Object positioning method, device, electronic device and storage medium

Abstract

The embodiment of the present application provides an object positioning method, device, electronic device and storage medium, the method includes: determining a first detection area and a second detection area within a preset range around a vehicle; determining multiple vehicle-side position information based on a target detection algorithm; detecting the first detection area based on a target detection algorithm and a first obstacle confidence threshold, and detecting the second detection area based on a target detection algorithm and a second obstacle confidence threshold through a roadside device to obtain multiple roadside position information; matching multiple vehicle-side positions with multiple roadside position information to obtain multiple matching combinations; inputting the first matching position information and the second matching position information in the multiple matching combinations into a position prediction model, performing position prediction processing, and obtaining multiple target position information corresponding to multiple target objects. The object positioning method provided by the embodiment of the present disclosure can improve detection efficiency and reduce missed detection and false detection.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an object positioning method, device, system and storage medium. Background Art

[0002] As automobile companies and Internet companies have started the research and development of intelligent connected vehicles, the most core technology is autonomous driving technology, and object perception and positioning in autonomous driving technology is also a crucial technology. At present, in the field of autonomous driving perception, the main sensors used include lidar. Lidar has become the main sensor for high-level autonomous driving due to its advantages of wide field of view, weak light dependence, and high ranging accuracy. However, most of the current mainstream autonomous driving solutions use 3D target detection algorithms based on deep learning, which have high requirements for computing power, high cost, and few chips that meet automotive regulations, which need to be solved urgently.

[0003] In order to avoid the problem of low positioning efficiency caused by the large amount of computing power required in the process of object perception and positioning, it is necessary to provide a technical solution to solve the problem of insufficient efficiency in the existing technology. Summary of the invention

[0004] In view of the defects in the prior art, the embodiments of the present disclosure provide an object positioning method, device, electronic device and storage medium, which can improve detection efficiency and reduce missed detection and false detection.

[0005] An embodiment of the present application provides an object positioning method, the method comprising: determining a first detection area and a second detection area within a preset range around a vehicle; wherein the first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area other than the first detection area in the preset range; based on a target detection algorithm, determining multiple vehicle-side position information; through a roadside device, detecting the first detection area based on the target detection algorithm and a first obstacle confidence threshold, and detecting the second detection area based on the target detection algorithm and the second obstacle confidence threshold, to obtain multiple roadside position information; matching the position information of each vehicle-side position information in the multiple vehicle-side position information with each roadside position information in the multiple roadside position information, to obtain multiple groups of matching combinations; determining multiple target objects corresponding to the multiple matching combinations; inputting the first matching position information and the second matching position information in the multiple matching combinations into a position prediction model, performing position prediction processing, to obtain multiple target position information corresponding to the multiple target objects.

[0006] Specifically, the roadside location information includes first roadside location information and second roadside location information; through the roadside equipment, the first detection area is detected based on the target detection algorithm and the first obstacle confidence threshold, and the second detection area is detected based on the target detection algorithm and the second obstacle confidence threshold to obtain multiple roadside location information, including: based on the target detection algorithm, the first detection area is detected to obtain multiple first candidate location information; the obstacle confidence of the first candidate location information is determined; if the obstacle confidence is greater than the first obstacle confidence threshold, the first candidate location information is determined to be the first roadside location information; based on the target detection algorithm and the second obstacle confidence threshold, the second detection area is detected to obtain multiple roadside location information, including: based on the target detection algorithm, the second detection area is detected to obtain multiple second candidate location information; the obstacle confidence of the second candidate location information is determined; if the obstacle confidence is greater than the second obstacle confidence threshold, the second candidate location information is determined to be the second roadside location information.

[0007] Specifically, the first obstacle confidence threshold is smaller than the second obstacle confidence threshold.

[0008] Specifically, position information matching is performed on each vehicle-end position information among multiple vehicle-end position information and each roadside position information among multiple roadside position information to obtain multiple groups of matching combinations, including: determining the current vehicle-end position information among multiple vehicle-end position information; position information matching is performed on the current vehicle-end position information and each roadside position information among multiple roadside position information to obtain a current matching combination that is successfully matched; traversing each vehicle-end position information among multiple vehicle-end position information to obtain multiple groups of matching combinations that are successfully matched.

[0009] Specifically, the current vehicle-side position information is matched with each roadside position information in multiple roadside position information to obtain a current matching combination that is successfully matched, including: determining the matching distance information between the current vehicle-side position information and each roadside position information in multiple roadside position information to obtain multiple matching distance information; determining the target matching distance information in the multiple matching distance information, the target matching distance information being the minimum matching distance information in the multiple matching distance information; when the target matching distance information is less than a preset distance threshold, determining that the current vehicle-side position information and the roadside position information corresponding to the target matching distance information are successfully matched; determining the successfully matched current vehicle-side position information and the roadside position information as the first matching position information and the second matching position information, respectively; and determining the matching combination based on the first matching position information and the second matching position information.

[0010] Specifically, the position prediction model is a model based on the Kalman filter algorithm.

[0011] Specifically, the roadside equipment includes any one of a laser radar, a millimeter wave radar and a radio frequency identification sensor.

[0012] Accordingly, an embodiment of the present application provides an object positioning device, characterized in that the device includes:

[0013] An area determination module is used to determine a first detection area and a second detection area within a preset range around the vehicle; wherein the first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area within the preset range except the first detection area;

[0014] A first position module, configured to detect a first detection area based on a target detection algorithm and a first obstacle confidence threshold, and obtain a plurality of vehicle-side position information;

[0015] A second position module is used to detect the second detection area based on the target detection algorithm and the second obstacle confidence threshold to obtain multiple roadside position information;

[0016] A vehicle-side position module, used to determine a plurality of vehicle-side position information based on a plurality of vehicle-side position information and a plurality of roadside position information;

[0017] An acquisition module, used to acquire multiple roadside location information sent by roadside equipment;

[0018] A matching module, used for matching each vehicle-side position information in the plurality of vehicle-side position information with each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations;

[0019] A target determination module, used to determine multiple target objects corresponding to multiple matching combinations;

[0020] The position prediction module is used to input the first matching position information and the second matching position information in the multiple matching combinations into the position prediction model, perform position prediction processing, and obtain multiple target position information corresponding to the multiple target objects.

[0021] Specifically, the first position module is further used to: detect the first detection area based on the target detection algorithm to obtain multiple first candidate position information; determine the obstacle confidence of the first candidate position information; if the obstacle confidence is greater than the first obstacle confidence threshold, determine that the first candidate position information is the vehicle-side position information;

[0022] The second position module is also used to: detect the second detection area based on the target detection algorithm to obtain multiple second candidate position information; determine the obstacle confidence of the second candidate position information; if the obstacle confidence is greater than the second obstacle confidence threshold, determine that the second candidate position information is roadside position information.

[0023] Specifically, the first obstacle confidence threshold is smaller than the second obstacle confidence threshold.

[0024] Specifically, the matching module is also used to: determine the current vehicle-end position information among multiple vehicle-end position information; perform position information matching between the current vehicle-end position information and each road-side position information among multiple road-side position information to obtain a current matching combination that is successfully matched; traverse each vehicle-end position information among multiple vehicle-end position information to obtain multiple groups of matching combinations that are successfully matched.

[0025] Specifically, the matching module is also used to: determine the matching distance information between the current vehicle-side position information and each roadside position information in multiple roadside position information to obtain multiple matching distance information; determine the target matching distance information in the multiple matching distance information, the target matching distance information is the minimum matching distance information in the multiple matching distance information; when the target matching distance information is less than a preset distance threshold, determine that the current vehicle-side position information and the roadside position information corresponding to the target matching distance information are successfully matched; determine the successfully matched current vehicle-side position information and roadside position information as the first matching position information and the second matching position information, respectively; determine the matching combination based on the first matching position information and the second matching position information.

[0026] Specifically, the position prediction model is a model based on the Kalman filter algorithm.

[0027] Specifically, the roadside equipment includes any one of a laser radar, a millimeter wave radar and a radio frequency identification sensor.

[0028] Accordingly, an embodiment of the present disclosure provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the above-mentioned object positioning method.

[0029] Accordingly, an embodiment of the present disclosure provides a computer-readable storage medium, in which at least one instruction, at least one program, code set or instruction set is stored, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the above-mentioned object positioning method.

[0030] The embodiments of the present application have the following beneficial effects:

[0031] (1) Reduce missed detections by appropriately lowering the obstacle output threshold for areas of interest or areas that have a greater impact on driving, and appropriately increase the obstacle output threshold for areas of no interest or areas that have a lesser impact on driving to reduce false detections;

[0032] (2) By matching the determined position information of the vehicle-mounted camera and the roadside sensing device, and predicting the target position of the target object based on the matched position information, it is possible to avoid position prediction based on a single position information, thereby improving the positioning accuracy of the target position;

[0033] (3) When the proportion of matched objects is low, the confidence of obstacles is reduced to avoid missed detection in complex environments or under poor detection conditions, thereby ensuring the comprehensiveness of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 This is a schematic diagram of an application scenario of an object positioning method provided in an embodiment of the present application;

[0036] Figure 2 It is a first flow chart of an object positioning method provided by an embodiment of the present application;

[0037] Figure 3 is a second flow chart of an object positioning method provided in an embodiment of the present application;

[0038] Figure 4 is a third flow chart of an object positioning method provided in an embodiment of the present application;

[0039] Figure 5 is a third flow chart of an object positioning method provided in an embodiment of the present application;

[0040] Figure 6 is a structural schematic diagram of an object positioning device provided in an embodiment of the present application;

[0041] Figure 7 It is a hardware structure block diagram of a server of an object positioning method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical scheme and advantages of the present application clearer, the embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiment is only one embodiment of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0043] The "embodiment" referred to herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present application. In the description of the embodiments of the present application, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "top", "bottom", etc. is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device / system or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present application. The terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may include one or more of the features explicitly or implicitly. Moreover, the terms "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" and "having" / "for" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system / apparatus, product or apparatus comprising a series of steps or units / modules is not necessarily limited to those steps or units / modules explicitly listed, but may include other steps or units / modules not explicitly listed or inherent to such process, method, product or apparatus.

[0044] The following is a specific embodiment of an object positioning method provided by the present application. Figure 1 , Figure 1 Schematic diagram of an application scenario of an object positioning method provided in an embodiment of the present application. Figure 1 As shown, the application environment may include a vehicle 101, a vehicle-mounted camera 102 arranged on the vehicle 101, a body controller 103 arranged on the vehicle, and a roadside perception device 104 arranged on the roadside.

[0045] In an optional embodiment, the on-board camera 102 on the vehicle 101 can be used to obtain image data of multiple objects around the vehicle 101, where the objects may be pedestrians, other vehicles other than the vehicle, roadside railings and other obstacles that interfere with the driving of the vehicle, and the image data of the objects may be expressed as information such as the object's position information, posture information, speed information, etc.

[0046] In an optional embodiment, the roadside sensing device 104 is installed on the roadside, and the roadside sensing device 104 can be used to collect point cloud data of multiple objects around the vehicle and vehicle identification information, and the point cloud data of each object can be used to determine a roadside position information of the object. Further, the roadside sensing device 104 can upload the collected point cloud data to the body controller 103 of the vehicle corresponding to the vehicle identification information based on the vehicle identification information.

[0047] In an optional embodiment, the vehicle body controller 103 can communicate with the vehicle-mounted camera 102 via a controller area network (CAN) bus. The roadside sensing device 104 can upload the acquired point cloud data to the vehicle body controller 103 via vehicle to X (V2X) wireless communication technology.

[0048] In addition, it should be noted that Figure 1 What is shown is only one application environment of the object positioning method provided by the present disclosure. In actual application, other application environments may also be included.

[0049] The following introduces an exemplary process of an object positioning method provided by the present application. Figure 2 It is a first flow chart of an object positioning method provided in an embodiment of the present application. The executor of an object positioning method may be vehicle 101. This specification provides method or process operation steps as shown in the embodiment or flow chart, but may include more or fewer operation steps based on conventional or non-creative labor. The sequence of steps listed in the embodiment is only one of many execution sequences and does not represent the only execution sequence. In actual execution, the method or process sequence shown in the embodiment or the drawings may be executed or executed in parallel (for example, in a parallel processor or multi-threaded processing environment). Specifically, Figure 2 As shown, the method includes:

[0050] S201: Determine a first detection area and a second detection area within a preset range around the vehicle.

[0051] The first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area other than the first detection area in the preset range. The preset range around the vehicle can be a range that can be detected by the on-board laser radar. Step S201 can be continuously executed based on a preset frequency, and the first detection area and the second detection area can be continuously updated.

[0052] In a specific implementation, the first detection area may include a vehicle passable area and a sidewalk area, and the second detection area may be other areas of the detection area except the first detection area. Map information within a preset range around the vehicle can be determined from map information, and multiple vehicle passable areas, sidewalk areas and other areas can be determined from the map information within the preset range, and the first detection area and the second detection area can be determined on this basis. Specifically, the first detection area may also include the lane or driving area where the vehicle is located in real time.

[0053] In another specific implementation, the traffic density within a preset range around the vehicle can be obtained based on high-precision map information and traffic statistics data, and the area within the preset range around the vehicle where the traffic density is greater than the preset density value is determined as the first detection area, and the area within the detection area where the traffic density is less than or equal to the preset density value is determined as the second detection area.

[0054] Specifically, step S201 may include: sending information of the first detection area and the second detection area to the roadside equipment.

[0055] S202: Determine multiple vehicle-side position information based on the target detection algorithm.

[0056] Specifically, the vehicle-mounted LiDAR can detect the environment within a preset range around the vehicle and obtain a LiDAR point cloud, where the detected targets may include multiple objects or objects within a preset range around the vehicle. The vehicle-side computing platform can perform real-time ground segmentation and non-ground point cloud obstacle clustering on the LiDAR point cloud by running traditional 3D target detection algorithms, thereby calculating the location information of the obstacle, that is, the vehicle-side location information. In addition, feature extraction can be performed on the point cloud within the target detection frame, and the vehicle-side location information corresponding to the obstacle target detection frame that is obviously misdetected can be eliminated based on the actual spatial characteristics of the point cloud of the clustered cluster.

[0057] S203: Using roadside equipment, detect a first detection area based on a target detection algorithm and a first obstacle confidence threshold, and detect a second detection area based on a target detection algorithm and a second obstacle confidence threshold to obtain multiple roadside location information.

[0058] Specifically, the roadside equipment can be a detection device installed at the roadside, and can be any one of a laser radar, a millimeter wave radar, and a radio frequency identification sensor. The roadside equipment can obtain point cloud data of multiple objects around the vehicle, and the point cloud data of each object can be used to determine a roadside position information of the object. For example, the high-performance computing platform at the roadside runs a deep learning-based 3D target detection algorithm, such as pointpillars, second, and pointRCNN, to detect the category, position, and length, width, and height parameters of the target detection box of the obstacle end-to-end, where the position information is the roadside position information. It can be understood that the object determined based on the vehicle-mounted laser radar and the object determined based on the roadside perception device can be the same or different.

[0059] The following is based on Figure 3 Step S203 is further described. Figure 3 As shown in the figure, Figure 3 In the illustrated embodiment, the roadside location information may include first roadside location information and second roadside location information, and the vehicle-side device may send information of the first detection area and the second detection area to the roadside device. Figure 3 The illustrated process further introduces step 203 with the roadside equipment as the execution subject, that is, how the vehicle determines multiple roadside location information through the roadside equipment. The exemplary process may include:

[0060] S301: Detect a first detection area based on a target detection algorithm to obtain a plurality of first candidate position information.

[0061] S302: Determine the obstacle confidence level of the first candidate position information.

[0062] Specifically, the target detection algorithm can be run through the roadside equipment to detect multiple objects in the first detection area, obtain multiple first candidate position information corresponding to the multiple objects, and output obstacle confidences corresponding to the multiple first detected objects.

[0063] S303: If the obstacle confidence is greater than a first obstacle confidence threshold, determine that the first candidate position information is the first roadside position information.

[0064] Specifically, the first obstacle confidence threshold may be a percentage value in the range of 0%-100%, for example, 50%. In an implementation where the first obstacle confidence is 50%, the first candidate position information corresponding to the obstacle confidence greater than 50% may be determined as the first roadside position information.

[0065] S304: Detecting the second detection area based on the target detection algorithm to obtain a plurality of second candidate position information;

[0066] S305: Determine the obstacle confidence level of the second candidate position information.

[0067] Specifically, the target detection algorithm can be run through the roadside equipment to detect multiple objects in the second detection area, obtain multiple second candidate position information corresponding to the multiple objects, and output obstacle confidences corresponding to the multiple second detected objects.

[0068] S306: If the obstacle confidence is greater than the second obstacle confidence threshold, determine that the second candidate location information is the second roadside location information.

[0069] Specifically, the second obstacle confidence threshold may be greater than the first obstacle confidence threshold. The second obstacle confidence threshold may be a percentage value in the range of 0%-100%, such as 75%. In an embodiment where the second obstacle confidence is 75%, the second candidate position information corresponding to the obstacle confidence greater than 75% may be determined as the second roadside position information.

[0070] After the roadside device determines the first roadside location information and the second roadside location information, the first roadside location information and the second roadside location information may be sent to the vehicle.

[0071] In the embodiment of the present application, by setting the first detection area and the second detection area, and setting different obstacle confidences for different detection areas, the obstacle output threshold can be appropriately lowered for the area of ​​interest or the area with a greater impact on driving to reduce missed detection, and the obstacle output threshold can be appropriately increased for the area of ​​non-interest or the area with a smaller impact on driving to reduce false detection.

[0072] S204: Match each vehicle-side position information in the plurality of vehicle-side position information with each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations.

[0073] Among them, each matching combination may include vehicle-side matching position information and roadside matching position information, the vehicle-side matching position information is one of multiple vehicle-side position information, and the roadside matching position information is the roadside position information among multiple roadside position information that matches the vehicle-side matching position information.

[0074] In a specific implementation, position information matching is performed on each vehicle-end position information in multiple vehicle-end position information and each roadside position information in multiple roadside position information. The position point data corresponding to each vehicle-end position information in multiple vehicle-end position information and the position point data corresponding to each roadside position information in multiple roadside position information are compared to obtain mutually matching vehicle-end matching position information and roadside matching position information.

[0075] In another specific implementation, the vehicle-side location information may be matched one-to-one with the roadside location information by running a matching algorithm, such as the Hungarian algorithm.

[0076] The following is based on Figure 4 Step S204 is further described. Figure 4 As illustrated, an exemplary process may include:

[0077] S401: Determine current vehicle-end position information among multiple vehicle-end position information.

[0078] Specifically, the vehicle-side position information that needs to be matched can be first determined from multiple vehicle-side position information as the current vehicle-side position information. The current vehicle-side position information can be determined in order of the size of the coordinates, wherein the vehicle-side position information that has been determined as the current vehicle-side position information can be excluded from the multiple alternative vehicle-side position information.

[0079] S402: Match the current vehicle-side position information with each piece of roadside position information in a plurality of roadside position information to obtain a current matching combination that has been successfully matched.

[0080] Specifically, the current vehicle-end position information can be matched with each roadside position information in multiple roadside position information respectively, so that when there is roadside position information in the multiple roadside position information that can match the current vehicle-end position information, it is determined that the current vehicle-end position information is matched successfully, and the current vehicle-end position information is combined with the roadside position information in the multiple roadside position information that matches the current vehicle-end position information to obtain a current matching combination that is successfully matched.

[0081] S403: Traverse each vehicle-end position information in the multiple vehicle-end position information to obtain multiple groups of successful matching combinations.

[0082] Specifically, after the current vehicle-end position information is matched, the process can return to step S401 and traverse each of the multiple vehicle-end position information until each of the multiple vehicle-end position information is determined as the current vehicle-end position information once, thereby determining multiple matching combinations.

[0083] In a specific implementation, after traversing each of the multiple vehicle-side position information and obtaining multiple groups of matching combinations that are successfully matched, the objects corresponding to the unmatched vehicle-side position information and road-side position information can be determined as the remaining target objects, wherein the objects corresponding to the vehicle-side position information and road-side position information can be determined by the vehicle-side laser radar and the road-side equipment when determining the vehicle-side position information and the road-side position information, respectively. In addition, the vehicle-side position information or the road-side position information corresponding to the remaining target objects can be determined as the remaining target positions.

[0084] In a specific implementation, if the proportion of the remaining target objects to all objects is greater than a preset proportion threshold, for example, 10%, where all objects are the sum of the remaining target objects and the objects corresponding to the matching combination, then the process returns to step S203 to re-execute the matching process, and a lower confidence threshold is used based on the original first obstacle confidence threshold and the original second obstacle confidence threshold to execute step S203, so as to avoid missed detection in complex environments or under poor detection conditions and ensure the comprehensiveness of detection. For example, the confidence threshold can be halved.

[0085] The following is based on Figure 4 Step S402 is further described. Specifically, Figure 5 As illustrated, an exemplary process may include:

[0086] S501: Determine matching distance information between current vehicle position information and each piece of roadside position information in a plurality of roadside position information to obtain a plurality of matching distance information.

[0087] Specifically, the matching distance information between the current vehicle position information and each roadside position information in the plurality of roadside position information can be determined based on the distance calculation formula between the spatial coordinate points. The matching distance information and each roadside position information are in a one-to-one correspondence.

[0088] S502: Determine target matching distance information from multiple matching distance information.

[0089] Specifically, the target matching distance information may be the minimum matching distance information among multiple matching distance information.

[0090] S503: When the target matching distance information is less than a preset distance threshold, it is determined that the current vehicle-side position information matches the roadside position information corresponding to the target matching distance information successfully.

[0091] Specifically, when the target matching distance information is less than the preset distance threshold, it means that the current vehicle position information and the roadside position information corresponding to the target matching distance information are determined to be the same object, and it is determined that the current vehicle position information and the roadside position information corresponding to the target matching distance information are matched successfully. If the target matching distance information is greater than or equal to the preset distance threshold, it is determined that the current vehicle position information and the roadside position information corresponding to the target matching distance information are matched unsuccessfully.

[0092] S504: Determine the current vehicle-side position information and roadside position information that have been successfully matched as first matching position information and second matching position information, respectively.

[0093] Specifically, the current vehicle-side position information and roadside position information that are successfully matched may be marked respectively, for example, as first matching position information and second matching position information.

[0094] S505: Determine a matching combination according to the first matching position information and the second matching position information.

[0095] Specifically, the first matching position information and the second matching position information that match each other can be determined as a matching combination. By using each of the multiple vehicle-side position information as the current position information to be matched, each of the multiple vehicle-side position information can be traversed to avoid missing the target object, thereby improving the vehicle's perception accuracy of surrounding objects and avoiding the missed target object from affecting the normal driving of the vehicle.

[0096] S205: Determine multiple target objects corresponding to the multiple matching combinations.

[0097] The target object corresponding to each matching combination is the object corresponding to the first matching position information in each matching combination. Specifically, each matching combination in the plurality of matching combinations corresponds to a target object. The type and object identification information of the object corresponding to the target object can be determined by the vehicle body controller based on the vehicle-mounted laser radar.

[0098] S206: Input the first matching position information and the second matching position information in the multiple matching combinations into a position prediction model, perform position prediction processing, and obtain multiple target position information corresponding to the multiple target objects.

[0099] Specifically, the position prediction model can be a model based on the Kalman filter algorithm. When the first matching position information and the second matching position information are input into the position prediction model, the first matching position information and the second matching position information are respectively used as two measurement values. The position prediction model is based on the input first matching position information and the second matching position information, and finally outputs an estimated value, and uses the estimated value as the target position of the target object corresponding to the first matching position information. It can be understood that each of the multiple target objects determined above will correspond to a target position, and then multiple target position information corresponding to the multiple target objects is determined. In an embodiment of the present application, by matching the determined position information of the vehicle-mounted camera and the roadside perception device, and predicting the target position of the target object based on the matched position information, it is possible to avoid position prediction based on a single position information, and thus improve the positioning accuracy of the target position.

[0100] Furthermore, the above-mentioned comparison of the position point data corresponding to each vehicle-end position information in the multiple vehicle-end position information with the position point data corresponding to each roadside position information in the multiple roadside position information to obtain mutually matching first matching position information and second matching position information is only a preferred implementation scheme. In a specific implementation scheme, the position point data corresponding to each roadside position information in the multiple roadside position information can also be compared with the position point data corresponding to each vehicle-end position information in the multiple vehicle-end position information to obtain mutually matching first matching positions and second matching positions. In this process, the type of object corresponding to the target object and the object identification information can be determined by the roadside equipment when determining the roadside position information, and after the second matching position information is determined in the matching process, the object corresponding to the second matching position information can be determined based on the identification information corresponding to the second matching position information, and the object is used as the target object of the matching combination. The rest of the specific matching process and matching principle can refer to the steps and principles of comparing the position point data corresponding to each vehicle-end position information in the multiple vehicle-end position information with the position point data corresponding to each roadside position information in the multiple roadside position information, which will not be repeated here.

[0101] In a specific implementation, after the first matching position information and the second matching position information in multiple matching combinations are input into a position prediction model and position prediction processing is performed to obtain multiple target position information corresponding to multiple target objects, the vehicle-side position information or roadside position information of the remaining target objects can also be determined as the remaining target positions, and the remaining target positions and the above-mentioned multiple target position information can be incorporated into the target position information set.

[0102] Accordingly, the present application provides an object positioning device. Figure 6 Schematic diagram of the structure of an object positioning device provided in an embodiment of the present application. Figure 6 As shown, the object positioning device 600 may include:

[0103] The area determination module 601 is used to determine a first detection area and a second detection area within a preset range around the vehicle; wherein the first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area in the preset range except the first detection area.

[0104] A first position module 602, configured to detect a first detection area based on a target detection algorithm and a first obstacle confidence threshold, and obtain a plurality of vehicle-side position information;

[0105] The second position module 603 is used to detect the second detection area based on the target detection algorithm and the second obstacle confidence threshold to obtain multiple roadside position information;

[0106] The vehicle end position module 604 is used to determine multiple vehicle end position information based on multiple vehicle end position information and multiple roadside position information;

[0107] The acquisition module 605 is used to acquire multiple roadside location information sent by the roadside equipment;

[0108] A matching module 606 is used to match each vehicle-side position information in the plurality of vehicle-side position information with each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations;

[0109] A target determination module 607, used to determine multiple target objects corresponding to multiple matching combinations;

[0110] The position prediction module 608 is used to input the first matching position information and the second matching position information in the multiple matching combinations into the position prediction model, perform position prediction processing, and obtain multiple target position information corresponding to the multiple target objects.

[0111] Specifically, the first position module 602 is also used to: based on the target detection algorithm, detect the first detection area to obtain multiple first candidate position information; determine the obstacle confidence of the first candidate position information; if the obstacle confidence is greater than the first obstacle confidence threshold, determine the first candidate position information as the vehicle-side position information; the second position module is also used to: based on the target detection algorithm, detect the second detection area to obtain multiple second candidate position information; determine the obstacle confidence of the second candidate position information; if the obstacle confidence is greater than the second obstacle confidence threshold, determine the second candidate position information as the roadside position information.

[0112] Specifically, the first obstacle confidence threshold is smaller than the second obstacle confidence threshold.

[0113] Specifically, the matching module 606 is also used to: determine the current vehicle-end position information among multiple vehicle-end position information; perform position information matching between the current vehicle-end position information and each road-side position information among multiple road-side position information to obtain a current matching combination that has successfully matched; and traverse each vehicle-end position information among multiple vehicle-end position information to obtain multiple groups of matching combinations that have successfully matched.

[0114] Specifically, the matching module 606 is also used to: determine the matching distance information between the current vehicle-side position information and each roadside position information in the multiple roadside position information to obtain multiple matching distance information; determine the target matching distance information in the multiple matching distance information, the target matching distance information is the minimum matching distance information in the multiple matching distance information; when the target matching distance information is less than a preset distance threshold, determine that the current vehicle-side position information and the roadside position information corresponding to the target matching distance information are successfully matched; determine the successfully matched current vehicle-side position information and roadside position information as the first matching position information and the second matching position information, respectively; determine the matching combination based on the first matching position information and the second matching position information.

[0115] Specifically, the position prediction model is a model based on the Kalman filter algorithm.

[0116] Specifically, the roadside equipment includes any one of a laser radar, a millimeter wave radar and a radio frequency identification sensor.

[0117] The device embodiments and method embodiments of the present application may be based on the same concept.

[0118] Correspondingly, an embodiment of the present disclosure also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the above-mentioned object positioning method.

[0119] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 7 is a hardware structure block diagram of a server of the object positioning method provided in the embodiment of the present application. Figure 7As shown, the server 700 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 710 (the central processing unit 710 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 730 for storing data, and one or more storage media 720 (such as one or more mass storage devices) for storing application programs 723 or data 722. Among them, the memory 730 and the storage medium 720 can be short-term storage or permanent storage. The program stored in the storage medium 720 may include one or more modules, each of which may include a series of instruction operations on the server. Furthermore, the central processing unit 710 can be configured to communicate with the storage medium 720 and execute a series of instruction operations in the storage medium 720 on the server 700. The server 700 may also include one or more power supplies 770, one or more wired or wireless network interfaces 750, one or more input and output interfaces 740, and / or one or more operating systems 721, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0120] The input / output interface 740 may be used to receive or send data via a network. The specific example of the network may include a wireless network provided by a communication provider of the server 700. In one example, the input / output interface 740 includes a network adapter (Network Interface Controller, NIC), which may be connected to other network devices via a base station so as to communicate with the Internet. In one example, the input / output interface 740 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0121] It can be understood by those skilled in the art that Figure 7 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 7 More or fewer components as shown, or with Figure 7 Different configurations shown.

[0122] The present application implements a storage medium, which can be set in a server to store at least one instruction, at least one program, code set or instruction set related to the object positioning method in the method embodiment, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by the processor to implement the above-mentioned object positioning method.

[0123] Specifically, in this embodiment, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a mobile hard disk, a magnetic disk, or an optical disk.

[0124] In the present invention, unless otherwise clearly specified and limited, the terms "connected", "connection" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be a connection between two elements or an interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0125] It should be noted that the order of the above embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The above description describes specific embodiments, and other embodiments are also within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in the order of different embodiments and can achieve the expected results. In addition, the processes depicted in the drawings do not necessarily require a specific order or connection order to achieve the desired results. In some embodiments, multi-task parallel processing is also possible or may be advantageous.

[0126] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device / system, since they are based on similarities to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0127] The above are preferred embodiments of the present invention. It should be noted that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An object positioning method, characterized in that: For a vehicle, the method comprises: Determine a first detection area and a second detection area within a preset range around the vehicle; wherein the first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area within the preset range except the first detection area; Determine the location information of multiple vehicles based on the target detection algorithm; By means of roadside equipment, the first detection area is detected based on a target detection algorithm and a first obstacle confidence threshold, and the second detection area is detected based on the target detection algorithm and a second obstacle confidence threshold, so as to obtain a plurality of roadside position information; the first obstacle confidence threshold is less than the second obstacle confidence threshold; Performing position information matching on each vehicle-side position information in the plurality of vehicle-side position information and each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations; Determine a plurality of target objects corresponding to the plurality of matching combinations; The current vehicle-side position information and roadside position information that have been successfully matched are respectively determined as the first matching position information and the second matching position information; the first matching position information and the second matching position information in multiple matching combinations are input into the position prediction model, and position prediction processing is performed to obtain multiple target position information corresponding to the multiple target objects.

2. The object positioning method according to claim 1, characterized in that: The roadside location information includes first roadside location information and second roadside location information. The first detection area is detected by the roadside equipment based on the target detection algorithm and the first obstacle confidence threshold, and the second detection area is detected based on the target detection algorithm and the second obstacle confidence threshold to obtain multiple roadside position information, including: Using a roadside device, based on a target detection algorithm, the first detection area is detected to obtain a plurality of first candidate position information; Determining an obstacle confidence level of the first candidate position information; If the obstacle confidence is greater than the first obstacle confidence threshold, determining the first candidate location information as the first roadside location information; Using the roadside equipment, based on the target detection algorithm, the second detection area is detected to obtain a plurality of second candidate position information; Determining an obstacle confidence level of the second candidate position information; If the obstacle confidence is greater than the second obstacle confidence threshold, the second candidate location information is determined to be the second roadside location information.

3. The object positioning method according to claim 1, characterized in that: The performing position information matching of each vehicle-side position information in the plurality of vehicle-side position information with each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations comprises: Determine current vehicle end position information among the multiple vehicle end position information; Performing position information matching of the current vehicle-side position information with each of the plurality of roadside position information to obtain a current matching combination with successful matching; Each vehicle-end position information in the multiple vehicle-end position information is traversed to obtain multiple groups of matching combinations that are successfully matched.

4. The object positioning method according to claim 3, characterized in that: The step of performing position information matching between the current vehicle-side position information and each roadside position information in the plurality of roadside position information to obtain a current matching combination that successfully matches includes: Determine matching distance information between the current vehicle-side position information and each roadside position information in the plurality of roadside position information to obtain a plurality of matching distance information; Determine target matching distance information among the multiple matching distance information, wherein the target matching distance information is the minimum matching distance information among the multiple matching distance information; When the target matching distance information is less than a preset distance threshold, determining that the current vehicle-side position information matches the roadside position information corresponding to the target matching distance information successfully; Determine the successfully matched current vehicle-side position information and roadside position information as first matching position information and second matching position information respectively; A matching combination is determined according to the first matching position information and the second matching position information.

5. The object positioning method according to claim 1, characterized in that: The position prediction model is a model based on the Kalman filter algorithm.

6. The object positioning method according to claim 1, characterized in that: The roadside equipment includes any one of a laser radar, a millimeter wave radar and a radio frequency identification sensor.

7. An object positioning device, characterized in that: The device comprises: An area determination module is used to determine a first detection area and a second detection area within a preset range around the vehicle; wherein the first detection area includes an area where the traffic density is greater than a preset threshold, and the second detection area is an area within the preset range except the first detection area; A first position module, configured to detect a first detection area based on a target detection algorithm and a first obstacle confidence threshold, and obtain a plurality of vehicle-side position information; A second position module, configured to detect a second detection area based on a target detection algorithm and a second obstacle confidence threshold to obtain a plurality of roadside position information; the first obstacle confidence threshold is less than the second obstacle confidence threshold; A vehicle-side position module, used to determine a plurality of vehicle-side position information based on a plurality of vehicle-side position information and a plurality of roadside position information; An acquisition module, used to acquire multiple roadside location information sent by roadside equipment; A matching module, used for matching each vehicle-side position information in the plurality of vehicle-side position information with each roadside position information in the plurality of roadside position information to obtain a plurality of matching combinations; A target determination module, used to determine multiple target objects corresponding to multiple matching combinations; The position prediction module is used to determine the current vehicle-side position information and roadside position information that have been successfully matched as the first matching position information and the second matching position information respectively; the first matching position information and the second matching position information in multiple matching combinations are input into the position prediction model, and position prediction processing is performed to obtain multiple target position information corresponding to multiple target objects.

8. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the object positioning method described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the object positioning method described in any one of claims 1-6.

Images