Target Tracking Filtering Method and Device

Abstract

The present application discloses a target tracking filtering method and apparatus. The method includes: obtaining first motion state data, second motion state data, and third motion state data of a target object; taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference to determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data; determining correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship; and determining target motion state data of the target object at the next moment according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data. In this way, the track stability of the target object can be effectively improved in real time, and it does not require a large amount of storage space, and the algorithm complexity is relatively low.

Description

Technical Field

[0001] This application belongs to the technical field of object detection and tracking, and particularly relates to an object tracking filtering method and device. Background Art

[0002] With the development and progress of technology, it has been possible to determine the motion state of an object based on the motion state data collected by sensors.

[0003] However, the data collected by sensors inevitably contains noise. Therefore, generally, the motion state data collected by sensors is filtered, and the motion state data of the object at the next moment is predicted based on the filtered motion state data. However, in the filtering process, there may be an unreasonable setting of filtering parameters. In the prediction process, there may also be a problem of inappropriate selection of prediction models. Even abnormal jumps may exist in the motion state data collected by sensors. These factors may all lead to poor track stability of the object.

[0004] In related technologies, usually, the motion state data after filtering is subjected to secondary smoothing processing to reduce the fluctuations in the motion state data after filtering, thereby improving the track stability of the object to a certain extent. However, since a certain amount of data needs to be stored for secondary smoothing processing, there will not only be a certain time delay and it cannot effectively improve the track stability of the object in real time, but also it will occupy more storage space. Summary of the Invention

[0005] The embodiments of this application provide an object tracking filtering method and device, which can effectively improve the track stability of an object in real time, and do not require a large amount of storage space, and the algorithm complexity is relatively low.

[0006] In a first aspect, the embodiments of this application provide an object tracking filtering method, which includes:

[0007] Obtain the first motion state data, the second motion state data, and the third motion state data of the object. The second motion state data is obtained by performing tracking filtering on the first motion state data, and the third motion state data is obtained by predicting the motion state of the object at the next moment based on the second motion state data.

[0008] Based on the first motion state data, the second motion state data, and the third motion state data, taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference, determine the relative position relationship of the first motion state data, the second motion state data, and the third motion state data.

[0009] Determine the correction factors corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship.

[0010] Determine the target motion state data of the target object at the next moment according to the correction factors, the first motion state data, the second motion state data, and the third motion state data.

[0011] In a second aspect, an embodiment of the present application provides a target tracking and filtering device, which includes:

[0012] An acquisition module, configured to acquire the first motion state data, the second motion state data, and the third motion state data of the target object. The second motion state data is obtained by performing tracking and filtering processing on the first motion state data, and the third motion state data is a prediction of the motion state of the target object at the next moment based on the second motion state data.

[0013] A first determination module, configured to determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data based on the first motion state data, the second motion state data, and the third motion state data, with any one of the first motion state data, the second motion state data, and the third motion state data as a reference.

[0014] A second determination module, configured to determine the correction factors corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship.

[0015] A third determination module, configured to determine the target motion state data of the target object at the next moment according to the correction factors, the first motion state data, the second motion state data, and the third motion state data.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, which includes: a processor and a memory storing computer program instructions.

[0017] When the processor executes the computer program instructions, the target tracking and filtering method shown in any one of the embodiments of the first aspect is implemented.

[0018] In a fourth aspect, an embodiment of the present application provides a computer storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the target tracking and filtering method shown in any one of the embodiments of the first aspect is implemented.

[0019] In a fifth aspect, an embodiment of the present application provides a computer program product, and when the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is caused to execute the target tracking and filtering method shown in any one of the embodiments of the first aspect.

[0020] The target tracking filtering method and device according to the embodiments of the present application can obtain the first motion state data, the second motion state data, and the third motion state data of the target object, determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data, and then determine the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data according to the relative position relationship. Then, according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, determine the target motion state data of the target object at the next moment, and then control the target object based on the target motion state data. Among them, the second motion state data is obtained by performing tracking filtering processing on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. The relative position relationship among the first motion state data, the second motion state data, and the third motion state data can reflect the real-time track deviation. Therefore, by correcting the first motion state data, the second motion state data, and the third motion state data according to the correction coefficients determined based on the relative position relationship, targeted correction can be performed on the real-time track deviation, and it is not necessary to store a certain amount of data for secondary smoothing processing. Therefore, the track stability of the target object can be effectively improved in real time, and it does not require a large amount of storage space, and the algorithm complexity is relatively low. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0022] Figure 1 is a flowchart of a target tracking filtering method provided by an embodiment of the present application.

[0023] Figure 2 is a schematic diagram of a curve showing the variation of the correction coefficient with the relative deviation amount provided by an embodiment of the present application.

[0024] Figure 3 is a schematic diagram of a curve of an experimental result provided by an embodiment of the present application.

[0025] Figure 4 is a schematic structural diagram of a target tracking filtering device provided by an embodiment of the present application.

[0026] Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The features and exemplary embodiments of various aspects of the present application will be described in detail below. To make the objectives, technical solutions, and advantages of the present application clearer and more understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application and not to limit the present application. For those skilled in the art, the present application can be implemented without some of these specific details. The following description of the embodiments is only intended to provide a better understanding of the present application by showing examples of the present application.

[0028] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0029] As described in the background art, in related fields such as robot navigation and intelligent assisted driving, sensors are essential data acquisition devices. For example, radar. Especially in an intelligent driving sensor system, millimeter-wave radar has good speed measurement capabilities for targets and good penetration capabilities for interference factors in complex environments such as rain and fog. Therefore, it has become one of the irreplaceable sensor choices in intelligent assisted driving solutions. Based on the basic principle of radar detection and considering the interference of factors such as noise during the detection process, the detection data of sensors such as radar is inevitably affected by factors such as measurement noise, resulting in certain changes or fluctuations in the measurement data. In engineering applications, sensors such as radar can be used to detect and track targets in a driving scenario. The accuracy of target detection and tracking has a significant impact on the stability of the track, and is therefore an important factor affecting the performance of an intelligent assisted driving system.

[0030] In order to improve the track stability, in traditional target detection and tracking algorithms, a filtering algorithm is usually used to filter the motion state data of the target object collected by the sensor, and the motion state data of the target object at the next moment is predicted based on the filtered motion state data. However, during the filtering process, there are often unreasonable settings of various filtering parameters, or the motion model used for prediction does not match the actual motion process of the detected target object. Moreover, there may be abnormal jump values (i.e., singular values) in the collected motion state data. These factors may all lead to certain fluctuations in the results output by the filtering algorithm, or there may be some abnormal deviations and jumps. Specifically, the track stability of the target object is still relatively poor.

[0031] In the related art, the common strategy for this phenomenon is usually to perform secondary smoothing processing on the filtered motion state data in order to reduce the fluctuations in the filtering results, thereby improving the track stability of the target object. However, this processing strategy requires storing a certain amount of filtered motion state data before secondary smoothing processing can be carried out. This will increase additional data storage and there is a certain time delay, and the results of the secondary smoothing processing cannot be output in real time. Therefore, the method adopted in the related art cannot identify the fluctuations or deviations in the filtering results in real time and cannot effectively improve the track stability of the target object in real time.

[0032] The embodiments of the present application provide a target tracking filtering method and device, which can obtain the first motion state data, the second motion state data, and the third motion state data of the target object, and determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data. Then, according to the relative position relationship, the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data are determined, and based on the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, the target motion state data of the target object at the next moment is determined, and then the target object is controlled based on the target motion state data. Among them, the second motion state data is obtained by performing tracking filtering on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. The relative position relationship among the first motion state data, the second motion state data, and the third motion state data can reflect the real-time track deviation. Therefore, by correcting the first motion state data, the second motion state data, and the third motion state data according to the correction coefficients determined based on this relative position relationship, targeted correction can be made for the real-time track deviation, and there is no need to store a certain amount of data for secondary smoothing processing. Therefore, the track stability of the target object can be effectively improved in real time, and there is no need to occupy a large amount of storage space, and the algorithm complexity is relatively low.

[0033] Figure 1The flowchart shows a target tracking and filtering method provided by an embodiment of the present application.

[0034] As Figure 1 shown, the execution subject of the target tracking and filtering method can be a target tracking and filtering device. The target tracking and filtering method may include the following steps:

[0035] S110, obtain the first motion state data, the second motion state data, and the third motion state data of the target object.

[0036] S120, based on the first motion state data, the second motion state data, and the third motion state data, taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference, determine the relative position relationship of the first motion state data, the second motion state data, and the third motion state data.

[0037] S130, according to the relative position relationship, determine the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively.

[0038] S140, according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, determine the target motion state data of the target object at the next moment.

[0039] Thus, the first motion state data, the second motion state data, and the third motion state data of the target object can be obtained, and the relative position relationship of the first motion state data, the second motion state data, and the third motion state data can be determined. Then, according to the relative position relationship, the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data are determined. And according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, the target motion state data of the target object at the next moment is determined. Then, the target object is controlled based on the target motion state data. Among them, the second motion state data is obtained by performing tracking and filtering processing on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. The relative position relationship of the first motion state data, the second motion state data, and the third motion state data can reflect the real-time track deviation. Therefore, by correcting the first motion state data, the second motion state data, and the third motion state data according to the correction coefficients determined based on the relative position relationship, targeted correction can be performed on the real-time track deviation, and there is no need to store a certain amount of data for secondary smoothing processing. Therefore, the track stability of the target object can be effectively improved in real time, and a large amount of storage space is not required, and the algorithm complexity is relatively low.

[0040] Regarding S110, the first motion state data may include, but is not limited to, motion state data such as the position and velocity of the target object at the current moment. The second motion state data may be obtained by performing tracking filtering on the first motion state data, and the third motion state data may be obtained by predicting the motion state of the target object at the next moment based on the second motion state data.

[0041] In some embodiments, in order to more accurately determine the first motion state data, the second motion state data, and the third motion state data, S110 may include:

[0042] Obtain the first motion state data collected by the sensor,

[0043] Perform tracking filtering on the first motion state data to obtain the second motion state data,

[0044] Input the second motion state data into the target motion state model, predict the motion state of the target object at the next moment, and output to obtain the third motion state data.

[0045] Exemplarily, the first motion state data may be subjected to tracking filtering by filtering algorithms such as mean filtering, Kalman filtering, and α-β filtering to obtain the second motion state data. The motion state of the target object at the next moment may be predicted by a uniform motion model or a uniformly accelerated motion model.

[0046] Of course, other filtering algorithms and motion state models may also be used, which are not limited herein.

[0047] In this way, the first motion state data, the second motion state data, and the third motion state data can be determined more accurately.

[0048] Regarding S120, the relative position relationship of the first motion state data, the second motion state data, and the third motion state data on the coordinate axis may reflect the relative deviation state between the first motion state data, the second motion state data, and the third motion state data.

[0049] If the first motion state data is X2, the second motion state data is X3, and the third motion state data is X1, the relative position relationships corresponding to various deviation states among the first motion state data, the second motion state data, and the third motion state data may be as shown in Table 1.

[0050] Table 1 - Relative Position Relationship Table

[0051] State Relative position relationship 1 <![CDATA[X1 X2 X3]]> 2 <![CDATA[X1 X3 X2]]> 3 <![CDATA[X2 X1 X3]]> 4 <![CDATA[X3 X1 X2]]> 5 <![CDATA[X2 X3 X1]]> 6 <![CDATA[X3 X2 X1]]>

[0052] In certain deviation states, the second motion state data X3 significantly deviates from the first motion state data X2 by a large distance. For example, states 3 and 4 in Table 1. For such deviation states, if the second motion state data X3 is directly output as the result, it will obviously cause the track to deviate seriously from the current position, resulting in poor track stability of the target object. Therefore, for situations similar to states 3 and 4, special optimization processing is required, while for other deviation states in Table 1, correction compensation can be performed based on the specific relative deviation amount.

[0053] In some embodiments, in order to more accurately determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data, S120 described above may include:

[0054] Determine the first relative deviation amount Δ1 = X1 - X2 between the first data X1 and the second data X2, and determine the second relative deviation amount Δ2 = X1 - X3 between the first data X1 and the third data X3.

[0055] Determine the product k = Δ1 × Δ2 of the first relative deviation amount and the second relative deviation amount.

[0056] Determine the relative position relationship according to the product.

[0057] Wherein, the first data can be any one of the first motion state data, the second motion state data, and the third motion state data, the second data can be any one of the first motion state data, the second motion state data, and the third motion state data except the first data, and the third data can be the data except the first data and the second data among the first motion state data, the second motion state data, and the third motion state data.

[0058] Here, taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference, the relative deviation amounts between this data and the other two data and the product between the relative deviation amounts can be determined, so as to determine the relative position relationship of the first motion state data, the second motion state data, and the third motion state data on the coordinate axis.

[0059] Exemplarily, the third motion state data can be taken as a reference, that is, the above-mentioned first data X1 can be the third motion state data. In addition, the second data X2 can be the first motion state data, and the third data X3 can be the second motion state data.

[0060] First, the first relative deviation amount Δ1 between X1 and X2 can be determined, and the second relative deviation amount Δ2 between X1 and X3 can be determined. Specifically:

[0061] Δ1 = X1 - X2

[0062] Δ2 = X1 - X3

[0063] Secondly, the product k of the first relative deviation amount Δ1 and the second relative deviation amount Δ2 can be determined. Specifically:

[0064] k = Δ1 × Δ2

[0065] Then, the relative position relationship can be determined based on the product k.

[0066] In this way, through the product of the first relative deviation amount and the second relative deviation amount, the relative position relationship of the first motion state data, the second motion state data, and the third motion state data on the coordinate axis can be determined more accurately.

[0067] In some embodiments, in order to more clearly determine the relative position relationship of the first motion state data, the second motion state data, and the third motion state data, the above-mentioned determination of the relative position relationship based on the product may include:

[0068] When the product k > 0, it is determined that the relative position relationship is that the second data and the third data are on the same side of the first data.

[0069] When the product k < 0, it is determined that the relative position relationship is that the second data and the third data are on different sides of the first data.

[0070] When the product k = 0, it is determined that the relative position relationship is that at least one of the second data and the third data coincides with the first data.

[0071] Here, taking the first data as a reference, the relative position relationship of the first data, the second data, and the third data on the coordinate axis can be divided into three categories: the second data and the third data are on the same side of the first data, the second data and the third data are on different sides of the first data, and at least one of the second data and the third data coincides with the first data.

[0072] Specifically, the first relative deviation amount is the relative deviation amount between the first data and the second data, and the second relative deviation amount is the relative deviation amount between the first data and the third data. If the product of the first relative deviation amount and the second relative deviation amount is greater than 0, it can indicate that the positive and negative signs of the first relative deviation amount and the second relative deviation amount are the same. Therefore, it can be determined that the second data and the third data are on the same side of the first data. If the product of the first relative deviation amount and the second relative deviation amount is less than 0, it can indicate that the positive and negative signs of the first relative deviation amount and the second relative deviation amount are opposite. Therefore, it can be determined that the second data and the third data are on different sides of the first data. If the product of the first relative deviation amount and the second relative deviation amount is equal to 0, it can indicate that at least one of the first relative deviation amount and the second relative deviation amount is 0. Therefore, it can be determined that at least one of the second data and the third data coincides with the first data.

[0073] Exemplarily, if k > 0, the relative position relationship is that X2 and X3 are on the same side of X1, belonging to one of the states 1, 2, 5, and 6 in Table 1. If k < 0, the relative position relationship is that X2 and X3 are on different sides of X1, belonging to one of the states 3 and 4 in Table 1. If k = 0, at least one of X2 and X3 coincides with X1 (not shown in Table 1).

[0074] In addition, for the case of k < 0, the positive and negative signs of Δ1 and Δ2 can be combined to further determine which one of the states 3 and 4 in Table 1 it specifically belongs to. If Δ1 > 0 and Δ2 < 0, it belongs to state 3. If Δ1 < 0 and Δ2 > 0, it belongs to state 4. Of course, in the target tracking filtering method provided by the embodiments of the present application, it is not necessary to determine specifically whether it belongs to state 3 or state 4.

[0075] In this way, through the positive and negative sign of the product of the first relative deviation amount and the second relative deviation amount, the relative position relationship of the first motion state data, the second motion state data, and the third motion state data on the coordinate axis can be determined more clearly.

[0076] Regarding S130, after determining the relative position relationship of the first motion state data, the second motion state data, and the third motion state data, the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data can be determined according to this relative position relationship.

[0077] In some embodiments, in order to correct the motion state data more effectively, the above S130 may include:

[0078] In the case where the relative position relationship is that the second data and the third data are on the same side of the first data, determine that the first correction coefficient corresponding to the first data is 1, determine the second correction coefficient corresponding to the second data according to the preset adjustment coefficient, the preset scaling coefficient, and the first relative deviation amount, and determine the third correction coefficient corresponding to the third data according to the preset adjustment coefficient, the preset scaling coefficient, and the second relative deviation amount.

[0079] In the case where the relative position relationship is that the second data and the third data are on different sides of the first data, determine that the first correction coefficient corresponding to the first data is 1, the third correction coefficient corresponding to the third data is 0, and determine the second correction coefficient corresponding to the second data according to the preset adjustment coefficient, the preset scaling coefficient, and the first relative deviation amount.

[0080] In the case where the relative position relationship is that at least one of the second data and the third data coincides with the first data, determine that the first correction coefficient corresponding to the first data and the third correction coefficient corresponding to the third data are both 1, and the second correction coefficient corresponding to the second data is 0.

[0081] Here, the first data is used as a reference, so the first correction coefficient corresponding to the first data may be 1. Then, based on different relative position relationships, the second correction coefficient corresponding to the second data and the third correction coefficient corresponding to the third data may be determined in different ways.

[0082] If the relative position relationship is that the second data and the third data are located on the same side of the first data, the correction coefficients of the second data and the third data can be determined based on the preset adjustment coefficient, the preset scaling coefficient and the relative deviation.

[0083] The specific values ​​of the preset adjustment coefficient and the preset scaling coefficient can be determined after debugging according to actual conditions.

[0084] If the relative position relationship is that the second data and the third data are located on different sides of the first data, then no matter it belongs to state 3 or state 4 in Table 1, the third data spans the first data and seriously deviates from the second data. At this time, the third data can be ignored, so it can be determined that the third correction coefficient corresponding to the third data is 0.

[0085] If the relative position relationship is that at least one of the second data and the third data coincides with the first data, the third data can be directly used as the target motion state data of the target object at the next moment without correction. Therefore, the second correction coefficient corresponding to the second data can be determined to be 0, and the third correction coefficient corresponding to the third data can be determined to be 1.

[0086] In this way, based on different relative position relationships, the correction coefficients are determined in different ways, and targeted corrections can be made based on the real-time deviation status.

[0087] In some embodiments, in order to more accurately determine the correction coefficient, the preset adjustment coefficient may be any value not less than 0.5 and not greater than 2, and the preset scaling coefficient may be any value not less than 0.1 and not greater than 0.5.

[0088] Here, the preset adjustment coefficient can be used to correct the situation where the relative deviation is small, and the preset scaling coefficient can be used to correct the situation where the relative deviation is large.

[0089] In this way, the presence of abnormal values ​​in the generated correction coefficient can be avoided by presetting the adjustment coefficient and the preset scaling coefficient, so that the correction coefficient can be determined more accurately.

[0090] In some embodiments, in order to improve the track stability of the target object, the second correction coefficient corresponding to the second data determined according to the preset adjustment coefficient, the preset scaling coefficient and the first relative deviation may include:

[0091] The second correction factor is determined based on the following formula:

[0092]

[0093] Determine the third correction coefficient corresponding to the third data according to the preset adjustment coefficient, the preset scaling coefficient, and the second relative deviation amount, including:

[0094] Determine the third correction coefficient based on the following formula:

[0095]

[0096] where a can be the second correction coefficient, b can be the third correction coefficient, Δ1 can be the first relative deviation amount, Δ2 can be the second relative deviation amount, ω can be the preset adjustment coefficient, α can be the preset scaling coefficient, and e can be the natural constant.

[0097] Here, a can be the second correction coefficient corresponding to the second data, and |Δ1| can be the absolute value of the first relative deviation amount between the second data and the first data. The larger |Δ1| is, the more the second data deviates from the first data. Therefore, it is necessary to decrease the second correction coefficient a of the second data. Based on the above preset adjustment coefficient ω and preset scaling coefficient α, a can decrease as |Δ1| increases and increase as |Δ1| decreases.

[0098] b can be the third correction coefficient corresponding to the third data, and |Δ2| can be the absolute value of the second relative deviation amount between the third data and the first data. The larger |Δ2| is, the more the third data deviates from the first data. Therefore, it is necessary to decrease the second correction coefficient b of the third data. Based on the above preset adjustment coefficient ω and preset scaling coefficient α, b can decrease as |Δ2| increases and increase as |Δ2| decreases.

[0099] Exemplarily, if k > 0, the correction coefficient corresponding to X1 can be 1, and the correction coefficient corresponding to X2 can be The correction coefficient corresponding to X3 can be

[0100] If k < 0, the correction coefficient corresponding to X1 can be 1, and the correction coefficient corresponding to X2 can be The correction coefficient corresponding to X3 can be 0.

[0101] If k = 0, the correction coefficient corresponding to X1 can be 1, the correction coefficient corresponding to X2 can be 0, and the correction coefficient corresponding to X3 can be 1.

[0102] In this way, by making the correction coefficient negatively correlated with the absolute value of the relative deviation amount, data with larger deviations can account for a smaller proportion, and data with smaller deviations can account for a larger proportion, which is beneficial to improving the trajectory stability of the target object.

[0103] Regarding S140, if the first data is the third motion state data, the second data is the first motion state data, and the third data is the second motion state data, then the third motion state data can be corrected based on the first correction coefficient, the first motion state data can be corrected based on the second correction coefficient, and the second motion state data can be corrected based on the third correction coefficient to determine the target motion state data of the target object at the next moment.

[0104] In some embodiments, in order to more accurately determine the target motion state data and thus improve the track stability of the target object, the above S140 may include:

[0105] According to the first correction coefficient, the second correction coefficient, and the third correction coefficient, determine the weights corresponding to the first data, the second data, and the third data respectively.

[0106] Based on the weights, perform weighted calculations on the first data, the second data, and the third data to obtain the target motion state data.

[0107] Among them, in some embodiments, in order to more accurately determine the weights corresponding to the first data, the second data, and the third data respectively, the above step of determining the weights corresponding to the first data, the second data, and the third data respectively according to the first correction coefficient, the second correction coefficient, and the third correction coefficient may include:

[0108] Determine the first weight corresponding to the first data as:

[0109]

[0110] Determine the second weight corresponding to the second data as:

[0111]

[0112] Determine the third weight corresponding to the third data as:

[0113]

[0114] Among them, a may be the second correction coefficient, and b may be the third correction coefficient.

[0115] In this way, through the above process, the weights corresponding to the first data, the second data, and the third data can be determined more accurately.

[0116] In some embodiments, in order to more accurately determine the target motion state data, the above step of performing weighted calculations on the first data, the second data, and the third data based on the weights to obtain the target motion state data may include:

[0117] Determine the target motion state data based on the following formula:

[0118]

[0119] Among them, X can be the target motion state data, X1 can be the first data, X2 can be the second data, and X3 can be the third data.

[0120] Exemplarily, if the preset adjustment coefficient ω = 0.75 and the preset scaling coefficient α = 0.2, the variation curve of the correction coefficient with different relative deviation amounts can be as Figure 2 shown. The variation curves of X1, X2, X3, and X can be as Figure 3 shown. In Figure 3 , the abscissa is time and the ordinate is the motion state data. From Figure 3 the curve change in, it can be seen that in the related art, directly using the filtered data X3 as the curve of the final motion state data has a large fluctuation, so the track stability is poor. However, the curve of the target motion state data X determined based on the target tracking filtering method provided in the embodiments of the present application is significantly smoother, obviously effectively improving the track stability.

[0121] In this way, based on the correction coefficients corresponding to each motion state data, the weights of each motion state data are respectively determined, and then weighted summation is performed, so that the target motion state data of the target object at the next moment can be determined more accurately, thereby effectively improving the track stability of the target object.

[0122] The target tracking filtering method provided by the embodiments of the present application can quickly identify the target track deviation state and set the correction coefficient specifically for the identified deviation state. The implementation process of the entire algorithm has a small calculation amount and a small data storage space occupation, does not require complex calculations, and has strong engineering feasibility. Compared with the track stability optimization algorithm that requires complex calculations in the related art, the algorithm in the target tracking filtering method provided by the embodiments of the present application has greatly improved in terms of real-time performance, algorithm complexity, data calculation amount, etc., and can perform targeted correction compensation in real time and quickly, effectively improving the track stability during the target tracking process.

[0123] Based on the same inventive concept, the embodiments of the present application also provide a target tracking filtering device. The following will be combined with Figure 4 to describe in detail the target tracking filtering device provided by the embodiments of the present application.

[0124] Figure 4 shows a schematic structural diagram of a target tracking filtering device provided by an embodiment of the present application.

[0125] As Figure 4 shown, the target tracking filtering device may include:

[0126] An acquisition module 401 is configured to acquire first motion state data, second motion state data, and third motion state data of a target object. The second motion state data is obtained by performing tracking filtering on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data.

[0127] A first determination module 402 is configured to determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data based on the first motion state data, the second motion state data, and the third motion state data, with any one of the first motion state data, the second motion state data, and the third motion state data as a reference.

[0128] A second determination module 403 is configured to determine correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship.

[0129] A third determination module 404 is configured to determine the target motion state data of the target object at the next moment according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data.

[0130] Thus, the first motion state data, the second motion state data, and the third motion state data of the target object can be acquired, the relative position relationship among the first motion state data, the second motion state data, and the third motion state data can be determined, then according to the relative position relationship, the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data can be determined, and according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, the target motion state data of the target object at the next moment can be determined. Among them, the second motion state data is obtained by performing tracking filtering on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. The relative position relationship among the first motion state data, the second motion state data, and the third motion state data can reflect the real-time track deviation. Therefore, by correcting the first motion state data, the second motion state data, and the third motion state data according to the correction coefficients determined based on the relative position relationship, targeted correction can be performed on the real-time track deviation, and it is not necessary to store a certain amount of data for secondary smoothing processing. Therefore, the track stability of the target object can be effectively improved in real time, and it does not require a large amount of storage space, and the algorithm complexity is relatively low.

[0131] In some embodiments, in order to more accurately determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data, the first determination module 402 may include:

[0132] The first determination sub-module is configured to determine a first relative deviation amount Δ1 = X1 - X2 between a first data X1 and a second data X2, and determine a second relative deviation amount Δ2 = X1 - X3 between the first data X1 and a third data X3.

[0133] The second determination sub-module is configured to determine a product k = Δ1 × Δ2 of the first relative deviation amount and the second relative deviation amount.

[0134] The third determination sub-module is configured to determine a relative position relationship according to the product.

[0135] Wherein, the first data is any one of a first motion state data, a second motion state data, and a third motion state data, the second data is any one of the first motion state data, the second motion state data, and the third motion state data except the first data, and the third data is the data among the first motion state data, the second motion state data, and the third motion state data except the first data and the second data.

[0136] In some embodiments, in order to more clearly determine the relative position relationship of the first motion state data, the second motion state data, and the third motion state data on the coordinate axis, the third determination sub-module may include:

[0137] The first determination unit is configured to, when the product k > 0, determine that the relative position relationship is that the second data and the third data are on the same side of the first data.

[0138] The second determination unit is configured to, when the product k < 0, determine that the relative position relationship is that the second data and the third data are on different sides of the first data.

[0139] The third determination unit is configured to, when the product k = 0, determine that the relative position relationship is that at least one of the second data and the third data coincides with the first data.

[0140] In some embodiments, in order to more effectively correct the motion state data, the second determination module 403 may include:

[0141] The fourth determination sub-module is configured to, when the relative position relationship is that the second data and the third data are on the same side of the first data, determine that the first correction coefficient corresponding to the first data is 1, determine a second correction coefficient corresponding to the second data according to a preset adjustment coefficient, a preset scaling coefficient, and the first relative deviation amount, and determine a third correction coefficient corresponding to the third data according to the preset adjustment coefficient, the preset scaling coefficient, and the second relative deviation amount.

[0142] A fifth determination sub-module, configured to determine that the first correction coefficient corresponding to the first data is 1, the third correction coefficient corresponding to the third data is 0, and determine the second correction coefficient corresponding to the second data according to a preset adjustment coefficient, a preset scaling coefficient, and a first relative deviation amount when the relative position relationship is that the second data and the third data are on different sides of the first data.

[0143] A sixth determination sub-module, configured to determine that both the first correction coefficient corresponding to the first data and the third correction coefficient corresponding to the third data are 1, and the second correction coefficient corresponding to the second data is 0 when at least one of the second data and the third data coincides with the first data in the relative position relationship.

[0144] In some embodiments, to more accurately determine the target motion state data and thus improve the track stability of the target object, the third determination module 404 may include:

[0145] A seventh determination sub-module, configured to determine the weights corresponding to the first data, the second data, and the third data respectively according to the first correction coefficient, the second correction coefficient, and the third correction coefficient.

[0146] A calculation sub-module, configured to perform weighted calculation on the first data, the second data, and the third data based on the weights to obtain the target motion state data.

[0147] In some embodiments, to improve the track stability of the target object, the fourth determination sub-module and the fifth determination sub-module may specifically be used for:

[0148] Determining the second correction coefficient based on the following formula:

[0149]

[0150] The fourth determination sub-module may specifically further be used for:

[0151] Determining the third correction coefficient based on the following formula:

[0152]

[0153] where a is the second correction coefficient, b is the third correction coefficient, Δ1 is the first relative deviation amount, Δ2 is the second relative deviation amount, ω is the preset adjustment coefficient, α is the preset scaling coefficient, and e is the natural constant.

[0154] In some embodiments, to more accurately determine the correction coefficient, the preset adjustment coefficient may be any value not less than 0.5 and not greater than 2, and the preset scaling coefficient may be any value not less than 0.1 and not greater than 0.5.

[0155] In some embodiments, to more accurately determine the weights corresponding to the first data, the second data, and the third data respectively, the seventh determination sub-module may include:

[0156] A fourth determination unit, configured to determine that the first weight corresponding to the first data is:

[0157]

[0158] A fifth determination unit, configured to determine that the second weight corresponding to the second data is:

[0159]

[0160] A sixth determination unit, configured to determine that the third weight corresponding to the third data is:

[0161]

[0162] Wherein, a is the second correction coefficient and b is the third correction coefficient.

[0163] In some embodiments, to more accurately determine the target motion state data, the calculation sub-module may include:

[0164] A calculation unit, configured to determine the target motion state data based on the following formula:

[0165]

[0166] Wherein, X is the target motion state data.

[0167] Figure 5 The schematic structural diagram of an electronic device provided by an embodiment of the present application is shown.

[0168] As Figure 5 shown, the electronic device 5 can implement the structural diagram of an exemplary hardware architecture of an electronic device according to the target tracking filtering method and the target tracking filtering device in the embodiments of the present application. The electronic device may refer to the electronic device in the embodiments of the present application.

[0169] The electronic device 5 may include a processor 501 and a memory 502 storing computer program instructions.

[0170] Specifically, the above-mentioned processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0171] Memory 502 may include a mass storage for data or instructions. By way of example and not limitation, memory 502 may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In a suitable case, memory 502 may include removable or non-removable (or fixed) media. In a suitable case, memory 502 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 502 is a non-volatile solid state memory. In a particular embodiment, memory 502 may include read only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, generally, memory 502 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of the present application.

[0172] The processor 501 reads and executes the computer program instructions stored in the memory 502 to implement any one of the target tracking filtering methods in the above embodiments.

[0173] In one example, the electronic device may further include a communication interface 503 and a bus 504. Among them, as Figure 5 shown, the processor 501, the memory 502, and the communication interface 503 are connected through the bus 504 to complete the communication with each other.

[0174] The communication interface 503 is mainly used to implement the communication between the modules, devices, units, and / or devices in the embodiments of the present application.

[0175] The bus 504 includes hardware, software, or both, and couples components of the electronic device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable bus or a combination of two or more of these. Where appropriate, the bus 504 may include one or more buses. Although embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0176] The electronic device can execute the target tracking filtering method in the embodiments of the present application, so as to implement the combination of Figures 1 to 4 the described target tracking filtering method and apparatus.

[0177] In addition, in combination with the target tracking filtering method in the above embodiments, embodiments of the present application can be implemented by providing a computer storage medium. Computer program instructions are stored on the computer storage medium, and when the computer program instructions are executed by a processor, any one of the target tracking filtering methods in the above embodiments is implemented.

[0178] It should be clear that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0179] The functional blocks shown in the above-described structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments for performing the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via a data signal carried in a carrier wave. A "machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0180] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, can be different from the order in the embodiments, or several steps can be executed simultaneously.

[0181] Aspects of the present application have been described above with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each block in the flowcharts and / or block diagrams, and the combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each block in the block diagrams and / or flowcharts, and the combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0182] As described above, this is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present application.

Claims

1. A target tracking filtering method, characterized in that, The method includes: Obtaining first motion state data, second motion state data, and third motion state data of a target object, where the second motion state data is obtained by performing tracking filtering processing on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. Based on the first motion state data, the second motion state data, and the third motion state data, taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference, determining the relative position relationship among the first motion state data, the second motion state data, and the third motion state data. According to the relative position relationship, determining correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively. According to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data, determining the target motion state data of the target object at the next moment.

2. The method according to claim 1, wherein The determining the relative position relationship among the first motion state data, the second motion state data, and the third motion state data based on the first motion state data, the second motion state data, and the third motion state data, taking any one of the first motion state data, the second motion state data, and the third motion state data as a reference, includes: Determining a first relative deviation amount Δ1 = X1 - X2 between a first data X1 and a second data X2, and determining a second relative deviation amount Δ2 = X1 - X3 between the first data X1 and a third data X3. Determining a product k = Δ1 × Δ2 of the first relative deviation amount and the second relative deviation amount. Determining the relative position relationship according to the product. Wherein, the first data is any one of the first motion state data, the second motion state data, and the third motion state data, the second data is any one of the first motion state data, the second motion state data, and the third motion state data except the first data, and the third data is the data among the first motion state data, the second motion state data, and the third motion state data except the first data and the second data.

3. The method according to claim 2, wherein The determining the relative position relationship according to the product includes: When the product k > 0, determining that the relative position relationship is that the second data and the third data are on the same side of the first data. When the product k < 0, determining that the relative position relationship is that the second data and the third data are on different sides of the first data. When the product k = 0, determining that the relative position relationship is that at least one of the second data and the third data coincides with the first data.

4. The method according to claim 3, characterized in that, The determining the correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship includes: When the relative position relationship is such that the second data and the third data are on the same side of the first data, determine that the first correction coefficient corresponding to the first data is 1, determine the second correction coefficient corresponding to the second data according to the preset adjustment coefficient, the preset scaling coefficient, and the first relative deviation amount, and determine the third correction coefficient corresponding to the third data according to the preset adjustment coefficient, the preset scaling coefficient, and the second relative deviation amount. When the relative position relationship is such that the second data and the third data are on different sides of the first data, determine that the first correction coefficient corresponding to the first data is 1, the third correction coefficient corresponding to the third data is 0, and determine the second correction coefficient corresponding to the second data according to the preset adjustment coefficient, the preset scaling coefficient, and the first relative deviation amount. When the relative position relationship is such that at least one of the second data and the third data coincides with the first data, determine that the first correction coefficient corresponding to the first data and the third correction coefficient corresponding to the third data are both 1, and the second correction coefficient corresponding to the second data is 0.

5. The method according to claim 4, wherein The determining the target motion state data of the target object at the next moment according to the correction coefficient, the first motion state data, the second motion state data, and the third motion state data includes: Determine the weights corresponding to the first data, the second data, and the third data respectively according to the first correction coefficient, the second correction coefficient, and the third correction coefficient. Based on the weights, perform a weighted calculation on the first data, the second data, and the third data to obtain the target motion state data.

6. The method according to claim 4, characterized in that, The determining the second correction coefficient corresponding to the second data according to the preset adjustment coefficient, the preset scaling coefficient, and the first relative deviation amount includes: Determine the second correction coefficient based on the following formula: The determining the third correction coefficient corresponding to the third data according to the preset adjustment coefficient, the preset scaling coefficient, and the second relative deviation amount includes: Determine the third correction coefficient based on the following formula: Where a is the second correction coefficient, b is the third correction coefficient, Δ1 is the first relative deviation amount, Δ2 is the second relative deviation amount, ω is the preset adjustment coefficient, α is the preset scaling coefficient, and e is the natural constant.

7. The method according to claim 6, wherein The preset adjustment coefficient is any value not less than 0.5 and not greater than 2, and the preset scaling coefficient is any value not less than 0.1 and not greater than 0.

5.

8. The method according to claim 5, wherein The determining the weights corresponding to the first data, the second data, and the third data respectively according to the first correction coefficient, the second correction coefficient, and the third correction coefficient includes: Determine the first weight corresponding to the first data as: Determine the second weight corresponding to the second data as: Determine the third weight corresponding to the third data as: Where a is the second correction coefficient and b is the third correction coefficient.

9. The method according to claim 8, wherein Performing weighted calculation on the first data, the second data, and the third data based on the weights to obtain the target motion state data includes: Determining the target motion state data based on the following formula: where X is the target motion state data.

10. A target tracking and filtering device, characterized in that, The device includes: An acquisition module, configured to acquire first motion state data, second motion state data, and third motion state data of a target object, where the second motion state data is obtained by performing tracking filtering processing on the first motion state data, and the third motion state data is obtained by predicting the motion state of the target object at the next moment based on the second motion state data. A first determination module, configured to determine the relative position relationship among the first motion state data, the second motion state data, and the third motion state data based on the first motion state data, the second motion state data, and the third motion state data, with any one of the first motion state data, the second motion state data, and the third motion state data as a reference. A second determination module, configured to determine correction coefficients corresponding to the first motion state data, the second motion state data, and the third motion state data respectively according to the relative position relationship. A third determination module, configured to determine the target motion state data of the target object at the next moment according to the correction coefficients, the first motion state data, the second motion state data, and the third motion state data.

Images