the methods described above, where the program file 41 may be stored in the storage medium in the form of a software product, and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or terminal devices, such as a computer, a server, a mobile phone, and a tablet.
In the embodiments provided in the present invention, it should be understood that the disclosed terminal, apparatus and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit. The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method of constructing a digital twin, comprising:
acquiring observation data of an observation object acquired by at least one sensor, wherein the at least one sensor comprises one sensor or a plurality of sensors of the same type or a plurality of sensors of different types;
based on Bayes principle, calculating posterior probability of the current moment according to the observation data, and obtaining the state of the observed object at the current moment by utilizing the posterior probability;
predicting the event of the observed object at the current moment according to the state and the observation data and a preset rule;
constructing a digital twin of the observation based on the state and the event.
2. The method for constructing a digital twin according to claim 1, wherein when the number of the sensors is one, the calculating a posterior probability of a current time from the observation data based on the bayesian principle includes:
acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment, which are obtained in advance;
calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment;
and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and a pre-acquired likelihood function of the current moment.
3. The method according to claim 1, wherein when the number of the sensors is two or more and the same type of sensor, the calculating the posterior probability of the current time from the observation data based on the bayesian principle includes:
acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment;
calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment;
and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment and the pre-acquired likelihood function of the sensor of the current moment.
4. The method according to claim 1, wherein when the number of the sensors is two or more and different types of sensors, if the first observation data of the main sensor and the second observation data of the auxiliary sensor are the data of the observed object collected from different angles, the calculating the posterior probability of the current time from the observation data based on the bayesian principle includes:
acquiring a prediction state and a prediction probability of a current moment based on a state prediction of a previous moment;
calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment;
and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
5. The method according to claim 1, wherein when the number of the sensors is two or more and different types of sensors, if the second observation data of the auxiliary sensor is a correction to the first observation data of the main sensor, the calculating the posterior probability of the current time from the observation data based on the bayesian principle includes:
confirming the corrected pose parameters of the primary sensor and the readings of the primary sensor for the observed object in the first observation data;
acquiring a predicted pose parameter of the main sensor, a predicted state of the observed object and a predicted probability at the current moment, wherein the predicted pose parameter is predicted based on the pose parameter of the main sensor at the previous moment and the state of the observed object;
calculating to obtain the prior probability of the current moment according to the prediction probability and the posterior probability of the previous moment;
and calculating to obtain the posterior probability of the current moment according to the prior probability of the current moment, the pre-acquired first likelihood function of the main sensor and the pre-acquired second likelihood function of the auxiliary sensor at the current moment.
6. The method for constructing a digital twin according to claim 1, wherein when events occurring at adjacent times of the observation object are independent of each other, the predicting of the event of the observation object at the current time according to the state and the observation data and a preset rule includes:
defining a first function α (t) and a second function β (t) from a relationship between the observation data, the state, the events, based on a Bayesian model: α (t) ═ p (x)t,lt,ut|at,rt)；β(t)＝p(xt,lt,ut|at-1,rt-1) (ii) a Wherein t represents time, xtRepresents said state, utRepresents the event,/tRepresenting the pose of the sensor itself, rtRepresenting the reading of the sensor for the observation, atA reading representing a sensor tag placed at a known fixed location;
utilizing the first function and the second function to iterate mutually to obtain an event posterior probability for deducing the state and the event based on an observation data time sequence;
and obtaining the event of the observed object at the current moment according to the posterior probability of the event.
7. The method for constructing the digital twin according to claim 1, wherein when events occurring at adjacent moments of the observation object are related to each other, the predicting of the event of the observation object at the current moment according to the state and the observation data and a preset rule includes:
defining a third function γ (t) from said states, events at time 0-t, based on a hidden Markov model:
wherein t represents time, x0:tRepresents said state, utRepresenting the event;
iterating the third function to obtain the probability of event inference;
and obtaining the event of the observed object at the current moment according to the inferred probability of the event.
8. A digital twinning construction apparatus, comprising:
the acquisition module is used for acquiring observation data of an observed object acquired by at least one sensor;
the state prediction module is used for calculating the posterior probability of the current moment according to the observation data based on the Bayesian principle and obtaining the state of the observed object at the current moment by utilizing the posterior probability;
the event prediction module is used for predicting the event of the observed object at the current moment according to the state and the observation data and a preset rule;
a construction module for constructing a digital twin of the observation based on the state and the event.
9. A terminal, characterized in that it comprises a processor, a memory coupled to the processor, in which are stored program instructions which, when executed by the processor, cause the processor to carry out the steps of the digital twinning construction method according to any of claims 1-7.
10. A storage medium characterized by storing a program file capable of implementing the digital twin construction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110218736.6A CN112949174A (en) | 2021-02-24 | 2021-02-24 | Digital twin construction method and device, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110218736.6A CN112949174A (en) | 2021-02-24 | 2021-02-24 | Digital twin construction method and device, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112949174A true CN112949174A (en) | 2021-06-11 |
Family
ID=76246536
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110218736.6A Withdrawn CN112949174A (en) | 2021-02-24 | 2021-02-24 | Digital twin construction method and device, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112949174A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113495577A (en) * | 2021-09-07 | 2021-10-12 | 南京航空航天大学 | Unmanned aerial vehicle cluster sensor model correction method for digital twin simulation |
| CN116187153A (en) * | 2022-11-14 | 2023-05-30 | 中国水利水电科学研究院 | Hydraulic structure digital twin model updating method based on hierarchical Bayes |
-
2021
- 2021-02-24 CN CN202110218736.6A patent/CN112949174A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113495577A (en) * | 2021-09-07 | 2021-10-12 | 南京航空航天大学 | Unmanned aerial vehicle cluster sensor model correction method for digital twin simulation |
| CN116187153A (en) * | 2022-11-14 | 2023-05-30 | 中国水利水电科学研究院 | Hydraulic structure digital twin model updating method based on hierarchical Bayes |
| CN116187153B (en) * | 2022-11-14 | 2023-08-29 | 中国水利水电科学研究院 | Hydraulic structure digital twin model updating method based on hierarchical Bayes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Karakra et al. | HospiT'Win: a predictive simulation-based digital twin for patients pathways in hospital | |
| CN112949174A (en) | Digital twin construction method and device, terminal and storage medium | |
| CN109211277B (en) | State determination method and device of visual inertial odometer and electronic equipment | |
| CN112232495B (en) | Prediction model training method, device, medium and computing equipment | |
| AU2009275668B2 (en) | Estimating states of a plurality of targets tracked by a plurality of sensors | |
| CN111795687B (en) | Robot map updating method and device, readable storage medium and robot | |
| JP2007215354A (en) | Method and processing program for estimating electrical load | |
| CN110059521B (en) | Target tracking method and device | |
| CN116594364B (en) | Optimization method and system based on prefabricated vegetable production control system | |
| JP7300550B2 (en) | METHOD AND APPARATUS FOR CONSTRUCTING SIGNS MAP BASED ON VISUAL SIGNS | |
| CN116088503B (en) | Dynamic obstacle detection method and robot | |
| CN109299669B (en) | Video face key point detection method and device based on double intelligent agents | |
| EP4170561A1 (en) | Method and device for improving performance of data processing model, storage medium and electronic device | |
| CN111508072A (en) | Map construction method and device, electronic equipment and storage medium | |
| CN116908777A (en) | Multi-robot random networking collaborative navigation method based on explicit communication with tag Bernoulli | |
| CN111158999A (en) | Method and device for calculating formatting time | |
| WO2019012740A1 (en) | Operation rule extraction device, operation rule extraction system, and operation rule extraction method | |
| CN110888957B (en) | Object positioning method and related device | |
| CN115267667B (en) | Underground high-precision positioning correction method, device, equipment and storage medium | |
| CN109343013B (en) | Spatial registration method and system based on restarting mechanism | |
| Berkvens et al. | Asynchronous, electromagnetic sensor fusion in RatSLAM | |
| CN115205419A (en) | Instant positioning and map construction method and device, electronic equipment and readable storage medium | |
| CN113406678B (en) | Filtering method, device, equipment and storage medium | |
| CN108572939B (en) | VI-SLAM optimization method, device, equipment and computer readable medium | |
| CN112445835B (en) | Business data processing method and device, network management server and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210611 |