Object position determination method and apparatus, electronic device, and storage medium

Abstract

The present disclosure provides an object position determination method and device, electronic equipment and storage medium, relates to the technical field of artificial intelligence, in particular to the technical field of roadside perception, computer vision and intelligent transportation. The specific implementation scheme of the object position determination method is as follows: obtaining at least two first initial positions detected by at least two perception devices for a movable object at a first time, at least two first relative position relationships between the at least two perception devices and the movable object, and a category of the movable object; adjusting a predetermined matrix corresponding to the category of the movable object according to each first relative position relationship to obtain each first matrix; fusing the at least two first initial positions based on each first matrix to generate a first fused position; and in response to the difference between the first fused position and the at least two first initial positions being less than a predetermined threshold value, determining that the first fused position is the position of the movable object at the first time.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and more particularly to the fields of roadside perception, computer vision, and intelligent transportation. Specifically, it relates to object location methods, devices, electronic devices, and storage media. Background Technology

[0002] Roadside perception utilizes various sensing devices, such as cameras, millimeter-wave radar, and lidar, to collect images for identification and combine them with roadside edge computing to achieve intelligent perception of traffic participants on that road segment. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for determining the location of an object.

[0004] According to one aspect of this disclosure, a method for determining the location of an object is provided, comprising: acquiring at least two first initial positions detected by at least two sensing devices for a movable object at a first moment and at least two first relative positional relationships between the at least two sensing devices and the movable object; adjusting predetermined matrices corresponding to the categories of the movable object according to each first relative positional relationship to obtain each first matrix; wherein the predetermined matrix indicates the statistical deviation between the perceived position and the actual position of the object for each category in different directions; fusing the at least two first initial positions based on each first matrix to generate a first fused position; and determining the first fused position as the location of the movable object at the first moment in response to the fact that the differences between the first fused position and the at least two first initial positions are all less than a predetermined threshold.

[0005] According to another aspect of this disclosure, an object location determination device is provided, comprising: a first acquisition module, an adjustment module, a first fusion module, and a first determination module.

[0006] The first acquisition module is used to acquire at least two first initial positions detected by at least two sensing devices for a movable object at a first moment and at least two first relative positional relationships between at least two sensing devices and the movable object.

[0007] An adjustment module is used for each first relative positional relationship to adjust the predetermined matrix corresponding to the category of the movable object to obtain each first matrix; wherein, the predetermined matrix indicates the statistical deviation between the perceived position and the actual position of the object for the category in different directions.

[0008] The first fusion module is used to fuse at least two first initial positions based on each first matrix to generate a first fused position.

[0009] The first determining module is configured to determine the first fusion position as the position of the movable object at the first moment in response to the fact that the difference between the first fusion position and at least two first initial positions is less than a predetermined threshold.

[0010] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described above.

[0011] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform the methods described above.

[0012] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described above.

[0013] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0014] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0015] Figure 1 This illustration schematically shows an exemplary system architecture for applying object location determination methods and apparatus according to embodiments of the present disclosure;

[0016] Figure 2 A flowchart illustrating an object location determination method according to an embodiment of the present disclosure is shown schematically.

[0017] Figure 3 A schematic diagram illustrating position fusion based on spatial adjustment according to an embodiment of the present disclosure is shown.

[0018] Figure 4 This illustration schematically shows a data processing diagram of location fusion based on spatial adjustment according to an embodiment of the present disclosure;

[0019] Figure 5 The illustration schematically shows a location fusion based on temporal and spatial adjustments according to an embodiment of the present disclosure;

[0020] Figure 6 A schematic diagram illustrating the determination of timing adjustment parameters according to an embodiment of the present disclosure is shown.

[0021] Figure 7 A schematic diagram illustrating location fusion based on time-series trajectories according to an embodiment of the present disclosure is shown.

[0022] Figure 8 A block diagram schematically illustrates an object position determination apparatus according to an embodiment of the present disclosure; and

[0023] Figure 9 A block diagram of an electronic device suitable for implementing an object location determination method according to an embodiment of the present disclosure is shown schematically. Detailed Implementation

[0024] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0025] In scenarios with complex traffic flow at intersections, multiple roadside sensing devices are typically deployed to detect the location of movable objects along the road. Because the pixel size of each target in an image depends on its relative position to the sensing device, different roadside sensing devices may perceive a different location for the same movable object compared to its actual physical location.

[0026] This bias can cause multiple roadside sensing devices to detect the same movable object at multiple locations at the same time, reducing the accuracy of the roadside sensing devices in identifying the location.

[0027] In view of this, the embodiments of this disclosure utilize the relative positions between different sensing devices and the movable object to dynamically adjust the predetermined statistical deviation corresponding to the category of the movable object, correcting the permissible deviation between the sensed position and the actual position in different directions. This allows the initial positions identified by multiple sensing devices to be correlated with each other, thereby achieving fusion between multiple initial positions. Only when the difference between the fused position and each initial position is less than a predetermined threshold is the fused position determined to be the location of the movable object, further improving the position recognition accuracy of the roadside sensing devices.

[0028] Figure 1 The illustration schematically shows an exemplary system architecture for an object location determination method and apparatus applicable according to embodiments of the present disclosure.

[0029] It is important to note that Figure 1The examples shown are merely examples of system architectures that can be applied to the embodiments of this disclosure, to help those skilled in the art understand the technical content of this disclosure, but do not mean that the embodiments of this disclosure cannot be used in other devices, systems, environments, or scenarios. For example, in another embodiment, an exemplary system architecture to which the object location determination method and apparatus can be applied may include a terminal device, but the terminal device can implement the object location determination method and apparatus provided by the embodiments of this disclosure without interacting with the server.

[0030] like Figure 1 As shown, the system architecture 100 according to this embodiment may include sensing devices 101, 102, 103, 104, a movable object 110, a network 105, and a server 106 configured in the intersection area. The network 105 serves as a medium for providing communication links between the sensing devices 101, 102, 103, 104, and the server 106. The network 105 may include various connection types, such as wired and / or wireless communication links, etc.

[0031] Sensing devices 101, 102, 103, and 104 interact with server 106 via network 105. For example, they can send the collected sensing locations to server 106 via network 105.

[0032] Server 106 can be a server that provides various services, such as analyzing and processing data such as the initial position perceived by the user using sensing devices 101, 102, 103, and 104, and the relative position between the sensing devices and the movable object, and feeding back the processing results (such as the fused position) to the sensing devices.

[0033] The object location determination method provided in this disclosure can generally be executed by server 106. Correspondingly, the object location determination device provided in this disclosure can generally be located in server 106. The object location determination method provided in this disclosure can also be executed by a server or server cluster that is different from server 106 and capable of communicating with sensing devices 101, 102, 103, 104 and / or server 106. Correspondingly, the object location determination device provided in this disclosure can also be located in a server or server cluster that is different from server 106 and capable of communicating with sensing devices 101, 102, 103, 104 and / or server 106.

[0034] For example, when a movable target 110 enters an intersection area, sensing devices 101, 102, 103, and 104 can respectively acquire images of the movable target to obtain its sensing location and the relative position between each sensing device and the movable target. Then, the acquired sensing locations and the relative positions between each sensing device and the movable target are sent to server 106, which fuses the sensing locations to generate a fused location. The fused location is then compared with the initial sensing location, and the final location of the movable object is output. Alternatively, a server or server cluster capable of communicating with sensing devices 101, 102, 103, 104, and / or server 106 can fuse the sensing locations to ultimately determine the location of the movable object.

[0035] The object location determination method provided in the embodiments of this disclosure can generally also be executed by any of the sensing devices 101, 102, 103, and 104.

[0036] For example, when the movable target 110 enters the intersection area, the sensing devices 101, 102, 103, and 104 can respectively acquire images of the movable target to obtain the sensing position and the relative position between each sensing device and the movable target. Then, the acquired sensing position and the relative position between each sensing device and the movable target are sent to one of the sensing devices with position fusion function. By executing the method provided in the embodiments of this disclosure, the location of the movable object is determined.

[0037] It should be understood that Figure 1 The number of sensing devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of sensing devices, networks, and servers can be included.

[0038] In the technical solution disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of user personal information comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and there is no violation of public order and good morals.

[0039] In the technical solution disclosed herein, the user's authorization or consent is obtained before acquiring or collecting the user's personal information.

[0040] Figure 2 A flowchart illustrating an object location determination method according to an embodiment of the present disclosure is shown schematically.

[0041] like Figure 2 As shown, the method includes operations S210 to S240.

[0042] In operation S210, at least two first initial positions detected by at least two sensing devices for the movable object at a first moment and at least two first relative positional relationships between the at least two sensing devices and the movable object are acquired.

[0043] In operation S220, according to the first relative positional relationship, the predetermined matrix corresponding to the category of the movable object is adjusted to obtain each first matrix.

[0044] In operation S230, based on each first matrix, at least two first initial positions are merged to generate a first merged position.

[0045] In operation S240, in response to the fact that the difference between the first fused position and at least two first initial positions is less than a predetermined threshold, the first fused position is determined to be the position of the movable object at the first moment.

[0046] According to embodiments of this disclosure, the sensing device may be a Road Side Unit (RSU) fixedly installed on a road segment. The movable object may be various targets moving in the road segment, such as vehicles, pedestrians, animals, etc.

[0047] According to embodiments of this disclosure, the first initial position can characterize the 2D position obtained by the sensing device by acquiring an image of a movable object and identifying the image, and then the 3D position obtained based on the mapping relationship between 2D space and 3D space.

[0048] According to embodiments of this disclosure, the first relative positional relationship can characterize the direction and distance of the first initial position relative to the installation position of the sensing device.

[0049] In complex traffic environments, such as intersections, movable objects may be obscured by large vehicles, or the images may be low-resolution due to the distance of the movable objects from the sensing devices. These factors can lead to lower accuracy in 2D position recognition, which in turn results in lower accuracy in 3D position mapping to 3D space.

[0050] Therefore, embodiments of this disclosure adjust a predetermined matrix based on the relative positional relationship between each sensing device and the movable object. This predetermined matrix indicates the statistical deviation between the perceived position and the actual position of objects of the same category as the movable target in different directions.

[0051] The predetermined matrix can be a covariance matrix, in which the elements represent the variance of the perceived position and the actual position in the direction of observation of the movable object relative to the sensing device, and the variance in the direction perpendicular to the observation direction.

[0052] In some embodiments, a mapping relationship between relative positional relationships and adjustment ranges can be pre-configured, and then the relative positional relationship between the installation location of each sensing device and the initial position of the movable object can be established.

[0053] The adjustment range can be calculated using any mathematical and statistical method based on the relative positional relationships between multiple related sensing locations and sensing devices and movable objects collected in the actual application scenario, which will not be elaborated here.

[0054] The adjustment range represents the degree of adjustment to the allowable deviation between the perceived location and the actual location. In practical applications, the accuracy of identifying targets far from the sensing device is often poor, leading to multiple sensing devices identifying different locations for the same target at the same time. Therefore, the adjustment range is generally greater than 0, meaning that the allowable deviation range needs to be expanded to achieve location correlation.

[0055] For example, if the initial position of the movable object is due north relative to the installation position of the sensing device, and the distance is 10m, the corresponding adjustment range can be to increase the variance of the movable object relative to the observation direction of the sensing device by a factor of n. Therefore, by increasing the variance value corresponding to the observation direction in the predetermined threshold by a factor of n, the updated first matrix is ​​obtained.

[0056] The variance between the perceived position and the actual position represented by the elements in each of the updated first matrices in the observation direction of the movable object relative to the sensing device is increased by a factor of n. This can be understood as increasing the allowable deviation range of the first initial position identified by each sensing device. In this way, two first initial positions that could not be spatially associated can be associated and then position fusion can be performed, thereby correcting the 3D positioning deviation caused by inaccurate image recognition position.

[0057] Since the adjustment range is also based on mathematical statistics, there may be errors. Therefore, the difference between the fused position and each initial position can be calculated separately. Only when the difference is less than a predetermined threshold is the first fused position determined as the position of the movable object at the first moment.

[0058] This embodiment utilizes the relative positions between different sensing devices and movable objects to dynamically adjust a predetermined statistical deviation corresponding to the category of the movable object. This corrects the permissible deviation between the sensed position and the actual position in different directions, enabling the initial positions identified by multiple sensing devices to be correlated and thus achieving fusion of multiple initial positions. Only when the difference between the fused position and each initial position is less than a predetermined threshold is the fused position determined to be the location of the movable object, further improving the position recognition accuracy of the roadside sensing devices.

[0059] The following is for reference. Figures 3-7 In conjunction with specific embodiments, Figure 2 The method shown will be further explained.

[0060] According to embodiments of this disclosure, adjusting predetermined matrices corresponding to the categories of movable objects based on each first relative positional relationship to obtain each first matrix may include the following operations: determining each first spatial adjustment parameter corresponding to each first relative positional relationship; adjusting predetermined matrices based on each first spatial adjustment parameter to obtain each first matrix.

[0061] In some embodiments, the adjustment range can be a spatial adjustment parameter corresponding to a single direction, or it can be a spatial adjustment parameter corresponding to all directions. The spatial adjustment parameters corresponding to all directions can be the same or different.

[0062] Therefore, the mapping relationship between relative positional relationships and spatial adjustment parameters can be pre-configured, and then the relative positional relationship between the installation position of each sensing device and the initial position of the movable object can be determined.

[0063] For example, if the initial position of the movable object is due north relative to the installation position of the sensing device and the distance is 10m, the corresponding spatial adjustment parameters may include: spatial adjustment parameter a1 of the movable object relative to the observation direction of the sensing device and spatial adjustment parameter b1 perpendicular to the observation direction.

[0064] Figure 3 A schematic diagram illustrating position fusion based on spatial adjustment according to an embodiment of the present disclosure is shown.

[0065] like Figure 3 As shown, this embodiment 300 includes a position association diagram 300A before spatial adjustment of the predetermined matrix and a position association diagram 300B after spatial adjustment of the matrix.

[0066] In the position association diagram 300A before spatial adjustment of the predetermined matrix, C1 represents the sensing device 301 installed at position A, and C2 represents the sensing device 302 installed at position B. 01 This represents the statistical deviation range 310 between the first initial position detected by sensing device 301 and the actual position. 02 This represents the statistical deviation range of 320 between the first initial position detected by sensing device 302 and the actual position. It can be seen that A... 01 With A 02 It is impossible to correlate them. This means that even if sensing devices 301 and 302 detect the same object and its position at the same time, according to the predetermined matrix indicating the statistical deviation range, A 01 With A 02They cannot be linked; the device's perception still indicates two different locations.

[0067] However, after adjustment based on spatial adjustment parameters, in the position association diagram 300B after spatial adjustment of the predetermined matrix, A1 represents the adjusted deviation range 311 between the first initial position detected by the sensing device 301 and the true position. This deviation range 311 is the adjustment range between A1 and the actual position. 01 The variance of the movable object relative to the viewing direction of the sensing device is increased by a1 times, and the variance perpendicular to the viewing direction is increased by b1 times. Similarly, this can also be applied to A. 02 Adjustments are made to obtain A2. A2 represents the adjusted deviation range 321 between the first initial position detected by the sensing device 302 and the actual position.

[0068] It can be seen that the adjusted A1 and A2 can be associated, indicating that when sensing device 301 and sensing device 302 detect the position of the same object at the same time, according to the adjusted deviation range indicated by the adjusted first matrix, A1 and A2 can be associated, and the result of device sensing can be a position.

[0069] According to embodiments of this disclosure, by configuring spatial adjustment parameters corresponding to relative positional relationships, a predetermined matrix is ​​dynamically adjusted based on changes in the relative position between the movable object and the sensing device. This increases the likelihood that the sensing positions identified by multiple sensing devices for movable objects far from or occluded by the sensing device can be correlated with each other, further improving the accuracy of sensing device fusion positioning.

[0070] Figure 4 The illustration shows a schematic diagram of location fusion based on spatial adjustment according to an embodiment of the present disclosure.

[0071] like Figure 4 As shown, in embodiment 400, firstly, a predetermined matrix A0411 is determined based on the category 410 of the movable object. Then, spatial adjustment parameters (a1, b1) 421a are determined based on the relative position R1 between sensing device C1 and the movable object, and spatial adjustment parameters (a2, b2) 421b are determined based on the relative position R2 between sensing device C2 and the movable object. Next, the predetermined matrix A0411 is adjusted based on the spatial adjustment parameters (a1, b1) 421a to obtain matrix A1422a. The predetermined matrix A0411 is adjusted based on the spatial adjustment parameters (a2, b2) 421b to obtain matrix A2422b.

[0072] According to embodiments of this disclosure, each first matrix is ​​obtained by adjusting a predetermined matrix according to each first spatial adjustment parameter, including the following operations: extracting the initial deviation of each direction from the predetermined matrix; adjusting the initial deviation of each direction according to the first spatial adjustment parameter of each direction to obtain the first target deviation of each direction; and updating the predetermined matrix according to the first target deviation of each direction to obtain the first matrix.

[0073] According to embodiments of this disclosure, each direction includes a first direction and a second direction. The first direction represents the observation direction of the movable object relative to the sensing device, and the second direction represents a direction perpendicular to the observation direction. The first spatial adjustment parameters include first spatial adjustment parameters for each direction. For example, spatial adjustment parameter a corresponding to the observation direction α1 of the movable object relative to the sensing device and spatial adjustment parameter b corresponding to the direction α2 perpendicular to the observation direction.

[0074] For example, the predetermined matrix can be a two-dimensional covariance matrix. The initial deviations in each direction extracted from the predetermined matrix can include the initial deviation λ1 corresponding to the observation direction α1 of the movable object relative to the sensing device and the initial deviation λ2 corresponding to the direction α2 perpendicular to the observation direction.

[0075] Then, based on the first spatial adjustment parameters for each direction, the initial deviations for each direction are adjusted accordingly to obtain the first target deviation aλ1 corresponding to the observation direction α1 of the movable object relative to the sensing device. The first target deviation bλ2 corresponds to the direction α2 perpendicular to the observation direction. The predetermined matrix is ​​then updated. When updating the predetermined matrix, the elements representing directions remain unchanged; only the initial deviations for each direction are updated to aλ1 and bλ2 for the corresponding directions.

[0076] In some embodiments, the predetermined matrix can also be a three-dimensional covariance matrix, which, compared to a two-dimensional covariance matrix, adds an initial deviation in the direction perpendicular to the plane containing α1α2. Correspondingly, when configuring the mapping relationship between relative positional relationships and spatial adjustment parameters, the spatial adjustment parameter in this direction can be added. The adjustment operation is similar to that of the two-dimensional covariance matrix and will not be described in detail here.

[0077] According to embodiments of this disclosure, spatial relaxation is performed on the movable object relative to the observation direction of the sensing device and the direction perpendicular to the observation direction, respectively. This reduces the impact of perception deviation caused by the different relative positions of the sensing device and the movable object on position association, and further improves the accuracy of fused position.

[0078] like Figure 4As shown, the position S2 (x2, y2) 412b detected by sensing device C2 and the position S1 412a (x1, y1) detected by sensing device C1 are fused based on matrices A1 422a and A2 422b to generate the fused position S. 12 431.

[0079] According to embodiments of this disclosure, fusing at least two first initial positions based on each first matrix to generate a first fused position includes: averaging the first target deviations in each direction in each first matrix to obtain the first average deviations in each direction; and fusing at least two first initial positions based on the first average deviations in each direction to generate the first fused position.

[0080] According to embodiments of this disclosure, the first matrix indicates a first target deviation of the first initial position of the movable object from its actual position in different directions.

[0081] For example: In the observation direction of the movable object relative to the sensing device, the first target deviation σ1 corresponding to the sensing device C1 is extracted from matrix A1, and the first target deviation σ2 corresponding to the sensing device C2 is extracted from matrix A2.

[0082] The first average deviation σ in this direction can be calculated according to equation (1):

[0083] (1)

[0084] Then, calculate the first fusion position coordinate x in this direction according to equation (2):

[0085] (2)

[0086] The calculation method for the first fusion position coordinate y, which is perpendicular to the viewing direction, is the same as the method for calculating x, and will not be elaborated here.

[0087] According to embodiments of this disclosure, by averaging the matrices obtained after spatial relaxation of multiple sensing devices in different directions, the influence of the relative positions of each sensing device and the movable object on the position association is comprehensively corrected, thereby improving the accuracy of the fused position.

[0088] Finally, operation S430 can be performed for the fusion position S. 12 431 can calculate the difference with positions S1412a and S2412b respectively. When the difference is less than the threshold, operation S431 is performed to determine the object position as the fused position S. 12 If the difference is greater than or equal to the threshold, operation S432 is executed to combine the time-series trajectory fusion.

[0089] According to embodiments of this disclosure, the method may further include the following operations: in response to the difference between the first fused position and at least two first initial positions being greater than or equal to a predetermined threshold, acquiring at least two second initial positions detected by at least two sensing devices for the movable object at a second time, at least two second relative positional relationships between the at least two sensing devices and the movable object, and the moving speed of the movable object; obtaining an estimated position of the movable object at the second time based on the moving speed of the movable object and the first fused position; fusing the estimated position with at least two second initial positions based on each second relative positional relationship to generate a second fused position; and determining the second fused position as the position of the movable object at the second time in response to the difference between the second fused position and at least two second initial positions being less than a predetermined threshold.

[0090] According to embodiments of this disclosure, the second moment is an adjacent moment that is later than the first moment.

[0091] Figure 5 The illustration shows a schematic diagram of location fusion based on temporal and spatial adjustments according to an embodiment of the present disclosure.

[0092] like Figure 5 As shown, this embodiment 500 includes a position association diagram 500A before spatial and temporal adjustments are made to the predetermined matrix and a position association diagram 500B after spatial and temporal adjustments are made to the matrix.

[0093] In the position association diagram 500A before spatial and temporal adjustments to the predetermined matrix, C1 represents sensing device 301 installed at position A, and C2 represents sensing device 302 installed at position B. 01 'Indicates the statistical deviation range 310 between the second initial position detected by sensing device 301 and the true position'. A 02 'This represents the statistical deviation range of 320° between the second initial position detected by sensing device 302 and the actual position.' It can be seen that A 01 'with A 02 'It is impossible to correlate them. This means that even if sensing devices 301 and 302 detect the same object and its position at the same time, according to the predetermined matrix indicating the statistical deviation range, A 01 'with A 02 'It is impossible to associate them; the device's perception result is still two locations.'

[0094] However, after adjustment based on spatial adjustment parameters, A3 represents the adjusted deviation range 311' between the second initial position detected by sensing device 301 and the true position. This deviation range 311' is the adjustment range between A and the actual position. 01The variance of the movable object relative to the viewing direction of the sensing device is increased by a1' times, and the variance perpendicular to the viewing direction is increased by b1' times. Similarly, this can also be applied to A. 02 After adjustment, A4 is obtained. A4 represents the adjusted deviation range 321' between the first initial position detected by sensing device 302 and the actual position. As can be seen in 500B, after spatial relaxation, A4 and A3 still cannot be correlated.

[0095] Therefore, in the 500A, the estimated position ES of the movable object at time t is calculated by combining the fused position at time t-1 with the additional moving velocity of the movable object. t 510.

[0096] In the position association diagram 500B after spatial and temporal adjustments to the predetermined matrix, it can be seen that the deviation ranges 511, A4, and A3 after temporal and spatial adjustments can be associated, indicating that when sensing devices 301 and 302 detect the position of the same object at the same time, the deviation ranges 511, A4, and A3 after temporal and spatial adjustments can be associated, and the result of device perception can be a position.

[0097] According to embodiments of this disclosure, the predetermined matrix is ​​relaxed in both spatial and temporal dimensions, thereby enabling dynamic adjustment of the predetermined matrix based on changes in the relative position between the movable object and the sensing device and changes in the temporal trajectory of the movable object. This increases the likelihood that the sensing positions identified by multiple sensing devices for movable objects far from the sensing devices or occluded movable objects can be correlated with each other, further improving the accuracy of the sensing device fusion positioning.

[0098] Unlike spatial adjustment parameters, temporal adjustment parameters are not pre-configured, but are calculated based on the fused and estimated positions of the mobile target at different times.

[0099] For example, obtaining the target time series adjustment parameters based on the difference between the fused position and the estimated position at each time step can include the following operations: obtaining the position difference at each time step based on the fused position and the estimated position at each time step; and processing the position difference at each time step to generate the target time series adjustment parameters.

[0100] Figure 6 A schematic diagram illustrating the determination of timing adjustment parameters according to an embodiment of the present disclosure is shown.

[0101] like Figure 6 As shown, based on S1 detected by sensing device C1 at time t-2 t-2 S2 detected by 610 and sensing device C2 at t-2 t-2620 is fused to obtain the fusion position S12. t-2 630. Then, based on the moving speed of the movable object, and based on the fusion position S12 t-2 630 Calculate the estimated position ES of the movable object at time t-1. t-1 640. Then, the S1 detected by sensing device C1 at time t-1 is... t-1 611 and sensing device C2 detected S2 at t-1 t-1 621 is fused to obtain the fusion position S12 t- 1631.

[0102] Next, based on the moving speed of the movable object, and based on the fusion position S12 t-1 631 Calculate the estimated position ES of a movable object at time t. t 641. Then, the S1 detected by sensing device C1 at time t is... t 612 and sensing device C2 detected S2 at t t 622 is fused to obtain the fusion position S12. t 632.

[0103] Finally, based on the fusion position S12 at time t-1 t-1 Estimated positions ES at times t31 and t-1 t-1 The estimated position at time 640 is calculated, and the position difference Pd1 is calculated at time 651. The fused position S12 is based on time t. t Estimated positions ES at times 632 and t t The estimated position of 641 is used to calculate the position difference Pd2 652. Furthermore, based on the position differences Pd1 651 and Pd2 652, any mathematical operation can be performed, such as averaging or logarithmic operations, to calculate the target timing adjustment parameter 660.

[0104] According to embodiments of this disclosure, based on the difference between the fused position and the estimated position at each time point, the temporal fluctuation between the fused position and the estimated position in the temporal movement trajectory of the movable object is obtained. The temporal fluctuation is then used to temporally relax the estimated position at adjacent time points, correcting the fluctuation deviation of the temporal trajectory and further improving the correlation between the perceived position and the temporal trajectory.

[0105] According to embodiments of this disclosure, fusing the estimated position with at least two second initial positions based on each second relative positional relationship to generate a second fused position may include the following operations: adjusting predetermined matrices corresponding to the categories of the movable object based on each second relative positional relationship between at least two sensing devices and the movable object to obtain each second matrix; determining the third matrix with the largest second target deviation in each direction from each second matrix; adjusting the third matrix according to target timing adjustment parameters to generate a fourth matrix; and fusing the estimated position with at least two second initial positions based on the fourth matrix and each second matrix to generate the second fused position.

[0106] In some embodiments, the target time-series adjustment parameters can be calculated based on the method described above, using the estimated position and fused position at each moment in the real-time movement trajectory.

[0107] In some embodiments, the target timing adjustment parameter may also be obtained by statistical calculation based on the trajectory position and fusion position at each moment in the historical movement trajectories of multiple objects of the same category as the movable object.

[0108] For example, the movable object can be a vehicle, and multiple sensing devices can be fixedly deployed in a certain intersection area. Then, the difference value between the trajectory position and the fused position of each vehicle passing through the intersection area during a historical period can be calculated at each historical moment. Next, using any statistical method, the calculated difference value of each vehicle at each historical moment is statistically calculated to obtain the target time series adjustment parameters.

[0109] In some embodiments, any mathematical operation can be used to fuse the estimated position with at least two second initial positions based on the fourth matrix and each of the second matrices to generate a second fused position.

[0110] In some embodiments, the second fusion position can also be calculated based on a method similar to that described above in formulas (1) and (2).

[0111] According to embodiments of this disclosure, the matrix with the largest deviation of the second target is corrected by using the target time series adjustment parameter, which reduces the impact of time series trajectory deviation on position fusion and achieves dual relaxation of the estimated position in terms of time and space, thereby further improving the accuracy of fusion positioning.

[0112] According to embodiments of this disclosure, based on each second relative positional relationship, a predetermined matrix corresponding to the category of the movable object is adjusted to obtain each second matrix. This may include the following operations: determining each second spatial adjustment parameter corresponding to each second relative positional relationship; and adjusting the predetermined matrix according to each second spatial adjustment parameter to obtain each second matrix.

[0113] For example, the second relative positional relationship can characterize the relative direction and relative distance between at least two sensing devices and a movable object at a second moment.

[0114] Based on the changes in the relative position between the sensing device and the movable object, the spatial adjustment parameters are dynamically matched to adjust the predetermined matrix, thereby realizing the dynamic correction of statistical deviations and further improving the correlation accuracy of the sensing positions of multiple sensing devices for the same movable object.

[0115] According to embodiments of this disclosure, adjusting a predetermined matrix according to each second spatial adjustment parameter to obtain each second matrix may include the following operations: extracting initial deviations in each direction from the predetermined matrix; adjusting the initial deviations in each direction according to the second spatial adjustment parameters in each direction to generate second target deviations in each direction; and updating the predetermined matrix according to the second target deviations in each direction to obtain each second matrix.

[0116] For each sensing position detected by each sensing device at each moment, spatial relaxation is required. Therefore, the process of adjusting the predetermined matrix based on each second relative position relationship to obtain each second matrix is ​​the same as the process of adjusting the predetermined matrix based on each first relative position relationship, and will not be elaborated here.

[0117] During temporal relaxation, since the estimated position calculated based on the fusion position at time t-1 is introduced, when performing position fusion, the difference between the fusion position at each time before time t-2 and the estimated position at each time can be used as follows: Figure 6 The method described herein calculates the target temporal adjustment parameters. Then, based on the target temporal adjustment parameters, the third matrix that determines the largest deviation of the second target in each direction among the second matrices is adjusted, thereby achieving temporal and spatial relaxation of the estimated position.

[0118] For example, the third matrix A can be adjusted according to the target timing adjustment parameters according to formula (3). max Generate the fourth matrix A es :

[0119] A es =γA max (3)

[0120] Where γ represents the target timing adjustment parameter.

[0121] According to embodiments of this disclosure, based on target timing adjustment parameters, the third matrix with the largest deviation of the second target in each direction in each second matrix is ​​adjusted, thereby realizing temporal and spatial relaxation of the estimated position, increasing the possibility that the positions of each sensing device are correlated in time and space, and further improving the accuracy of fusion positioning.

[0122] According to embodiments of this disclosure, fusing the estimated position with at least two second initial positions based on the fourth matrix and each of the second matrices to generate a second fused position may include the following operations: fusing the estimated position with the second initial position detected by the first sensing device based on the second matrix and the fourth matrix corresponding to the first sensing device to generate an intermediate fused position; and fusing the intermediate fused position with the second initial position detected by the second sensing device based on the second matrix and the fourth matrix corresponding to the second sensing device to generate a second fused position.

[0123] Figure 7 The illustration shows a schematic diagram of location fusion based on time-series trajectories according to an embodiment of the present disclosure.

[0124] like Figure 7 As shown, in embodiment 700, based on the moving speed of the movable object, according to the fusion position S at time t-1 t-1 701, the estimated position ES at time t is calculated. t 702.

[0125] According to embodiments of this disclosure, the estimated position ES is calculated based on a second matrix and a fourth matrix corresponding to the first sensing device. t 702 is fused with the second initial position (position St1703) detected by the first sensing device to generate an intermediate fused position Es. t It may include the following operations: averaging the second target deviation in each direction of the second matrix corresponding to the first sensing device and the third target deviation in each direction of the fourth matrix to obtain the second average deviation in each direction; and fusing the estimated position with the second initial position detected by the first sensing device based on the second average deviation in each direction to generate an intermediate fused position.

[0126] For example: it can be based on matrix A3 and matrix A es First, calculate the second average deviation in each direction using formula (1) described above. Then, calculate the intermediate fusion position using formula (2). Further details are omitted here.

[0127] Then, based on the second matrix and the fourth matrix corresponding to the second sensing device, the intermediate fusion position is fused with the second initial position detected by the second sensing device to generate the second fusion position. This may include the following operations: averaging the second target deviation in each direction in the second matrix corresponding to the second sensing device and the third target deviation in each direction in the fourth matrix to obtain the third average deviation in each direction; and fusing the intermediate fusion position with the second initial position detected by the second sensing device based on the third average deviation in each direction to generate the second fusion position.

[0128] For example: it can be based on matrix A4 and matrix A es The second fusion position is calculated according to formulas (1) and (2) described above.

[0129] In this embodiment of the disclosure, since the target adjustment timing parameter is a number greater than 1, matrix A es It is calculated based on the target deviation matrix with the largest value among all matrices. Therefore, matrix A es The deviation among the second matrices obtained at time t is the largest. When calculating the intermediate fusion position and the second fusion position, A is directly used. es The matrix can be calculated as a matrix with the estimated position or intermediate fusion position.

[0130] According to embodiments of this disclosure, the estimated position is cascaded and fused with the initial position detected by each sensing device based on the dynamically corrected matrix, thereby gradually correcting the initial position of each sensing device, reducing the amount of data processing at each step, and improving the efficiency of fusion positioning.

[0131] Figure 8 A block diagram of an object location determination apparatus according to an embodiment of the present disclosure is shown schematically.

[0132] like Figure 8 As shown, the position determination device 900 may include: a first acquisition module 810, an adjustment module 820, a first fusion module 830, and a first determination module.

[0133] The first acquisition module 810 is used to acquire at least two first initial positions detected by at least two sensing devices for a movable object at a first moment and at least two first relative positional relationships between at least two sensing devices and the movable object.

[0134] The adjustment module 820 is used for each first relative position relationship to adjust the predetermined matrix corresponding to the category of the movable object to obtain each first matrix; wherein, the predetermined matrix indicates the statistical deviation between the perceived position and the actual position of the object for the category in different directions.

[0135] The first fusion module 830 is used to fuse at least two first initial positions based on each first matrix to generate a first fused position.

[0136] The first determining module 840 is configured to determine the first fused position as the position of the movable object at the first moment in response to the fact that the difference between the first fused position and at least two first initial positions is less than a predetermined threshold.

[0137] According to embodiments of this disclosure, the adjustment module may include a first determining submodule and a first adjusting submodule. The first determining submodule is used to determine each first spatial adjustment parameter corresponding to each first relative positional relationship. The first adjusting submodule is used to adjust predetermined matrices according to each first spatial adjustment parameter to obtain each first matrix.

[0138] According to embodiments of this disclosure, the first spatial adjustment parameters include first spatial adjustment parameters for each direction; each direction includes a first direction and a second direction, the first direction representing the observation direction of the movable object relative to the sensing device, and the second direction representing a direction perpendicular to the observation direction. The first adjustment submodule may include a first extraction unit, a first adjustment unit, and a first update unit.

[0139] The first extraction unit is used to extract the initial deviations in each direction from a predetermined matrix. The first adjustment unit is used to adjust the initial deviations in each direction according to the first spatial adjustment parameters in each direction, thereby obtaining the first target deviations in each direction. The first update unit is used to update the predetermined matrix according to the first target deviations in each direction, thereby obtaining the first matrix.

[0140] According to embodiments of this disclosure, the first matrix indicates the target deviation between the first initial position and the true position of the movable object in different directions. The first fusion module may include a first processing unit and a first fusion unit.

[0141] The first processing unit is used to average the first target deviations in each direction of each first matrix to obtain the first average deviation in each direction. The first fusion unit is used to fuse at least two first initial positions based on the first average deviations in each direction to generate a first fused position.

[0142] According to embodiments of this disclosure, the apparatus may further include a second acquisition module, an estimation module, a second fusion module, and a second determination module.

[0143] The second acquisition module is used to acquire, in response to the fact that the difference between the first fusion position and at least two first initial positions is greater than or equal to a predetermined threshold, at least two second initial positions detected by at least two sensing devices for the movable object at a second time, at least two second relative positional relationships between at least two sensing devices and the movable object, and the moving speed of the movable object; wherein the second time is an adjacent time that is later than the first time.

[0144] The estimation module is used to obtain the estimated position of the movable object at the second moment based on the moving speed of the movable object and the first fused position.

[0145] The second fusion module is used to fuse the estimated position with at least two second initial positions based on the second relative positional relationships to generate a second fused position.

[0146] The second determining module is used to determine the second fusion position as the position of the movable object at the second time moment in response to the difference between the second fusion position and at least two second initial positions being less than a predetermined threshold.

[0147] According to embodiments of this disclosure, the second fusion module may include: a second adjustment submodule, a second determination submodule, a third adjustment submodule, and a second fusion submodule.

[0148] The second adjustment submodule is used for each second relative position relationship, adjusting the predetermined matrix corresponding to the category of the movable object to obtain each second matrix.

[0149] The second determination submodule is used to determine the third matrix with the largest deviation of the second target in each direction from each second matrix.

[0150] The third adjustment submodule is used to adjust the third matrix according to the target time series adjustment parameters to generate the fourth matrix; wherein, the target time series adjustment parameters are obtained based on the difference between the fused position at each time and the estimated position at each time.

[0151] The second fusion submodule is used to fuse the estimated position with at least two second initial positions based on the fourth matrix and each of the second matrices to generate a second fused position.

[0152] According to embodiments of this disclosure, the second adjustment submodule may include the following operations: a first determining unit and a second adjustment unit.

[0153] The first determining unit is used to determine each second spatial adjustment parameter corresponding to each second relative position relationship. The second adjusting unit is used to adjust the predetermined matrix according to each second spatial adjustment parameter to obtain each second matrix.

[0154] According to embodiments of this disclosure, the second spatial adjustment parameters include second spatial adjustment parameters for each direction; each direction includes a first direction and a second direction, the first direction representing the observation direction of the movable object relative to the sensing device, and the second direction representing a direction perpendicular to the observation direction. The second adjustment unit includes: an extraction subunit, an adjustment subunit, and an update subunit.

[0155] The extraction sub-unit is used to extract the initial deviations in each direction from a predetermined matrix. The adjustment sub-unit is used to adjust the initial deviations in each direction according to the second space adjustment parameters in each direction, thereby generating the second target deviations in each direction. The update sub-unit is used to update the predetermined matrix according to the second target deviations in each direction, thereby obtaining the second matrices.

[0156] According to embodiments of this disclosure, the second fusion module further includes a difference calculation submodule and a generation submodule. The difference calculation submodule is used to obtain the position difference at each time step based on the fused position at each time step and the estimated position at each time step. The generation submodule is used to process the position differences at each time step and generate target time-series adjustment parameters.

[0157] According to embodiments of this disclosure, at least two sensing devices include a first sensing device and a second sensing device. The second fusion submodule may include a second fusion unit and a third fusion unit.

[0158] The second fusion unit is used to fuse the estimated position with the second initial position detected by the first sensing device based on the second matrix and the fourth matrix corresponding to the first sensing device to generate an intermediate fused position.

[0159] The third fusion unit is used to fuse the intermediate fusion position with the second initial position detected by the second sensing device based on the second matrix and the fourth matrix corresponding to the second sensing device, so as to generate a second fusion position.

[0160] According to embodiments of this disclosure, the second fusion unit may include a first averaging subunit and a first fusion subunit. The first averaging subunit is used to average the second target deviations in each direction of the second matrix corresponding to the first sensing device and the third target deviations in each direction of the fourth matrix, respectively, to obtain second average deviations in each direction. The first fusion subunit is used to fuse the estimated position with the second initial position detected by the first sensing device based on the second average deviations in each direction, to generate an intermediate fused position.

[0161] According to embodiments of this disclosure, the third fusion unit may include: a second averaging subunit and a second fusion subunit. The second averaging subunit is used to average the second target deviations in each direction of the second matrix corresponding to the second sensing device and the third target deviations in each direction of the fourth matrix, respectively, to obtain a third average deviation in each direction. The second fusion subunit is used to fuse the intermediate fusion position with the second initial position detected by the second sensing device based on the third average deviation in each direction, to generate a second fusion position.

[0162] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0163] According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method described above.

[0164] According to embodiments of the present disclosure, a non-transitory computer-readable storage medium stores computer instructions, wherein the computer instructions are used to cause a computer to perform the method described above.

[0165] According to an embodiment of this disclosure, a computer program product includes a computer program that, when executed by a processor, implements the method described above.

[0166] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0167] like Figure 9 As shown, device 900 includes a computing unit 901, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 902 or a computer program loaded from storage unit 908 into random access memory (RAM) 903. RAM 903 may also store various programs and data required for the operation of device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / output (I / O) interface 905 is also connected to bus 904.

[0168] Multiple components in device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, etc.; output unit 907, such as various types of monitors, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0169] The computing unit 901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as the object location determination method. For example, in some embodiments, the object location determination method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the object location determination method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the object location determination method by any other suitable means (e.g., by means of firmware).

[0170] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0171] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0172] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0173] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0174] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0175] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, distributed system servers, or servers incorporating blockchain technology.

[0176] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0177] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining the location of an object, comprising: Acquire at least two first initial positions detected by at least two sensing devices for a movable object at a first moment, at least two first relative positional relationships between the at least two sensing devices and the movable object, and the category of the movable object; Based on each of the first relative positional relationships, a predetermined matrix corresponding to the category of the movable object is adjusted to obtain each first matrix; wherein, the predetermined matrix indicates the statistical deviation between the perceived position and the actual position of the object for the category in different directions; wherein, the first matrix indicates the target deviation between the first initial position and the actual position of the movable object in different directions; The first target deviation in each direction in each of the first matrices is averaged to obtain the first average deviation in each direction; The at least two initial positions are fused based on the first average deviation in each direction to generate a first fused position; and In response to the fact that the difference between the first fused position and each of the at least two first initial positions is less than a predetermined threshold, the first fused position is determined to be the position of the movable object at the first moment.

2. The method according to claim 1, wherein, The step of adjusting the predetermined matrix corresponding to the category of the movable object according to each of the first relative positional relationships to obtain each first matrix includes: Determine each first spatial adjustment parameter corresponding to each of the first relative positional relationships; Each first matrix is ​​obtained by adjusting the predetermined matrix according to the first space adjustment parameters.

3. The method according to claim 2, wherein, The first spatial adjustment parameters include first spatial adjustment parameters for each direction; each direction includes a first direction and a second direction; the first direction represents the viewing direction of the movable object relative to the sensing device; The second direction represents the direction perpendicular to the observation direction; The step of adjusting the predetermined matrix according to each of the first spatial adjustment parameters to obtain each first matrix includes: Extract the initial deviations in each direction from the predetermined matrix; Based on the first spatial adjustment parameters for each direction, the initial deviations for each direction are adjusted accordingly to obtain the first target deviation for each direction; and The predetermined matrix is ​​updated based on the first target deviation in each direction to obtain the first matrix.

4. The method according to any one of claims 1 to 3, further comprising: In response to the fact that the differences between the first fusion position and the at least two first initial positions are all greater than or equal to a predetermined threshold, at least two second initial positions detected by at least two sensing devices for the movable object at a second time, at least two second relative positional relationships between the at least two sensing devices and the movable object, and the moving speed of the movable object are obtained; wherein, the second time is an adjacent time later than the first time. Based on the moving speed of the movable object and the first fused position, the estimated position of the movable object at the second moment is obtained; Based on the respective second relative positional relationships, the estimated position is fused with at least two second initial positions to generate a second fused position; and In response to the fact that the difference between the second fused position and the at least two second initial positions is less than the predetermined threshold, the second fused position is determined to be the position of the movable object at the second moment.

5. The method according to claim 4, wherein, The step of fusing the estimated position with at least two second initial positions based on the respective second relative positional relationships to generate a second fused position includes: Based on the second relative positional relationships between at least two sensing devices and the movable object, the predetermined matrices corresponding to the categories of the movable object are adjusted respectively to obtain each second matrix; From each of the second matrices, determine the third matrix that has the largest deviation of the second target in each direction; The third matrix is ​​adjusted according to the target time-series adjustment parameters to generate a fourth matrix; wherein the target time-series adjustment parameters are obtained based on the difference between the fusion position and the estimated position at each time step; and Based on the fourth matrix and each of the second matrices, the estimated position is fused with the at least two second initial positions to generate the second fused position.

6. The method according to claim 5, wherein, The step of adjusting the predetermined matrices corresponding to the categories of the movable objects based on the respective second relative positional relationships to obtain each second matrix includes: Determine each second spatial adjustment parameter corresponding to each of the second relative positional relationships; and Each second space adjustment parameter is used to adjust the predetermined matrix to obtain each second matrix.

7. The method according to claim 6, wherein, The second spatial adjustment parameters include second spatial adjustment parameters in each direction; each direction includes: a first direction and a second direction; the first direction represents the viewing direction of the movable object relative to the sensing device; the second direction represents a direction perpendicular to the viewing direction; The step of adjusting the predetermined matrix according to each of the second spatial adjustment parameters to obtain each second matrix includes: Extract the initial deviations in each direction from the predetermined matrix; Based on the second spatial adjustment parameters for each direction, the initial deviations for each direction are adjusted accordingly to generate the second target deviations for each direction; and The predetermined matrix is ​​updated based on the second target deviation in each direction to obtain each second matrix.

8. The method according to claim 7, wherein, The target time-series adjustment parameters are obtained based on the difference between the fused position and the estimated position at each time step, including: The positional differences at each time step are obtained based on the fused position and the estimated position at each time step; and The positional differences at each time point are processed to generate the target timing adjustment parameters.

9. The method according to claim 5, wherein, The at least two sensing devices include a first sensing device and a second sensing device; The step of fusing the estimated position with the at least two second initial positions based on the fourth matrix and each of the second matrices to generate the second fused position includes: Based on the second matrix and the fourth matrix corresponding to the first sensing device, the estimated position is fused with the second initial position detected by the first sensing device to generate an intermediate fused position; and Based on the second matrix and the fourth matrix corresponding to the second sensing device, the intermediate fusion position is fused with the second initial position detected by the second sensing device to generate the second fusion position.

10. The method according to claim 9, wherein, The step of fusing the estimated position with the second initial position detected by the first sensing device based on the second matrix and the fourth matrix corresponding to the first sensing device to generate an intermediate fused position includes: The second target deviations in each direction of the second matrix corresponding to the first sensing device and the third target deviations in each direction of the fourth matrix are averaged to obtain the second average deviation in each direction; and The estimated position is fused with the second initial position detected by the first sensing device based on the second average deviation in each direction to generate the intermediate fused position.

11. The method according to claim 9, wherein, The step of fusing the intermediate fusion position with the second initial position detected by the second sensing device based on the second matrix and the fourth matrix corresponding to the second sensing device to generate the second fusion position includes: The second target deviations in each direction of the second matrix corresponding to the second sensing device and the third target deviations in each direction of the fourth matrix are averaged to obtain the third average deviation in each direction; and The intermediate fusion position is fused with the second initial position detected by the second sensing device based on the third average deviation in each direction to generate the second fusion position.

12. An object position determination device, comprising: The first acquisition module is used to acquire at least two first initial positions detected by at least two sensing devices for a movable object at a first moment, at least two first relative positional relationships between the at least two sensing devices and the movable object, and the category of the movable object; An adjustment module is used to adjust a predetermined matrix corresponding to the category of the movable object for each of the first relative positional relationships, thereby obtaining each first matrix; wherein the predetermined matrix indicates the statistical deviation between the perceived position and the actual position of the object for the category in different directions; wherein the first matrix indicates the target deviation between the first initial position and the actual position of the movable object in different directions. The first fusion module includes: A first processing unit is configured to average the first target deviations in each direction of each of the first matrices to obtain the first average deviation in each direction; and A first fusion unit is configured to fuse the at least two first initial positions based on a first average deviation in each direction to generate a first fused position; and A first determining module is configured to determine the first fused position as the position of the movable object at the first moment in response to the fact that the difference between the first fused position and the at least two first initial positions is less than a predetermined threshold.

13. The apparatus according to claim 12, wherein, The adjustment module includes: A first determining submodule is used to determine each first spatial adjustment parameter corresponding to each of the first relative positional relationships; and The first adjustment submodule is used to adjust the predetermined matrix according to each of the first spatial adjustment parameters to obtain each first matrix.

14. The apparatus according to claim 13, wherein, The first spatial adjustment parameters include first spatial adjustment parameters for each direction; each direction includes a first direction and a second direction; the first direction represents the viewing direction of the movable object relative to the sensing device; The second direction represents the direction perpendicular to the observation direction; The first adjustment submodule includes: The first extraction unit is used to extract the initial deviations in each direction from the predetermined matrix; The first adjustment unit is used to adjust the initial deviation of each direction according to the first spatial adjustment parameters of each direction, thereby obtaining the first target deviation of each direction; and The first update unit is used to update the predetermined matrix according to the first target deviation in each direction to obtain the first matrix.

15. The apparatus according to any one of claims 12 to 14, wherein, The device further includes: The second acquisition module is configured to, in response to the fact that the differences between the first fused position and the at least two first initial positions are all greater than or equal to a predetermined threshold, acquire at least two second initial positions detected by at least two sensing devices for the movable object at a second time, at least two second relative positional relationships between the at least two sensing devices and the movable object, and the moving speed of the movable object; wherein the second time is an adjacent time later than the first time. An estimation module is used to obtain an estimated position of the movable object at the second moment based on the moving speed of the movable object and the first fused position; The second fusion module is used to fuse the estimated position with at least two second initial positions based on each of the second relative positional relationships to generate a second fused position; and The second determining module is configured to determine the second fused position as the position of the movable object at the second moment in response to the difference between the second fused position and the at least two second initial positions being less than the predetermined threshold.

16. The apparatus according to claim 15, wherein, The second fusion module includes: The second adjustment submodule is used to adjust the predetermined matrix corresponding to the category of the movable object for each of the second relative positional relationships, so as to obtain each second matrix. The second determining submodule is used to determine the third matrix with the largest second target deviation in each direction from each of the second matrices; The third adjustment submodule is used to adjust the third matrix according to the target time-series adjustment parameters to generate a fourth matrix; wherein the target time-series adjustment parameters are obtained based on the difference between the fusion position and the estimated position at each time step; and The second fusion submodule is used to fuse the estimated position with the at least two second initial positions based on the fourth matrix and each of the second matrices to generate the second fused position.

17. The apparatus according to claim 16, wherein, The second adjustment submodule includes: The first determining unit is configured to determine each second spatial adjustment parameter corresponding to each of the second relative positional relationships; and The second adjustment unit is used to adjust the predetermined matrix according to each of the second spatial adjustment parameters to obtain each second matrix.

18. The apparatus according to claim 17, wherein, The second spatial adjustment parameters include second spatial adjustment parameters in each direction; each direction includes: a first direction and a second direction; the first direction represents the viewing direction of the movable object relative to the sensing device; the second direction represents a direction perpendicular to the viewing direction; The second adjustment unit includes: Extract sub-units for extracting initial deviations in each direction from the predetermined matrix; An adjustment subunit is configured to adjust the initial deviations in each direction according to the second spatial adjustment parameters in each direction, thereby generating the second target deviations in each direction; and An update sub-unit is used to update the predetermined matrix according to the second target deviation in each direction, thereby obtaining each of the second matrices.

19. The apparatus according to claim 16, wherein, The second fusion module also includes: The difference calculation submodule is used to obtain the position difference at each time step based on the fused position and the estimated position at each time step; and A generation submodule is used to process the positional differences at each time point and generate the target timing adjustment parameters.

20. The apparatus according to claim 16, wherein, The at least two sensing devices include a first sensing device and a second sensing device; The second fusion submodule includes: The second fusion unit is used to fuse the estimated position with the second initial position detected by the first sensing device based on the second matrix and the fourth matrix corresponding to the first sensing device to generate an intermediate fused position; as well as The third fusion unit is used to fuse the intermediate fusion position with the second initial position detected by the second sensing device based on the second matrix and the fourth matrix corresponding to the second sensing device, so as to generate the second fusion position.

21. The apparatus according to claim 20, wherein, The second fusion unit includes: The first averaging subunit is used to average the second target deviation in each direction of the second matrix corresponding to the first sensing device and the third target deviation in each direction of the fourth matrix, respectively, to obtain the second average deviation in each direction; and The first fusion subunit is used to fuse the estimated position with the second initial position detected by the first sensing device based on the second average deviation of each direction to generate the intermediate fused position.

22. The apparatus according to claim 20, wherein, The third fusion unit includes: The second averaging subunit is used to average the second target deviation in each direction of the second matrix corresponding to the second sensing device and the third target deviation in each direction of the fourth matrix, respectively, to obtain the third average deviation in each direction; and The second fusion subunit is used to fuse the intermediate fusion position with the second initial position detected by the second sensing device based on the third average deviation in each direction, so as to generate the second fusion position.

23. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method according to any one of claims 1-11.

24. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-11.

25. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-11.

Images