Target fusion method, device, vehicle and storage medium

Abstract

The present invention provides a target fusion method, device, vehicle and storage medium, belonging to the field of intelligent driving technology, and the method includes: obtaining first target information collected by a first vehicle-mounted sensor and second target information collected by a second vehicle-mounted sensor, wherein the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors; when the first target information and the second target information are in the same perception area, determining a fusion condition corresponding to the perception area; when the first target information and the second target information meet the fusion condition, fusing the first target information and the second target information to generate tracking target information. The present invention improves the efficiency and accuracy of target fusion.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and particularly relates to a target fusion method, device, vehicle and storage medium. Background Art

[0002] With the development of science and technology, autonomous driving technology has been widely applied. In autonomous driving technology, the target perception technology is one of the core technologies, and among them, the multi-target fusion technology is the key point of the target perception technology.

[0003] In the existing multi-target fusion technologies, the vehicle-mounted sensing module is used to collect the information of the obstacles in front of the vehicle, and then based on the attribute information of the obstacles in front of the vehicle, the obstacles in front of the vehicle are fused. There is a large amount of obstacle information, and the obstacles are not distinguished, so the matching of the obstacles is not accurate enough. Moreover, the attribute information of the obstacles is extracted through deep learning, which increases the computing power consumption and time consumption, and cannot be applied to the autonomous driving controller that does not support the AI algorithm.

[0004] Therefore, it can be seen that in the existing technology during multi-target fusion, there are technical problems of time-consuming and laborious matching of target obstacles

[0005] and inaccurate matching. Summary of the Invention

[0006] In view of this, it is necessary to provide a target fusion method, device, vehicle and storage medium to solve the technical problems of time-consuming and laborious matching of target obstacles and inaccurate matching in the existing technology during multi-target fusion.

[0007] To solve the above technical problems, on the one hand, the present invention provides a target fusion method, including:

[0008] Obtain the first target information collected by the first vehicle-mounted sensor and the second target information collected by the second vehicle-mounted sensor, where the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors;

[0009] When the first target information and the second target information are in the same perception area, determine the fusion condition corresponding to the perception area;

[0010] When the first target information and the second target information meet the fusion condition, fuse the first target information and the second target information to generate tracking target information.

[0011] As a possible implementation manner of the present invention, in this implementation manner, the step of determining the fusion condition corresponding to the perception area when the first target information and the second target information are in the same perception area includes:

[0012] Determine a first sensing area where the first target object information is located and a second sensing area where the second target object information is located;

[0013] Determine whether the first sensing area and the second sensing area are the same sensing area;

[0014] If so, determine a fusion condition corresponding to the same sensing area;

[0015] If not, do not fuse the first target object information and the second target object information.

[0016] As a possible implementation manner of the present invention, in this implementation manner, the determining a fusion condition corresponding to the same sensing area includes:

[0017] Determine the orientation information and distance information of the same sensing area from the vehicle;

[0018] Based on the orientation information and the distance information, determine a fusion condition corresponding to the same sensing area.

[0019] As a possible implementation manner of the present invention, in this implementation manner, when the first target object information and the second target object information meet the fusion condition, fusing the first target object information and the second target object information to generate tracking target object information includes:

[0020] When the first Euclidean distance between the first target object corresponding to the first target object information and the second target object corresponding to the second target object information in the same coordinate system is less than or equal to a first preset distance threshold, determine that the first target object and the second target object are the same target object;

[0021] When there is no corresponding tracking target object information for the identity identifier of the first target object and / or the identity identifier of the second target object in a preset tracking target object list, fuse the first target object information and the second target object information to generate tracking target object information, and the identity identifier of the tracking target object includes the identity identifier of the first target object and the identity identifier of the second target object.

[0022] As a possible implementation manner of the present invention, in this implementation manner, when there is corresponding tracking target object information for the identity identifier of the first target object and / or the identity identifier of the second target object in a preset tracking target object list, the method includes:

[0023] Update the tracking target object information based on the first target object information and the second target object information.

[0024] As a possible implementation manner of the present invention, in this implementation manner, when the first Euclidean distance is greater than the first preset distance threshold, and the identity identifier of the first target and / or the identity identifier of the second target has corresponding tracking target information in a preset tracking target list, the method includes:

[0025] When the second Euclidean distance between the first target and / or the second target and the tracking target in the same coordinate system is less than or equal to the second preset distance threshold, update the tracking target information based on the first target information and / or the second target information.

[0026] As a possible implementation manner of the present invention, in this implementation manner, when the distance between the same sensing area and the vehicle is greater than the third preset distance threshold, the method includes:

[0027] Update the first target information, the second target information, and the tracking target information based on the historical association relationship among the first target information, the second target information, and the tracking target information.

[0028] On the other hand, the present invention also provides a target fusion device, including:

[0029] A target information acquisition module, configured to acquire first target information collected by a first vehicle-mounted sensor and second target information collected by a second vehicle-mounted sensor, where the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target sensing sensors;

[0030] A fusion condition determination module, configured to determine a fusion condition corresponding to the sensing area when the first target information and the second target information are in the same sensing area;

[0031] A fusion module, configured to fuse the first target information and the second target information to generate tracking target information when the first target information and the second target information meet the fusion condition.

[0032] On the other hand, the present invention also provides a vehicle, including a memory and a processor, where

[0033] The memory is used to store a program;

[0034] The processor is coupled to the memory and is configured to execute the program stored in the memory to implement the steps in the target fusion method in any of the above implementation manners.

[0035] On the other hand, the present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions, and when the programs or instructions are executed by a processor, the steps in the target fusion method described in any of the above implementation manners can be implemented.

[0036] The beneficial effects of the present invention are as follows: The target fusion method provided by the present invention judges the regions where the first target object information and the second target object information are located. When the first target object information and the second target object information are in the same region, corresponding fusion conditions are determined based on this same region. When the first target object information and the second target object information meet the fusion conditions, the first target object information and the second target object information are fused to generate tracking target object information. By dividing the sensing regions of vehicle-mounted sensors and adopting different fusion methods for different regions, the matching efficiency, matching accuracy are improved, and the matching error rate is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0038] Figure 1 It is a schematic flowchart of a target fusion method provided by an embodiment of the present invention;

[0039] Figure 2 It is a schematic flowchart of a method for judging whether to fuse provided by an embodiment of the present invention;

[0040] Figure 3 It is a schematic diagram of a divided region provided by an embodiment of the present invention;

[0041] Figure 4 It is a schematic flowchart of a method for determining fusion conditions provided by an embodiment of the present invention;

[0042] Figure 5 It is a schematic flowchart of a possible implementation method of S103 provided by an embodiment of the present invention;

[0043] Figure 6 It is a schematic diagram of target object fusion provided by an embodiment of the present invention;

[0044] Figure 7 It is a schematic structural diagram of a target fusion device provided by an embodiment of the present invention;

[0045] Figure 8 It is a schematic structural diagram of a vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] In the description of the embodiments of the present invention, unless otherwise specified, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0048] The descriptions such as "first" and "second" involved in the embodiments of the present invention are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Therefore, the technical features defined with "first" and "second" may explicitly or implicitly include at least one such feature.

[0049] Referring to "embodiments" herein means that the specific features, structures, or characteristics described in connection with the embodiments may be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0050] The present invention provides a target fusion method, device, vehicle, and storage medium, which will be described separately below.

[0051] Figure 1 is a schematic flowchart of an embodiment of the target fusion method provided by the present invention, as Figure 1 shown, the target fusion method includes:

[0052] S101, obtaining first target object information collected by a first vehicle-mounted sensor and second target object information collected by a second vehicle-mounted sensor, where the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors;

[0053] S102, when the first target object information and the second target object information are in the same perception area, determining a fusion condition corresponding to the perception area;

[0054] S103, when the first target object information and the second target object information meet the fusion condition, fusing the first target object information and the second target object information to generate tracking target object information.

[0055] The target fusion method provided by the embodiments of the present invention can be applied to the field of autonomous driving, such as target fusion in autonomous driving technology, mainly for identifying and fusing targets in the forward area of the vehicle and maintaining a tracking list.

[0056] In the embodiments of the present invention, the vehicle is equipped with at least two types of target perception sensors, such as an image acquisition sensor and a millimeter-wave radar sensor. The types and models of the image acquisition sensor and the millimeter-wave radar sensor are not limited in the present invention. Optionally, the first vehicle-mounted sensor is an image acquisition sensor, and the second vehicle-mounted sensor is a millimeter-wave radar sensor. The first target information collected by the first vehicle-mounted sensor is the image information of the first target, and the second target information collected by the second vehicle-mounted sensor is the millimeter-wave radar point cloud information of the second target. Among them, the first target and the second target may be the same target or different targets. In particular, when the first target and the second target are the same target, the first target information and the second target information can be fused to obtain the tracking target information.

[0057] In the embodiments of the present invention, as Figure 2 shown, step S102 includes:

[0058] Determine the first perception area of the first target information and the second perception area where the second target information is located;

[0059] Judge whether the first perception area and the second perception area are the same perception area;

[0060] If so, determine the fusion conditions corresponding to the same perception area;

[0061] If not, do not fuse the first target information and the second target information.

[0062] In the embodiments of the present invention, in order to improve the accuracy of target matching of the target fusion method provided by the embodiments of the present invention, the perception areas of the vehicle-mounted sensors are processed in a sub-region manner. Generally, the forward area of the vehicle is processed in a sub-region manner. Among them, the sub-region method can be selected according to the actual situation. As a possible implementation manner of the present invention, as Figure 3 shown, the forward area of the vehicle is divided into sub-regions, including the vehicle's close-range area and the vehicle's medium- and long-range area. Among them, since the accuracy requirement for target fusion in the vehicle's close-range area is relatively high, the vehicle's close-range area is further divided into the left front side area, the directly front area, and the right front side area. It should be understood that the forward area of the vehicle is the area that can be perceived by both the first vehicle-mounted sensor and the second vehicle-mounted sensor.

[0063] For the convenience of description, taking a specific implementation as an example, if the first target object information collected by the first vehicle-mounted sensor is in the left front area and the second target object information collected by the second vehicle-mounted sensor is in the right front area, it indicates that the distance between the first target object and the second target object is far and they are not the same target object. Therefore, there is no need to fuse the first target object information and the second target object information. When the first target object information collected by the first vehicle-mounted sensor is in the left front area and the second target object information collected by the second vehicle-mounted sensor is in the left front area, it indicates that the first target object and the second target object may be the same target object. It is necessary to determine the fusion conditions in this area and then determine whether to fuse the first target object information and the second target object information based on these fusion conditions.

[0064] In the embodiment of the present invention, the tracked target object information refers to the information of the target object existing in front of the vehicle obtained after fusing the target object information collected by the first vehicle-mounted sensor and the second vehicle-mounted sensor. It is necessary to continuously track this target object so that the autonomous driving system can make corresponding responses based on the actions of this target object. Generally, there are multiple target objects in front of the vehicle. Under the perception of the first vehicle-mounted sensor and the second vehicle-mounted sensor, multiple tracked target object information will be generated. These multiple tracked target object information can be included in the tracked target object information list together for subsequent maintenance of the tracked object information.

[0065] In the embodiment of the present application, after dividing the front area of the vehicle into regions, each region corresponds to a tracked target object list. When it is determined that the target object information collected by the vehicle-mounted sensor belongs to a certain perception region, during the subsequent stage of matching the target object information with the tracked target object information in the tracked target object list, only compare the target object information with the tracked target object information in the tracked target object list corresponding to its perception region, rather than comparing it with the entire tracked target object list, which can greatly reduce the workload of target comparison and fusion. Specifically, taking an embodiment as an example, assume that there are M detection targets in a certain frame, N targets for maintaining the tracking list. Among them, M detection targets are divided into the directly forward, left front side, and right front side, assumed to be , , , where . N In the tracking lists, according to the directly forward, left front side, and right front side, assume , , where . The previously widely used matching method is to perform an association operation on the M detection targets with the N tracking targets in sequence, and its calculation dimension is ; After dividing each sub-region, the embodiments of the present invention calculate as . Obviously, the matching method provided by the embodiments of the present invention has less computational complexity and faster computational efficiency.

[0066] In the embodiments of the present invention, the fusion condition of the first target information and the second target information is related to the sensing region where the first target and the second target are located, and the fusion condition of different sensing regions is related to the azimuth relationship and distance information between the sensing region and the vehicle. Specifically, as Figure 4 shown, determining the fusion condition corresponding to the same sensing region includes:

[0067] S401, determining the azimuth information and distance information between the same sensing region and the vehicle;

[0068] S402, determining the fusion condition corresponding to the same sensing region based on the azimuth information and distance information.

[0069] In the embodiments of the present invention, based on the different azimuths and different distances between the sensing region and the vehicle, the target fusion condition in this sensing region may also be different. The target fusion condition refers to the condition for determining that the first target information collected by the first vehicle-mounted sensor and the second target information collected by the second vehicle-mounted sensor are information of the same target.

[0070] In the embodiments of the present invention, for the fusion condition of a certain sensing region, taking a specific embodiment as an example, as Figure 5 shown, when the first target information and the second target information meet the fusion condition, fusing the first target information and the second target information to generate tracking target information includes:

[0071] S501, when the first Euclidean distance between the first target corresponding to the first target information and the second target corresponding to the second target information in the same coordinate system is less than or equal to the first preset distance threshold, determining that the first target and the second target are the same target;

[0072] S502, when there is no corresponding tracking target information for the identity identifier of the first target and / or the identity identifier of the second target in the preset tracking target list, fusing the first target information and the second target information to generate tracking target information, and the identity identifier of the tracking target includes the identity identifier of the first target and the identity identifier of the second target.

[0073] In the embodiments of the present invention, the first preset distance threshold is different for different sensing regions. Generally, the first preset distance threshold corresponding to the close-range region of the vehicle is smaller, and the first preset distance threshold corresponding to the long-range region of the vehicle is larger.

[0074] In an embodiment of the present invention, the first target object information includes the coordinate information of the first target object in the vehicle coordinate system and the identity identification information of the first target object in the first vehicle-mounted sensor, and the second target object information includes the coordinate information of the second target object in the vehicle coordinate system and the identity identification information of the second target object in the second vehicle-mounted sensor. When, in the vehicle coordinate system, the Euclidean distance between the first target object and the second target object calculated based on the coordinates of the first target object and the coordinates of the second target object is less than or equal to the first preset distance threshold, it is determined that the first target object and the second target object are the same target object, that is, the sensor information collected by the first vehicle-mounted sensor and the second vehicle-mounted sensor is of the same target object, and the first target object information and the second target object information can be fused to obtain the tracking target object information. Optionally, before fusing the first target object information and the second target object information, it is possible to first check whether there is a tracking target object corresponding to the identity identification of the first target object and / or the identity identification of the second target object in a preset tracking target object list. Generally, when the first target object and the second target object first appear in the forward area of the vehicle, there is no tracking target object information corresponding to the first target object and the second target object in the tracking object list, then the first target object information and the second target object information can be fused to generate tracking target object information, and the identity identification of the tracking target object includes the identity identification of the first target object and the identity identification of the second target object.

[0075] As a possible implementation manner of the present invention, in this implementation manner, when there is corresponding tracking target object information in the preset tracking target object list for the identity identification of the first target object and / or the identity identification of the second target object, the method includes:

[0076] Updating the tracking target object information based on the first target object information and the second target object information.

[0077] In an embodiment of the present invention, when it is determined that the first target object and the second target object are the same target object, the first target object information and the second target object information are fused. When there is corresponding tracking target object information in the preset tracking target object list for the identity identification of the first target object and / or the identity identification of the second target object, such as Figure 6As shown in the figure, the first vehicle-mounted sensor is a camera, and the second vehicle-mounted sensor is a millimeter-wave radar. For the same target object, the identity identification information in the first target object information collected by the camera for this target object is 1, and the identity identification information in the second target object information collected by the millimeter-wave radar for this target object is 3. Then, the identity identification in the fused tracking target object information is (1, 3). Based on the historical matching relationship, the identity identification of the tracking target object information including (1, 3) is 7. The information of the tracking target object 7 can be directly updated according to the target object information of target object 1 and target object 3. When there is no tracking target object information that simultaneously includes the identity identification (1, 3), check whether there is tracking target object information that separately includes the identity identification 1 or the identity identification 3 in the previous frame of tracking target object list. For example, there is a tracking target object 8 (2, 4) in the previous frame. In the target object information collected by the camera this time, there is a target object with the identity identification of 2, and there is no target object with the identity identification of 4 in the target object information collected by the millimeter-wave radar. Then, calculate the Euclidean distance in the vehicle coordinate system between the target object with the identity identification of 2 collected by the camera and the tracking target object 8 (2, 4). When this Euclidean distance is less than or equal to the preset camera distance threshold, use the tracking target object 8 (2, ) to update the tracking target object 8 (2, 4). Similarly, when there is no target object with the identity identification of 2 in the target object information collected by the camera this time, and there is a target object with the identity identification of 4 in the target object information collected by the millimeter-wave radar, calculate the Euclidean distance in the vehicle coordinate system between the target object with the identity identification of 4 collected by the millimeter-wave radar and the tracking target object 8 (2, 4). When this Euclidean distance is less than or equal to the preset radar distance threshold, use the tracking target object 8 (, 4) to update the tracking target object 8 (2, 4). Optionally, when there is no target object with the identity identification of 2 in the target object information collected by the camera this time, and there is no target object with the identity identification of 4 in the target object information collected by the millimeter-wave radar, do not update the tracking target object 8 (2, 4). Of course, the above embodiments are only a relatively simple possible fusion method of the present application. In actual applications, the camera and the millimeter-wave radar will simultaneously collect the information of multiple target objects, and multiple target objects need to be processed simultaneously. The processing logic is the same as that of the embodiments of the present application, which will not be elaborated here.

[0078] As a possible implementation manner of the present invention, in this implementation manner, when the first Euclidean distance is greater than the first preset distance threshold, and the identity identification of the first target object and / or the identity identification of the second target object has corresponding tracking target object information in the preset tracking target object list, the method includes:

[0079] When the second Euclidean distance between the first target object and / or the second target object and the tracking target object in the same coordinate system is less than or equal to the second preset distance threshold, update the tracking target object information based on the first target object information and / or the second target object information.

[0080] In an embodiment of the present invention, as in the foregoing embodiment, in the previous frame, the first target 1 and the second target 3 are the same target, and the fused tracking target is 7(1, 3). However, in the current frame, if the Euclidean distance between the first target 1 collected by the camera and the second target 3 collected by the millimeter-wave radar in the vehicle coordinate system is greater than the first preset distance threshold, it is possible that in the current frame, the recognition of the target fails, and it is necessary to re-match the first target or the second target corresponding to the tracking target 7(1, 3); taking the second target 3 collected by the millimeter-wave radar as an example, calculate the Euclidean distance between the second target 3 and the tracking target 7(1, 3) in the vehicle coordinate system. When the Euclidean distance is less than or equal to the preset radar distance threshold, update the tracking target 7(1, 3) with the target information of the second target 3 to obtain the tracking target 7(, 3). If the Euclidean distance is greater than the preset radar distance threshold, search for the associated identity identification information of the tracking target 7(1, 3) in the previous few frames. If the associated identity identification information of the tracking target 7(1, 3) in the previous few frames is all (1, 3), and at this time the Euclidean distance between the second target 3 and the tracking target 7(1, 3) is less than or equal to the preset radar distance threshold, update the tracking target 7(1, 3) with the target information of the second target 3 to obtain the tracking target 7(, 3), otherwise, do not update the tracking target 7(1, 3). Similarly, the above method is also used for the first target information collected by the camera, and the principle is the same, so the present invention will not be elaborated here.

[0081] The target fusion method provided by the present invention determines by obtaining and judging the regions where the first target information and the second target information are located. When the first target information and the second target information are in the same region, corresponding fusion conditions are determined based on the same region. When the first target information and the second target information meet the fusion conditions, the first target information and the second target information are fused to generate tracking target information. By dividing the sensing regions of the vehicle-mounted sensors, different fusion methods are adopted for different regions to improve the matching efficiency, matching accuracy, and reduce the matching error rate. At the same time, based on the historical matching relationship of the tracking target, the tracking target is updated and maintained to further improve the matching efficiency and matching speed.

[0082] As a possible implementation manner of the present invention, in this implementation manner, when the distance between the same sensing region and the vehicle is greater than the third preset distance threshold, the method includes:

[0083] Based on the historical association relationship among the first target information, the second target information, and the tracking target information, update the first target information, the second target information, and the tracking target information.

[0084] In an embodiment of the present invention, when the sensing areas where the first target object and the second target object are located are in front of the vehicle and at a relatively long distance from the vehicle, due to possible errors in the information of the first target object collected by the camera and the information of the second target object collected by the millimeter-wave radar in the case of long distance, when updating the information of the tracking target object in the tracking target object list, it can be updated based on the association relationship between the tracking target object and the identity identifiers of the first target object and the second target object in the previous few frames. Specifically, for example, for the tracking target object 7(1, 3), it was at a relatively long distance in the left front of the vehicle in the previous few frames. In the current frame, the tracking target object 7 is at a relatively long distance directly in front of the vehicle, and both are at the junction of the left front and the front. Then, the information of the first target object collected by the camera and the information of the second target object collected by the millimeter-wave radar may be one in the left front and the other directly in front. Since the two are not in the same sensing area, they cannot be fused. Based on the association relationship of the previous few frames of the tracking target object 7(1, 3), the information of the first target object and the second target object associated with it are both 1 and 3. Then, the target object information of the first target object 1 and the second target object 3 can be directly used to update the tracking target object 7(1, 3).

[0085] The embodiment of the present invention uses a sliding window method to maintain and verify the information of the tracking target object, reducing the risk of errors.

[0086] To better implement the target fusion method in the embodiment of the present invention, correspondingly, on the basis of the target fusion method, as Figure 7 shown, the embodiment of the present invention also provides a target fusion device. The target fusion device 700 includes:

[0087] A target object information acquisition module 710, configured to acquire the information of the first target object collected by the first vehicle-mounted sensor and the information of the second target object collected by the second vehicle-mounted sensor, where the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target sensing sensors;

[0088] A fusion condition determination module 720, configured to determine the fusion condition corresponding to the sensing area when the information of the first target object and the information of the second target object are in the same sensing area;

[0089] A fusion module 730, configured to fuse the information of the first target object and the information of the second target object to generate the information of the tracking target object when the information of the first target object and the information of the second target object meet the fusion condition.

[0090] The target fusion device 700 provided in the above embodiment can implement the technical solutions described in the above target fusion method embodiment. The specific implementation principles of the above modules or units can be referred to the corresponding content in the above target fusion method embodiment, and will not be elaborated here.

[0091] The target fusion device provided by the present invention determines the regions where the first target information and the second target information are located. When the first target information and the second target information are in the same region, corresponding fusion conditions are determined based on this same region. When the first target information and the second target information meet these fusion conditions, the first target information and the second target information are fused to generate tracking target information. By dividing the sensing region of the vehicle-mounted sensor and adopting different fusion methods for different regions, the matching efficiency, matching accuracy are improved, and the matching error rate is reduced.

[0092] As Figure 8 shown, the present invention also correspondingly provides a vehicle 800. The vehicle 800 includes a processor 801, a memory 802, and a display 803. Figure 8 Only some components of the vehicle 800 are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0093] In some embodiments, the processor 801 can be a central processing unit (CPU), a microprocessor, or other data processing chips, and is used to run the program code stored in the memory 802 or process data, such as the target fusion method in the present invention.

[0094] In some embodiments, the processor 801 can be a single server or a server group. The server group can be centralized or distributed. In some embodiments, the processor 801 can be local or remote. In some embodiments, the processor 801 can be implemented on a cloud platform. In one embodiment, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination of the above.

[0095] In some embodiments, the memory 802 can be an internal storage unit of the vehicle 800, such as the hard disk or memory of the electronic device in the vehicle 800. In some other embodiments, the memory 802 can also be an external storage device of the vehicle 800, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device in the vehicle 800.

[0096] Furthermore, the memory 802 can also include both the internal storage unit of the vehicle 800 and the external storage device. The memory 802 is used to store the application software installed in the vehicle 800 and various types of data.

[0097] The display 803 can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) toucher, etc. in some embodiments. The display 803 is used to display information of the vehicle 800 and to display a visual user interface. Components 801 - 803 of the vehicle 800 communicate with each other via a system bus.

[0098] In one embodiment, when the processor 801 executes the target fusion program in the memory 802, the following steps can be implemented:

[0099] Obtain first target object information collected by a first vehicle-mounted sensor and second target object information collected by a second vehicle-mounted sensor, where the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors;

[0100] When the first target object information and the second target object information are in the same perception area, determine the fusion conditions corresponding to the perception area;

[0101] When the first target object information and the second target object information meet the fusion conditions, fuse the first target object information and the second target object information to generate tracking target object information.

[0102] It should be understood that when the processor 801 executes the target fusion program in the memory 802, in addition to the above functions, other functions can also be implemented. For details, refer to the description of the corresponding method embodiments above.

[0103] Correspondingly, an embodiment of the present invention further provides a computer-readable storage medium. The computer-readable storage medium is used to store computer-readable programs or instructions. When the programs or instructions are executed by a processor, the steps or functions in the target fusion method provided by the above method embodiments can be implemented.

[0104] Those skilled in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware (such as a processor, a controller, etc.) through a computer program. The computer program can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium is a magnetic disk, an optical disk, a read-only memory, or a random access memory, etc.

[0105] The above has introduced in detail the target fusion method, device, vehicle, and storage medium provided by the present invention. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A target fusion method, characterized in that: include: Acquire first target information collected by a first vehicle-mounted sensor and second target information collected by a second vehicle-mounted sensor, wherein the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors; When the first target object information and the second target object information are in the same perception area, determining a fusion condition corresponding to the perception area; When the first target object information and the second target object information meet the fusion condition, fusing the first target object information and the second target object information to generate tracking target object information; When the first target object information and the second target object information meet the fusion condition, fusing the first target object information and the second target object information to generate tracking target object information includes: When a first Euclidean distance between a first target object corresponding to the first target object information and a second target object corresponding to the second target object information in the same coordinate system is less than or equal to a first preset distance threshold, determining that the first target object and the second target object are the same target object; When the identity of the first target object and / or the identity of the second target object does not have corresponding tracking target information in the preset tracking target list, the first target object information and the second target object information are fused to generate tracking target information, and the identity of the tracking target includes the identity of the first target object and the identity of the second target object.

2. The target fusion method according to claim 1, characterized in that: When the first target object information and the second target object information are in the same perception area, determining a fusion condition corresponding to the perception area includes: Determine a first perception area where the first target object information is located and a second perception area where the second target object information is located; Determining whether the first perception area and the second perception area are the same perception area; If yes, determining a fusion condition corresponding to the same perception area; If not, the first target object information and the second target object information are not fused.

3. The target fusion method according to claim 2, characterized in that: The determining of the fusion condition corresponding to the same perception area includes: Determining the position information and distance information between the same sensing area and the vehicle; A fusion condition corresponding to the same perception area is determined based on the orientation information and the distance information.

4. The target fusion method according to claim 1, characterized in that: When the identity identifier of the first target object and / or the identity identifier of the second target object has corresponding tracking target information in a preset tracking target list, the method includes: The tracking target object information is updated based on the first target object information and the second target object information.

5. The target fusion method according to claim 1, characterized in that: When the first Euclidean distance is greater than a first preset distance threshold, and the identity of the first target object and / or the identity of the second target object has corresponding tracking target information in a preset tracking target list, the method includes: When a second Euclidean distance between the first target and / or the second target and the tracked target in the same coordinate system is less than or equal to a second preset distance threshold, the tracked target information is updated based on the first target information and / or the second target information.

6. The target fusion method according to claim 1, characterized in that: When the distance between the same perception area and the vehicle is greater than a third preset distance threshold, the method includes: The first target information, the second target information, and the tracking target information are updated based on the historical association relationship between the first target information, the second target information, and the tracking target information.

7. A target fusion device, applicable to the target fusion method according to any one of claims 1 to 6, characterized in that: include: A target information acquisition module, used to acquire first target information collected by a first vehicle-mounted sensor and second target information collected by a second vehicle-mounted sensor, wherein the first vehicle-mounted sensor and the second vehicle-mounted sensor are two different types of target perception sensors; a fusion condition determination module, configured to determine a fusion condition corresponding to the perception area when the first target object information and the second target object information are in the same perception area; A fusion module is used to fuse the first target object information and the second target object information to generate tracking target object information when the first target object information and the second target object information meet the fusion condition.

8. A vehicle, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the target fusion method described in any one of claims 1 to 6 above.

9. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the target fusion method described in any one of claims 1 to 6 above.

Images