WO2020037610A1

WO2020037610A1 - Motion trajectory prediction method for target object, and monitoring platform

Info

Publication number: WO2020037610A1
Application number: PCT/CN2018/101977
Authority: WO
Inventors: 邬奇峰; 李昊南; 刘尧
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2018-08-23
Filing date: 2018-08-23
Publication date: 2020-02-27
Also published as: CN110785765A

Abstract

Disclosed in the present invention are a motion trajectory prediction method for a target object, and a monitoring platform, said method comprising: acquiring, according to a trajectory model, an expected motion state of a target object in a lost state, the target object being an object monitored by a monitoring platform, and the trajectory model comprising model coefficients; acquiring an observation position of the target object in a monitored state; determining the model coefficients according to the observation position and the expected motion state, and updating the trajectory model on the basis of the determined model coefficients; and predicting, according to the updated trajectory model, the motion trajectory of the target object in the lost state. The present invention uses the expected motion state and the observation position to determine the model coefficients of the trajectory model of the target object, effectively improving the accuracy of determination of the model coefficients, obtaining a more accurate trajectory model of the target object, and thereby effectively improving the accuracy of motion trajectory prediction for the target object.

Description

Prediction method and monitoring platform of target object's motion trajectory

Technical field

The present invention relates to the field of monitoring technology, and in particular, to a method and a monitoring platform for predicting a motion trajectory of a target object.

Background technique

With the continuous progress of science and technology, the monitoring platform (fixed monitoring platform or mobile monitoring platform) can realize the monitoring of the target object, and some monitoring platforms can further follow the target object.

In complex environments (such as woods, indoor environments, etc.), the target object may be blocked by obstacles. The monitoring platform cannot observe the target object because the monitoring platform cannot obtain the motion data or images of the target object, and the target object is in a lost state. However, when the target object is in a lost state, the motion trajectory prediction method of the target object in the prior art cannot accurately predict the motion trajectory of the target object in the lost state.

Summary of the Invention

The invention provides a method and a monitoring platform for predicting the motion trajectory of a target object to improve the accuracy of prediction of the motion trajectory of the target object.

A first aspect of the present invention is to provide a method for predicting a motion trajectory of a target object, which is applied to a monitoring platform and includes:

Obtaining the expected motion state of the target object in a lost state according to a trajectory model, wherein the target object is a monitoring object of the monitoring platform, and the trajectory model includes a model coefficient;

Obtaining an observation position when the target object is in a monitoring state;

Determining the model coefficient according to the observation position and the expected motion state, and updating the trajectory model based on the determined model coefficient;

The motion trajectory of the target object in the lost state is predicted according to the updated trajectory model.

A second aspect of the present invention is to provide a monitoring platform, including:

Memory for storing computer programs;

A processor for running a computer program stored in the memory to implement:

A third aspect of the present invention is to provide a monitoring platform, including:

A processing module configured to obtain an expected motion state of a target object when the target object is in a lost state according to a trajectory model, wherein the target object is a monitoring object of the monitoring platform, and the trajectory model includes a model coefficient;

An acquisition module, configured to acquire an observation position when the target object is in a monitoring state;

A determining module, configured to determine the model coefficient according to the observation position and the expected motion state, and update the trajectory model based on the determined model coefficient;

The prediction module is configured to predict a motion trajectory of the target object in a lost state according to the updated trajectory model.

A fourth aspect of the present invention is to provide a computer-readable storage medium, where the computer-readable storage medium stores program instructions, and the program instructions are used to implement the method for predicting a motion track of a target object according to the first aspect.

The method and monitoring platform for predicting the trajectory of a target object provided by the present invention determine the model coefficients based on the expected motion state and observation position of the target object, update the trajectory model based on the determined model coefficients, and predict the target object based on the updated trajectory model The trajectory when in a lost state. Among them, determining the model coefficients of the trajectory model of the target object through the expected motion state and observation position can effectively improve the accuracy of the determination of the model coefficients, and then obtain a more accurate trajectory model of the target object, which can effectively improve the target. The accuracy of the object's motion trajectory prediction.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic flowchart of a method for predicting a motion trajectory of a target object according to an embodiment of the present invention; FIG.

FIG. 2 is a schematic flowchart of determining the model coefficient according to the observation position and an expected motion state according to an embodiment of the present invention; FIG.

FIG. 3 is a first schematic flowchart of obtaining an observation position when the target object is in a monitoring state according to an embodiment of the present invention; FIG.

FIG. 4 is a second schematic flowchart of obtaining an observation position of the target object in a monitoring state according to an embodiment of the present invention; FIG.

5 is a schematic flowchart of another method for predicting a motion trajectory of a target object according to an embodiment of the present invention;

6 is a schematic structural diagram of a monitoring platform according to an embodiment of the present invention;

FIG. 7 is a schematic structural diagram of another monitoring platform according to an embodiment of the present invention.

detailed description

In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described clearly and completely in combination with the drawings in the embodiments of the present invention. It is a part of the embodiments of the present invention, but not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the drawings. In the case where there is no conflict between the embodiments, the following embodiments and features in the embodiments can be combined with each other.

The target object is the monitoring object of the monitoring platform. The monitoring platform may include a fixed monitoring platform, for example, a monitoring device provided on a fixed object (such as a building or a pole). For example, the monitoring device may be provided on a building. Video surveillance equipment. The monitoring platform may include a movable platform, where the movable platform may be a monitoring device that is moved by an external force, such as a handheld gimbal; the movable platform may be a monitoring device that is moved by a power system configured by itself, specifically, The movable platform may include a mobile robot. Further, the mobile robot may include a land mobile robot, an underwater or surface robot, a drone or a space robot, and so on. The target object may be an object having specific characteristics, for example, the target object may include a user, a car, an animal, and the like.

According to whether the monitoring platform can monitor the target object, the status of the target object can include the monitoring state and the missing state.

When the target object is in a monitoring state, the monitoring platform can observe the target object, for example, the position of the target object is observed, that is, the observation position P _{i of the} target object is obtained.

When the target object is lost, the monitoring platform cannot observe the target object, for example, the position of the target object cannot be observed. In complex environments (such as woods, indoor environments, etc.), the target object may be blocked by obstacles. Because the monitoring platform cannot observe the target object, the target object is lost. For example, the monitoring platform cannot observe the target object. The monitoring platform cannot obtain the observation position of the target object.

The trajectory model of the target object can be used to represent the motion trajectory of the target object. The trajectory model can specifically be a function that uses time as an independent variable to indicate the relationship between the position and time of the target object. The trajectory model may include model coefficients, and the model coefficients may be parameters other than independent variables in the trajectory model. Further, the trajectory model may include a polynomial trajectory model. In this case, the model coefficients are polynomial coefficients and constant terms in the polynomial trajectory model. For example, taking the trajectory of the target object on the X axis as an example, the trajectory model may be a polynomial trajectory model T (t) = p ₀ + p ₁ t + p ₂ t ² + p ₃ t ³ + ... + p _n t ⁿ , wherein the model coefficients may be p ₀ , p ₁ , p ₂ , p ₃ ... p _n . In some embodiments, the trajectory model may be other types of models, for example, the trajectory model may be an exponential trajectory model.

When the target object is in a lost state, in order to continuously monitor the target object, it is necessary to predict the motion trajectory of the target object. At this time, the trajectory model of the target object can be used to predict the motion trajectory of the target object. Specifically, when the model coefficients in the trajectory model of the target object are known, the target object is in a lost state, that is, after the target object is lost. At any time, the trajectory of the target object can be predicted according to the trajectory model of the target object. For example, by substituting the moment after the target object is lost into the trajectory model of the target object, it can predict the position of the target object when it is lost.

In the prior art, when predicting the motion trajectory of a target object, a model coefficient in the trajectory model of the target object is determined by running a minimum fitting algorithm on the position constraint of the target object, that is, the motion trajectory of the target object is simulated. Together. The position constraint may be determined according to a position error between the observation position and an expected position corresponding to the observation position, wherein the expected position is determined according to a trajectory model of a target object. For example, a position error between the observation position and an expected position corresponding to the observation position is T (t _i ) -p _i , and the position constraint may be:

Run a minimization fitting algorithm on the position constraints of the target object

To determine the model coefficients in the trajectory model of the target object. However, in the prior art, the constraint is only imposed on the position dimension, and the motion state of the target object when it is lost is not considered. This results in inaccurate motion trajectories, and the trajectory model determined by the fit cannot reflect that the target objects are lost. State of motion. For example, when the target object in the lost state moves at approximately uniform speed, since the motion state of the target object in the lost state is not taken into account, it can be calculated that the target object in the lost state may have large motion acceleration according to the fitted motion trajectory.

In order to solve the above technical problems, FIG. 1 is a schematic flowchart of a method for predicting a motion trajectory of a target object according to an embodiment of the present invention. Referring to FIG. 1, this embodiment provides a method for predicting a motion trajectory of a target object. Apply to monitoring platform. Specifically, the method includes:

S101: Obtain an expected motion state when the target object is in a lost state according to the trajectory model of the target object, where the target object is a monitoring object of the monitoring platform, and the trajectory model includes a model coefficient;

Specifically, when the target object is in a lost state, the monitoring platform may obtain an expected motion state of the target object according to the trajectory model of the target object. The expected motion state of the target object may include at least one of a speed, an acceleration, and a change amount of the acceleration. Specifically, those skilled in the art may set it according to specific design requirements and use requirements, and details are not described herein again. For example, when the expected motion state is acceleration, the acceleration model of the target object can be obtained according to the trajectory model of the target object, and the moment when the target object is in the missing state is substituted into the acceleration model of the target object, that is, the target object can be obtained Lost state acceleration.

S102: Obtain an observation position when the target object is in a monitoring state;

Specifically, when the target object is in a monitoring state, the monitoring platform can observe the target object. As mentioned earlier, the monitoring platform can obtain the observation position of the target object.

S103: Determine a model coefficient according to the observation position and the expected motion state, and update the trajectory model based on the determined model coefficient;

Specifically, after the observation position and the expected motion state are obtained, the observation position and the expected motion state may be analyzed and processed to determine a model coefficient of the trajectory model, and the trajectory model may be further updated based on the determined model coefficient. After obtaining the updated trajectory model, the trajectory model of the target object has been determined.

S104: Predict the motion trajectory of the target object in a lost state according to the updated trajectory model.

Specifically, when the target object is in a lost state, the updated trajectory model may be used to predict the motion trajectory of the target object to obtain the motion trajectory when the target object is in a lost state. For example, the time when the target object is in a lost state can be substituted into the updated trajectory model, that is, the position of the target object can be predicted, and then the motion trajectory of the target object in the lost state can be predicted by the predicted position of the target object.

The method for predicting the trajectory of a target object provided in this embodiment determines a model coefficient based on the expected motion state and observation position of the target object, updates the trajectory model based on the determined model coefficient, and can predict that the target object is missing based on the updated trajectory model. The motion trajectory in the state. Among them, determining the model coefficients of the trajectory model of the target object through the expected motion state and observation position can effectively improve the accuracy of the determination of the model coefficients, and then obtain a more accurate trajectory model of the target object, which can effectively improve the target. The accuracy of the object's motion trajectory prediction.

FIG. 2 is a schematic flowchart of determining a model coefficient according to an observation position and an expected motion state according to an embodiment of the present invention; based on the foregoing embodiment, and referring to FIG. 2, it can be known that the specific determination method of the model coefficient is not performed in this embodiment. By definition, a person skilled in the art may set it according to specific design requirements. Optionally, in this embodiment, determining the model coefficient according to the observation position and the expected motion state may include:

S201: Obtain a position error between an observation position and an expected position when the target object is in a monitoring state, and determine a position constraint of the target object according to the position error, where the expected position is a target object corresponding to the observation position determined according to the trajectory model. position;

Specifically, the monitoring platform can obtain the observation position P _i when the target object is in the monitoring state, and the monitoring platform can obtain the expected position of the target object corresponding to the observation position according to the trajectory model of the target object as T (t _i ). For example, at time t _i when the target object is in the monitoring state, the monitoring platform acquires the observation position P _i when the target object is in the monitoring state, and can substitute t _i into the trajectory model of the target object, that is, it can obtain the observation position The expected position of the corresponding target object. At this time, the expected position of the target object is the expected position where the target object is in a monitoring state.

Further, the monitoring platform may obtain a position error between the observed position and the expected position, and the position error may be T (t _i ) -p _i or p _i -T (t _i ), and then the target object is determined according to the position error. Location constraints. Optionally, the position constraint may be:

Optionally, the position constraint is

It can be understood that those skilled in the art may also use other methods to determine the position constraint of the target object according to the position error according to the specific usage scenario and design requirements, and details are not described herein again.

S202: Determine a motion state constraint of the target object according to the expected motion state;

Specifically, in order to ensure that the fitted motion trajectory more accurately reflects the true motion state of the target object, the motion state constraint of the target object may be determined according to the expected motion state. The specific determination method of the motion state constraint of the target object can be implemented in the following feasible ways:

A feasible method is to determine the integral term according to the expected motion state, and perform integral operation on the integral term within a preset time period to obtain the motion state constraint of the target object, wherein the preset time period is when the target object is in a lost state. period.

Specifically, after the expected motion state is obtained, an integral term corresponding to the expected motion state may be determined. For example: when the expected motion state is acceleration, the determined integral term corresponding to the expected motion state may be | T ⁽²⁾ (t) | or | T ⁽²⁾ (t) | ² , and then the said Integral term performs integral operation to obtain the motion state constraint of the target object

or

Wherein, (t _l , t _m ) is a preset time period, and the preset time period is a time period when the target object is in a lost state. Optionally, t _l is the initial time when the target object is in the lost state, that is, the time when the monitoring platform determines that the target object is in the lost state, and t _m is the intermediate time when the target object is in the lost state.

Another feasible way is to determine a plurality of accumulation terms according to the expected movement state, and perform accumulation operations on the plurality of accumulation terms to obtain a movement state constraint of the target object.

Specifically, the expected motion state of the target object at multiple moments can be obtained, so that multiple accumulation terms can be determined, and accumulation operations are performed on the multiple accumulation terms to obtain the movement state constraints of the target object. For example, when the expected motion state is acceleration, an accumulation term | T ⁽²⁾ (t _j ) | or | T ⁽²⁾ (t _j ) corresponding to the expected motion state T ⁽²⁾ (t _j ) may be obtained. | ² and perform accumulation operations on multiple accumulation terms to obtain the motion state constraints of the target object

or

S203: Run a minimization fitting algorithm based on position constraints and motion state constraints to determine model coefficients.

Specifically, after acquiring the position constraint and the motion state constraint, a minimization fitting algorithm may be run based on the position constraint and the motion state constraint to determine a model coefficient of the trajectory model. For convenience of explanation, the position constraint is abbreviated as A and the motion state constraint is abbreviated as B. A fitting term may be determined based on the position constraint A and the motion state constraint B, and a minimizing fitting algorithm is run on the fitting term. For example, the fitted term may be (A + λ _t B), and a minimization fitting algorithm min (A + λ _t B) is run on the fitted term to determine a model coefficient, where λ _t may be a prediction Set the coefficient.

When the expected motion state is acceleration, it is assumed that when the target object is in a lost state, the acceleration of the target object is small, which is approximately uniform motion. When running the minimization fitting algorithm min (A + λ _t B), to minimize A + λ _t B, the motion state constraint should be as small as possible, that is, the acceleration of the target object in the missing state should be as small as possible.

When the expected motion state is a change amount of acceleration, it is assumed that when the target object is in a lost state, the change amount of acceleration of the target object is small, which is similar to a uniformly variable motion. When running the minimization fitting algorithm min (A + λ _t B), to minimize A + λ _t B, the motion state constraint should be as small as possible, that is, the change amount of the acceleration of the target object in the missing state should be as small as possible .

Compared with the prior art, the trajectory prediction method provided by the embodiment of the present invention can determine the model coefficients of the trajectory model more accurately by considering both the position constraint and the motion state constraint, thereby obtaining a more accurate trajectory model of the target object. Effectively improve the accuracy of the target's motion trajectory prediction.

Based on the above embodiment and continuing to refer to FIG. 1, in this embodiment, after predicting the motion trajectory of the target object in a lost state according to the updated trajectory model, the method further includes: sending the motion trajectory to the control The terminal so that the control terminal displays the motion track on the interactive interface.

Specifically, after the monitoring platform obtains the motion trajectory of the target object, the monitoring platform may send the motion trajectory to the control terminal. The control terminal may be a terminal device with a display device, such as one or more of a smart phone, a tablet computer, a laptop computer, a desktop computer, and a wearable device. After the control terminal receives the motion track, it controls The display device on the terminal can display the motion trajectory, so that the user can intuitively obtain the motion trajectory of the target object in the lost state.

FIG. 3 is a schematic flowchart of acquiring an observation position when a target object is in a monitoring state according to an embodiment of the present invention. Based on the above embodiment, it can be known that the monitoring platform includes a photographing device, and the photographing device may include a camera, a video camera, and so on. Wherein, the obtaining the observation position of the target object when it is in a monitoring state includes:

S301: Acquire an image output by the shooting device;

S302: Obtain an observation position of a target object according to an image, where the image includes the target object.

Specifically, the processor of the monitoring platform may acquire an image output by the shooting device, where the image includes a target object, and the processor may identify the target object in the image to obtain a position of the target object in the image, and further Ground, the observation position of the target object is acquired according to the position of the target object in the image.

It can be understood that when the target object is in an image output by the shooting device, the target object is in a monitoring state, and the monitoring platform can obtain the observation position of the target object through the image. When the target object is not in the image output by the shooting device, the monitoring platform cannot obtain the observation position of the target object through the image, and the target object is in a lost state.

FIG. 4 is a second schematic flowchart of obtaining an observation position of a target object in a monitoring state according to an embodiment of the present invention. Based on the above embodiment, it can be known from continuing to refer to FIG. 4 that acquiring the observation position when the target object is in a monitoring state may include:

S401: Acquire motion data sent by a control terminal carried by a target object;

S402: Obtain the observation position of the target object according to the motion data.

Specifically, the target object carries a control terminal, wherein the control terminal is configured with a motion sensor, and the motion sensor can detect the motion data of the target object and output the motion data. Wherein, the motion data may include at least one of motion position information, speed information, and acceleration information, and the motion sensor may include a satellite positioning receiver, an inertial measurement unit, a photoelectric code disc, and the like. The control terminal may send the motion data to the monitoring platform. After the monitoring platform obtains the motion data, it may analyze and process the motion data to obtain the observation position of the target object.

It can be understood that when the monitoring platform can receive the motion data sent by the control terminal, the target object is in a monitoring state, and the monitoring platform can obtain the observation position of the target object through the image. When the monitoring platform cannot receive the motion data sent by the control terminal, the monitoring platform cannot obtain the observation position of the target object through the image, and the target object is in a lost state.

FIG. 5 is a schematic flowchart of another method for predicting a motion trajectory of a target object according to an embodiment of the present invention. Based on the above embodiments, and referring to FIG. 5 continuously, in order to improve the practicability of the method, the method may further include:

S501: When the target object is in a lost state, determine the motion trajectory of the mobile robot according to the motion trajectory;

As mentioned above, the monitoring platform may include a mobile robot, and the target object may be a follower of the mobile robot. When the target object is in a monitoring state, the processor of the mobile robot may determine the motion trajectory of the mobile robot according to the observation position of the target object, and control the mobile robot to move according to the motion trajectory to follow the target object. However, when the target object is in a missing state, the mobile robot cannot observe the target object, and when the target object needs to be followed, the processor of the mobile robot can determine the motion trajectory of the mobile robot according to the predicted motion trajectory.

Optionally, determining the motion trajectory of the mobile robot according to the motion trajectory may include: determining the motion trajectory as the motion trajectory of the mobile robot. Specifically, the processor of the mobile robot may determine the predicted motion trajectory of the target object as the motion trajectory of the mobile robot.

S502: Control the mobile robot to move according to the motion trajectory.

The processor of the mobile robot may control the mobile robot to move according to the determined motion trajectory to follow the target object. Since the motion trajectory is determined according to the predicted motion trajectory of the target object, the mobile robot is When the determined motion trajectory moves, the probability of the mobile robot encountering an obstacle can be effectively reduced, and at the same time, the probability of retrieving and recovering the monitoring after the target object is blocked can be improved, thereby ensuring the success rate of following.

FIG. 6 is a schematic structural diagram of a monitoring platform according to an embodiment of the present invention. As can be seen with reference to FIG. 6, this embodiment provides a monitoring platform. The monitoring platform may be a movable platform, and the movable platform may include a mobile platform. Robot; specifically, the monitoring platform may include:

A memory 601, configured to store a computer program;

The processor 602 is configured to run a computer program stored in the memory 601 to implement: obtaining an expected motion state of the target object when the target object is in a lost state according to a trajectory model, wherein the target object is a monitoring object of the monitoring platform, and the trajectory model includes model coefficients; The observation position of the target object in the monitoring state; the model coefficients are determined according to the observation position and the expected motion state, and the trajectory model is updated based on the determined model coefficients; the motion trajectory of the target object in the lost state is predicted based on the updated trajectory model.

The trajectory model may be a polynomial trajectory model; the expected motion state may include: acceleration and / or a change amount of the acceleration.

Further, when the processor 602 determines the model coefficient according to the observation position and the expected motion state, the processor 602 is configured to:

Obtain a position error between the observation position and the expected position when the target object is in the monitoring state, and determine the position constraint of the target object according to the position error, where the expected position is the position of the target object corresponding to the observation position determined according to the trajectory model;

Determine the movement state constraints of the target object according to the expected movement state;

A minimization fitting algorithm is run based on position constraints and motion state constraints to determine model coefficients.

In addition, when the processor 602 determines the motion state constraint of the target object according to the expected motion state, an implementable manner is that the processor 602 is configured to:

The integral term is determined according to the expected motion state, and the integral term is integrated within a preset time period to obtain the motion state constraint of the target object, wherein the preset time period is a time period when the target object is in a lost state.

Another implementable manner is that the processor 602 is configured to:

Multiple accumulation terms are determined according to the expected movement state, and accumulation operations are performed on the multiple accumulation terms to obtain a movement state constraint of the target object.

Further, the processor 602 is configured to:

After predicting the motion trajectory when the target object is in a lost state according to the updated trajectory model, the motion trajectory is sent to the control terminal so that the control terminal displays the motion trajectory on the interactive interface.

In addition, when the processor 602 acquires an observation position when the target object is in a monitoring state, an implementable manner is that the monitoring platform includes a photographing device, and the processor 602 is configured to:

Obtaining an image output by the shooting device;

Obtain the observation position of the target object according to the image, where the image includes the target object.

When the processor 602 obtains the observation position when the target object is in a monitoring state, another implementable manner is that the processor 602 is configured to:

Acquiring motion data sent by a control terminal carried by a target object;

Obtain the observation position of the target object based on the motion data.

Further, the processor 602 is configured to:

When the target object is in a lost state, determine the motion trajectory of the mobile robot according to the motion trajectory;

Control the mobile robot to move according to the motion trajectory.

Wherein, when the processor 602 determines the motion trajectory of the mobile robot according to the motion trajectory, the processor 602 is configured to:

The motion trajectory is determined as the motion trajectory of the mobile robot.

The monitoring platform in this embodiment may be used to execute the technical solution of the embodiment shown in FIG. 1-5 in the foregoing method, and the implementation principles and technical effects thereof are similar, and details are not described herein again.

FIG. 7 is a schematic structural diagram of another monitoring platform according to an embodiment of the present invention. Referring to FIG. 7, this embodiment provides another monitoring platform. The monitoring platform may include:

A processing module 701 is configured to obtain an expected motion state of a target object when the target object is in a lost state according to a trajectory model, where the target object is a monitoring object of a monitoring platform, and the trajectory model includes a model coefficient;

An obtaining module 702, configured to obtain an observation position when the target object is in a monitoring state;

A determining module 703, configured to determine a model coefficient according to the observation position and the expected motion state, and update the trajectory model based on the determined model coefficient;

The prediction module 704 is configured to predict a motion trajectory of the target object in a lost state according to the updated trajectory model.

Another aspect of this embodiment provides a computer-readable storage medium. The computer-readable storage medium stores program instructions, and the program instructions are used to implement a method for predicting a motion trajectory of a target object in any one of the foregoing embodiments.

The technical solutions and technical features in each of the above embodiments may be singular or combined in the case of conflict with the present, as long as they do not exceed the scope of the knowledge of those skilled in the art, they all belong to the equivalent embodiments within the protection scope of this application .

In the several embodiments provided by the present invention, it should be understood that the related apparatuses and methods disclosed may be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be another division manner. For example, multiple units or components may be divided. The combination can either be integrated into another system, or some features can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist separately physically, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional unit.

When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention essentially or part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium , Including a number of instructions to cause the computer processor 101 (processor) to perform all or part of the steps of the method described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and other media that can store program codes.

The above description is only an embodiment of the present invention, and thus does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the description and drawings of the present invention, or directly or indirectly applied to other related technologies The same applies to the fields of patent protection of the present invention.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, but not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: It is still possible to modify the technical solutions described in the foregoing embodiments, or to replace some or all of the technical features equivalently; and these modifications or replacements do not depart from the essence of the corresponding technical solutions of the technical solutions of the embodiments of the present invention. range.

Claims

A method for predicting the trajectory of a target object, which is applied to a monitoring platform, is characterized in that it includes:

Obtaining the expected motion state of the target object in a lost state according to a trajectory model, wherein the target object is a monitoring object of the monitoring platform, and the trajectory model includes a model coefficient;

Obtaining an observation position when the target object is in a monitoring state;

Determining the model coefficient according to the observation position and the expected motion state, and updating the trajectory model based on the determined model coefficient;

The motion trajectory of the target object in the lost state is predicted according to the updated trajectory model.
The method according to claim 1, wherein the expected motion state comprises: acceleration and / or a change amount of acceleration.
The method according to claim 1 or 2, wherein the determining the model coefficient according to the observation position and an expected motion state comprises:

Acquiring a position error between an observed position and an expected position of the target object in a monitoring state, and determining a position constraint of the target object according to the position error, wherein the expected position is determined according to the trajectory model A position of a target object corresponding to the observation position;

Determining a movement state constraint of the target object according to the expected movement state;

A minimization fitting algorithm is run based on the position constraints and the motion state constraints to determine the model coefficients.
The method according to claim 3, wherein determining the movement state constraint of the target object according to the expected movement state comprises:

An integral term is determined according to the expected movement state, and an integral operation is performed on the integral term within a preset period of time to obtain a movement state constraint of the target object, wherein the preset period of time is that the target object is lost The time period in the state.
The method according to claim 3, wherein determining the movement state constraint of the target object according to the expected movement state comprises:

A plurality of accumulation terms are determined according to the expected movement state, and accumulation operations are performed on the plurality of accumulation terms to obtain a movement state constraint of the target object.
The method according to any one of claims 1 to 5, wherein after predicting a motion trajectory when the target object is in a lost state according to an updated trajectory model, the method further comprises:

Sending the motion trajectory to a control terminal so that the control terminal displays the motion trajectory on an interactive interface.
The method according to any one of claims 1-6, wherein the monitoring platform comprises a photographing device, and the acquiring an observation position when the target object is in a monitoring state comprises:

Acquiring an image output by the photographing device;

Observing the observation position of the target object according to the image, wherein the image includes the target object.
The method according to any one of claims 1-6, wherein the acquiring an observation position when the target object is in a monitoring state comprises:

Acquiring motion data sent by a control terminal carried by the target object;

Obtain an observation position of the target object according to the motion data.
The method according to any one of claims 1-8, wherein the monitoring platform is a movable platform.
The method of claim 9, wherein the movable platform comprises a mobile robot.
The method according to claim 10, further comprising:

Determining the motion trajectory of the mobile robot according to the predicted motion trajectory when the target object is in a lost state;

Controlling the mobile robot to move according to the motion trajectory.
The method according to claim 11, wherein determining the motion trajectory of the mobile robot according to the motion trajectory comprises:

The motion trajectory is determined as a motion trajectory of the mobile robot.
The method according to any one of claims 1-12, wherein the trajectory model is a polynomial trajectory model.
A monitoring platform includes:

Memory for storing computer programs;

A processor for running a computer program stored in the memory to implement:

Obtaining the expected motion state of the target object in a lost state according to a trajectory model, wherein the target object is a monitoring object of the monitoring platform, and the trajectory model includes a model coefficient;

Obtaining an observation position when the target object is in a monitoring state;

Determining the model coefficient according to the observation position and the expected motion state, and updating the trajectory model based on the determined model coefficient;

The motion trajectory of the target object in the lost state is predicted according to the updated trajectory model.
The monitoring platform according to claim 14, wherein the expected motion state comprises: acceleration and / or a change amount of acceleration.
The monitoring platform according to claim 14 or 15, wherein when the processor determines the model coefficient according to the observation position and an expected motion state, the processor is specifically configured to:

Acquiring a position error between an observed position and an expected position of the target object in a monitoring state, and determining a position constraint of the target object according to the position error, wherein the expected position is determined according to the trajectory model A position of a target object corresponding to the observation position;

Determining a movement state constraint of the target object according to the expected movement state;

A minimization fitting algorithm is run based on the position constraints and the motion state constraints to determine the model coefficients.
The monitoring platform according to claim 16, wherein when the processor determines a movement state constraint of the target object according to the expected movement state, the processor is specifically configured to:

An integral term is determined according to the expected movement state, and an integral operation is performed on the integral term within a preset period of time to obtain a movement state constraint of the target object, wherein the preset period of time is that the target object is lost The time period in the state.
The monitoring platform according to claim 16, wherein when the processor determines a movement state constraint of the target object according to the expected movement state, the processor is specifically configured to:

A plurality of accumulation terms are determined according to the expected movement state, and accumulation operations are performed on the plurality of accumulation terms to obtain a movement state constraint of the target object.
The monitoring platform according to any one of claims 14 to 18, wherein the processor is further configured to:

After predicting a motion trajectory when the target object is in a lost state according to the updated trajectory model, the motion trajectory is sent to a control terminal so that the control terminal displays the motion trajectory on an interactive interface.
The monitoring platform according to any one of claims 14 to 19, wherein the monitoring platform comprises a photographing device, and the processor acquires an observation position when the target object is in a monitoring state, and is specifically configured to:

Acquiring an image output by the photographing device;

Observing the observation position of the target object according to the image, wherein the image includes the target object.
The monitoring platform according to any one of claims 14 to 19, wherein when the processor obtains an observation position when the target object is in a monitoring state, the processor is specifically configured to:

Acquiring motion data sent by a control terminal carried by the target object;

Obtain an observation position of the target object according to the motion data.
The monitoring platform according to any one of claims 14-21, wherein the monitoring platform is a movable platform.
The monitoring platform according to claim 22, wherein the movable platform comprises a mobile robot.
The monitoring platform according to claim 23, wherein the processor is further configured to:

Determining the motion trajectory of the mobile robot according to the predicted motion trajectory when the target object is in a lost state;

Controlling the mobile robot to move according to the motion trajectory.
The monitoring platform according to claim 24, wherein when the processor determines the motion trajectory of the mobile robot according to the motion trajectory, the processor is specifically configured to:

The motion trajectory is determined as a motion trajectory of the mobile robot.
The monitoring platform according to any one of claims 14 to 25, wherein the trajectory model is a polynomial trajectory model.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores program instructions, and the program instructions are used to implement a method for predicting a motion track of a target object according to any one of claims 1-13. .