Vehicle control method and device, electronic equipment and storage medium

Abstract

The present disclosure provides a vehicle control method and device, electronic equipment and storage medium, the method comprising: obtaining first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in a historical frame before a current frame, first object recognition results corresponding to the first point cloud data, and second point cloud data collected in the current frame; determining second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information and the first object recognition results; and controlling a driving state of the target vehicle based on the second blind area information.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, in particular, to a vehicle control method and device, electronic equipment and storage medium. BACKGROUND

[0002] With the improvement of living standards, cars have become an indispensable part of human society, and high-precision vehicle positioning and navigation are important parts of vehicle intelligence and automation, and are the basis for intelligent vehicle perception, control, path planning and other modules.

[0003] Generally, the positioning of obstacles can be based on the radar loaded on the vehicle, but due to the influence of obstacle shielding and the vertical angle resolution of the radar itself, there will be a radar blind area in the point cloud data collected by the radar, and the obstacles in the blind area cannot be detected, which brings risks to the safe driving of the vehicle. SUMMARY

[0004] The embodiments of the present disclosure at least provide a vehicle control method, device, electronic equipment and storage medium.

[0005] In a first aspect, the embodiments of the present disclosure provide a vehicle control method, comprising:

[0006] obtaining first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in a historical frame before a current frame, first object recognition results corresponding to the first point cloud data, and second point cloud data collected in the current frame;

[0007] determining second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information and the first object recognition results;

[0008] controlling a driving state of the target vehicle based on the second blind area information.

[0009] In this aspect, the second blind area information corresponding to the second point cloud data can be determined through the second point cloud data, the first blind area information and the first object recognition results, the blind area of the second point cloud data is monitored, and thus the driving state of the target vehicle can be controlled based on the second blind area information, and the probability of dangerous accidents is reduced.

[0010] In a possible implementation, the first blind area information includes position information of a first blind area in the first point cloud data, and the second blind area information includes position information of a second blind area in the second point cloud data.

[0011] The determination of the second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information and the first object recognition results comprises:

[0012] determine a second object recognition result corresponding to the second point cloud data and position information of the second blind area.

[0013] In a possible implementation, the first blind area information further includes object information in the first blind area; and the second blind area information further includes target object information in the second blind area.

[0014] The determining of the second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information, and the first object recognition result includes:

[0015] determining initial object information in the second blind area based on the position information of the second blind area, the position information of the first blind area, and the object information in the first blind area;

[0016] updating the initial object information based on the object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area to obtain target object information in the second blind area.

[0017] In this aspect, the position information of the second blind area, the position information of the first blind area, and the object information in the first blind area are used to enable the second blind area to inherit the object information in the first blind area of the historical frame to obtain initial object information in the second blind area. Then, the initial object information is updated based on the target object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area to obtain target object information in the second blind area, so as to determine the obstacle in the blind area and control the vehicle driving state according to the target object information, thereby effectively improving the safety of vehicle driving.

[0018] In a possible implementation, the position information of the second blind area corresponding to the second point cloud data is determined by the following steps:

[0019] determining information of an obstacle in a set range from the target vehicle based on the second point cloud data;

[0020] determining the position information of the blind area corresponding to the second point cloud data based on the beam information emitted by the radar and the determined information of the obstacle.

[0021] In this implementation, the information of the obstacle in the set range from the target vehicle in the second point cloud data and the beam information emitted by the radar are used to more accurately determine the position information of the blind area corresponding to the second point cloud data, so as to accurately determine the radar blind area.

[0022] In a possible implementation, the first blind area includes at least one first sub-blind area, and the second blind area includes at least one second sub-blind area;

[0023] The initial object information in the second blind area is determined based on the position information of the second blind area, the position information of the first blind area, and the object information in the first blind area.

[0024] The association between each second sub-blind area and each first sub-blind area is determined based on the position information of the first blind area and the position information of the second blind area.

[0025] The initial object information in each second sub-blind area is determined based on the determined association and the target object information in each first sub-blind area.

[0026] In this implementation, the association between each second sub-blind area and each first sub-blind area can be determined more accurately based on the position information of the first blind area and the position information of the second blind area, and the initial object information in each second sub-blind area can be determined based on the determined association, so that the object information in each blind area can be associated in time sequence, and the object information in each blind area can be more accurately transmitted in time, that is, the initial object information can be more accurately determined.

[0027] In a possible implementation, the association between each second sub-blind area and each first sub-blind area is determined based on the position information of the first blind area and the position information of the second blind area.

[0028] An overlapping area between each second sub-blind area and each first sub-blind area is determined based on the position information of the first blind area and the position information of the second blind area.

[0029] For any second sub-blind area, the association between the second sub-blind area and each first sub-blind area is determined based on the area of the overlapping area between the second sub-blind area and each first sub-blind area. In this implementation, whether each second sub-blind area overlaps with each first sub-blind area is determined, and the association between each second sub-blind area and each first sub-blind area is determined based on the area of the overlapping area, so that the determined association between each sub-blind area is more accurate.

[0030] In a possible implementation, the initial object information in each second sub-blind area is determined based on the determined association and the target object information in each first sub-blind area.

[0031] For any second sub-blind area, based on object information in each first sub-blind area and an association relationship between the second sub-blind area and each first sub-blind area, initial object information in the second sub-blind area is determined.

[0032] In this embodiment, based on object information in the first sub-blind area and the determined association relationship, target object information in the first sub-blind area that needs to be inherited by the second sub-blind area can be determined, and the accuracy of the initial object information is improved.

[0033] In a possible implementation, the target object information includes observable object information and unobservable object information; and the initial object information includes observable object information and unobservable object information.

[0034] The updating of the initial object information based on the object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area to obtain target object information in the second blind area includes:

[0035] The observable object information in the initial object information is updated based on the first object recognition result, the second object recognition result, and the position information of the second blind area.

[0036] The unobservable object information in the initial object information is updated based on the first object recognition result, the second object recognition result, the observable object information in the initial object information, and the unobservable object information in the object information in the first blind area.

[0037] The updated initial object information is determined as the target object information in the second blind area.

[0038] In this embodiment, the observable object information in the inherited initial object information is updated based on the first object recognition result, the second object recognition result, and the position information of the second blind area, and the unobservable object information is updated by determining the change of unobservable objects between the current frame and the historical frame based on the first object recognition result, the second object recognition result, and the observable object information in the initial object information, so that the obstacles in the blind area can be accurately determined.

[0039] In a possible implementation, the control of the driving state of the target vehicle based on the second blind area information includes:

[0040] The type and quantity of unobservable objects in each second sub-blind area are determined based on the unobservable object information of the target object information in the second blind area.

[0041] determine a danger level corresponding to each of the second sub-blind areas based on a type and a quantity of the unobservable objects in each of the second sub-blind areas;

[0042] control a driving state of the target vehicle based on the danger level corresponding to each of the second sub-blind areas.

[0043] In this embodiment, the type and the quantity of the unobservable objects in the second sub-blind area affect the safety and the driving of the target vehicle, and therefore, the danger level corresponding to each of the second sub-blind areas can be accurately determined based on the type and the quantity of the unobservable objects in the second sub-blind area, and then the driving state of the target vehicle is controlled based on the danger level, which can effectively improve the driving safety of the target vehicle.

[0044] In a second aspect, the embodiments of the present disclosure further provide a vehicle control device, comprising:

[0045] an acquisition module configured to acquire first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in historical frames before a current frame, a first object recognition result corresponding to the first point cloud data, and second point cloud data collected in the current frame;

[0046] a determination module configured to determine second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information, and the first object recognition result;

[0047] a control module configured to control a driving state of the target vehicle based on the second blind area information.

[0048] In a third aspect, the embodiments of the present disclosure further provide an electronic device, comprising a processor, a memory, and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the first aspect or any possible implementation manner of the first aspect.

[0049] In a fourth aspect, the embodiments of the present disclosure further provide a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by a processor to perform the steps of the first aspect or any possible implementation manner of the first aspect.

[0050] For the effects of the vehicle control device, the electronic device, and the computer readable storage medium, refer to the description of the vehicle control method, and details are not repeated here.

[0051] In order to make the above objectives, characteristics and advantages of the present disclosure more apparent and easy to understand, the following preferred embodiments are specifically described below with reference to the accompanying drawings. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings used in the embodiments will be briefly described below. These drawings are incorporated in and constitute a part of this specification. They illustrate embodiments conforming to this disclosure and, together with the specification, serve to explain the technical solutions of this disclosure. It should be understood that the following drawings only show some embodiments of this disclosure and should not be considered as limiting the scope. Those skilled in the art can obtain other related drawings based on these drawings without creative effort.

[0053] Figure 1 A flowchart of a vehicle control method provided by an embodiment of this disclosure is shown;

[0054] Figure 2 A schematic diagram of the blind zone provided in the embodiments of this disclosure is shown;

[0055] Figure 3 A schematic diagram illustrating the state of the target object provided in an embodiment of this disclosure is shown;

[0056] Figure 4 A schematic diagram of a vehicle control device provided in an embodiment of this disclosure is shown;

[0057] Figure 5 A schematic diagram of an electronic device provided in an embodiment of the present disclosure is shown. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. The components of the embodiments of this disclosure described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed disclosure, but merely represents selected embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0059] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0060] The term "and / or", merely describes an associated relationship, which means that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, and C, which means including any one or more elements selected from the set consisting of A, B, and C.

[0061] In order to solve the problem that the prior art cannot detect obstacles in the radar blind area, which brings risks to the safe driving of the vehicle, the present disclosure provides a vehicle control method, device, electronic equipment and storage medium. The present disclosure first acquires first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in a historical frame before a current frame, a first object recognition result corresponding to the first point cloud data, and second point cloud data collected in the current frame. Then, based on the second point cloud data, the first blind area information, and the first object recognition result, the second blind area information corresponding to the second point cloud data is determined. Finally, based on the second blind area information, the driving state of the target vehicle is controlled. The present disclosure can determine the second blind area information corresponding to the second point cloud data by using the second point cloud data, the first blind area information, and the first object recognition result, monitor the blind area of the second point cloud data, and thus control the driving state of the target vehicle based on the second blind area information, thereby reducing the probability of dangerous accidents.

[0062] The vehicle control method, device, electronic equipment and storage medium disclosed by the present disclosure will be described below through specific embodiments.

[0063] As shown in Figure 1 The present disclosure discloses a vehicle control method, which can be applied to electronic equipment with computing capability, such as servers, vehicle-mounted computers, etc. Specifically, the vehicle control method can include the following steps:

[0064] S110, acquiring first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in a historical frame before a current frame, a first object recognition result corresponding to the first point cloud data, and second point cloud data collected in the current frame.

[0065] The target vehicle can be an autonomous vehicle, and the radar can be a laser radar. The laser radar has high resolution, high ranging accuracy, and good detection performance, and is the most important sensor on an autonomous vehicle. The radar can continuously collect point cloud data. The current frame and the historical frame can be two consecutive frames or two non-consecutive frames. In the case of two non-consecutive frames, the time interval between the two frames needs to be within a preset range, so as to prevent the difference between the two frames of point cloud data from being too large and improve the timeliness of the subsequent inheritance of the target object information of the historical frame.

[0066] Due to the influence of an obstacle and the limitation of the vertical angle resolution of the laser radar, there can be a blind area in the point cloud data collected by the laser radar. The blind area can have multiple sub-blind areas, and the top view of each sub-blind area can be a polygon with an irregular shape. Referring to FIG. 1, which is a schematic diagram of a blind area provided by an embodiment of the present disclosure. Figure 2 Figure 2 In the example shown in FIG. 1, the target vehicle detects two obstacles by the laser radar, and forms two sub-blind areas. The position information of the blind area can include the coordinates of the blind area in the point cloud data and the height information of the blind area.

[0067] The blind area information can include the position information of the blind area and the object information in the blind area. The object in the blind area can include an observable object and an unobservable object. The unobservable object is an object that is completely blocked by the obstacle and cannot be detected by the radar. The observable object is an object that is detected by the radar in the blind area, such as the obstacle itself that causes the blind area, or due to errors, the blind area and the obstacle profile are not completely consistent, resulting in a gap, so that some objects behind the obstacle can be detected.

[0068] The target object information in the blind area can be determined by the point cloud data. For example, a trained object recognition model can be used to recognize the point cloud data to obtain the target object information in the point cloud data.

[0069] The target object information can include the position, shape, orientation angle, speed, and category of the target object.

[0070] S120, based on the second point cloud data, the first blind area information, and the first object recognition result, determining second blind area information corresponding to the second point cloud data.

[0071] The first blind area information can include the position information of the first blind area in the first point cloud data, and the second blind area information can include the position information of the second blind area in the second point cloud data. The position information of the second blind area and the second object recognition result can be determined by an object recognition algorithm.

[0072] ​In a possible embodiment, the position information of the second blind area corresponding to the second point cloud data can be determined through the following steps:

[0073] Based on the second point cloud data, information of an obstacle within a set range from the target vehicle in the second point cloud data is determined.

[0074] Based on the line beam information emitted by the radar and the determined information of the obstacle, position information of a blind area corresponding to the second point cloud data is determined.

[0075] The radar can obtain position information of each point constituting an obstacle contour in a set coordinate system, and in this way, contour information of an obstacle within a set range from the target vehicle can be obtained based on the point cloud data.

[0076] The line beam information can include the number of line beams and the height from the ground of the line beams emitted by the radar at each rotation angle, and can be represented by a line height map established in advance. For example, a grid map of a bird's eye view of a ground surface region within a set range from the target vehicle can be constructed in advance, and then a line height map corresponding to the grid map can be generated based on the line beam information emitted by the radar, where the line height map includes three dimensions, the first two dimensions represent the row position and column position of each grid in the line height map, and the third dimension represents the number of line beams included in each grid. In addition, the height of each line beam included in the grid in the grid is also recorded.

[0077] The number of line beams corresponding to each grid refers to the number of line beams emitted by the radar device and entering the grid, without considering the existence of an obstacle in the grid, but only according to the installation position, installation angle and arrangement angle of the radar transmitter.

[0078] In this embodiment, the position information of the second blind area corresponding to the second point cloud data can be determined based on the information of the obstacle within a set range from the target vehicle in the second point cloud data and the line beam information emitted by the radar, so that the determination of the radar blind area is realized.

[0079] The step of determining the position information of the first blind area of the first point cloud data can be the same as the step of determining the position information of the second blind area.

[0080] The first blind area information can further include object information in the first blind area, and the second blind area information can further include target object information in the second blind area.

[0081] The second blind area information corresponding to the second point cloud data can be determined based on the second point cloud data, the first blind area information, and the first object recognition result.

[0082] The initial object information in the second blind area is determined based on the position information of the second blind area, the position information of the first blind area, and the object information in the first blind area.

[0083] The target object information in the second blind area is obtained by updating the initial object information based on the object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area.

[0084] The initial object information can be object information inherited from a historical frame. In a possible embodiment, the first blind area can include at least one first sub-blind area, and the second blind area can include at least one second sub-blind area. The initial object information can be determined by the following steps:

[0085] The association between each second sub-blind area and each first sub-blind area is determined based on the position information of the first blind area and the position information of the second blind area.

[0086] The initial object information in each second sub-blind area is determined based on the determined association and the target object information in each first sub-blind area.

[0087] The association can include appearance (the sub-blind area does not exist in a historical frame, but exists in a current frame), disappearance (the sub-blind area exists in a historical frame, but does not exist in a current frame), one-to-one (one sub-blind area in a historical frame corresponds to one sub-blind area in a current frame, and no other associated sub-blind area exists), splitting (one sub-blind area in a historical frame is associated with multiple sub-blind areas in a current frame), fusion (multiple sub-blind areas in a historical frame are associated with only one sub-blind area in a current frame), splitting and fusion (splitting and fusion occur simultaneously), according to the determined association, it can be determined which target object information in the first sub-blind area should be inherited by the second sub-blind area.

[0088] In this embodiment, the association between each second sub-blind area and each first sub-blind area is determined based on the position information of the first blind area and the position information of the second blind area. Then, the initial object information in each second sub-blind area is determined according to the determined association. In this way, the object information in each blind area can be associated in time sequence, so that the object information in the blind area can be transmitted in time.

[0089] In some possible embodiments, the association between each second sub-blind area and each first sub-blind area can be determined by the following steps:

[0090] determine whether each of the second sub-blind areas overlaps with each of the first sub-blind areas based on the position information of the first blind area and the position information of the second blind area;

[0091] For any second sub-blind area, determine the correlation between the second sub-blind area and each first sub-blind area based on the area of the overlapping region between the second sub-blind area and each first sub-blind area. For example, if there are m sub-blind areas in the historical frame and n sub-blind areas in the current frame, first convert the position information of the first blind area in the historical frame to the coordinate system of the position information of the second blind area in the current frame, then determine the correlation matrix of m*n. If there is an overlap between a first sub-blind area and a second sub-blind area in terms of position, i.e. the area of the overlapping region is greater than 0, set the corresponding position in the correlation matrix to 1, otherwise set it to 0. By solving the correlation matrix, the correlation between each second sub-blind area and each first sub-blind area can be obtained, and the successor of each first sub-blind area (the second sub-blind area overlapping with the first sub-blind area) and the predecessor of each second sub-blind area (the first sub-blind area overlapping with the second sub-blind area) can be obtained.

[0092] For example, the correlation between different sub-blind areas can be represented by assigning identification information to the sub-blind areas. For example, if a second sub-blind area has no predecessor, it can be assigned new identification information; if a second sub-blind area has only one predecessor, it can inherit the identification information of the predecessor; if a second sub-blind area has multiple predecessors (fusion relationship), it can inherit the latest identification information of the predecessors and the target object information of all predecessors; if a first sub-blind area has multiple successors (split relationship), it can assign new identification information to all successors of the first sub-blind area; if fusion and split relationships exist simultaneously, new identification information can be assigned to the second sub-blind area of the successor.

[0093] After determining the correlation, the initial object information in the second sub-blind area can be determined based on the object information in each of the first sub-blind areas and the correlation between the second sub-blind area and each first sub-blind area.

[0094] For example, the correlation type (such as fusion, split, etc.) between the second sub-blind area and each first sub-blind area can be used to determine whether the first sub-blind area is the predecessor of the second sub-blind area, and then the initial object information in the second sub-blind area can be determined based on the target object information of each predecessor.

[0095] In the case of inheriting the identity of the predecessor, the target object information of the predecessor of the second sub-blind area is used as the initial object information of the second sub-blind area, and in the case of fusion relationship, the target object information of all predecessors of the second sub-blind area can be fused and then inherited by the second sub-blind area.

[0096] The embodiment determines the association relationship between each second sub-blind area and each first sub-blind area by judging whether there is an overlapping area between each second sub-blind area and each first sub-blind area, so that the determined association relationship between each sub-blind area is more accurate.

[0097] After inheriting the target object information of the historical frame, since there is a certain time difference between the historical frame and the current frame, the objects detected by the radar can move, some objects can enter the blind area, and some objects can leave the blind area, so it is necessary to update the determined initial object information. In some possible embodiments, the initial object information can be updated by the following steps:

[0098] Based on the first object recognition result, the second object recognition result, and the position information of the second blind area, the observable object information in the initial object information is updated;

[0099] Based on the first object recognition result, the second object recognition result, the observable object information in the initial object information, and the unobservable object information in the target object information in the first blind area, the unobservable object information in the initial object information is updated;

[0100] The updated initial object information is determined as the target object information in the second blind area.

[0101] This step can be implemented by tracking each target object between different frames. The tracking mode can include explicit tracking and implicit tracking. The explicit tracking can use the object recognition result of the radar as the tracking result and record it in the observable object information of the corresponding blind area. The implicit tracking can use the latest object recognition result of the radar as the tracking result and record it in the unobservable object information of the corresponding blind area.

[0102] Specifically, the first object recognition result and the second object recognition result can be used to track the objects in the point cloud data, determine the same object in the first point cloud data and the second point cloud data, and the same object recognized can be represented by the same object identifier. When the appearance position of a target object is determined to be in a second sub-blind area, the target object can be added to the observable object information of the second sub-blind area to update the observable object information in the initial object information.

[0103] Further, the unobservable object information in the initial object information can be updated. For example, if a target object is observed in the first point cloud data and the second point cloud data, and exists in the observable object information of a second sub-blind area and the unobservable object information of a first sub-blind area, it indicates that the target object enters the blind area, disappears, and is detected in the blind area again. For this kind of target object, a display tracking method can be used, and the second object recognition result of the target object is used as tracking information. If a target object is observed in the first point cloud data and the second point cloud data, and exists in the observable object information of a second sub-blind area, and does not exist in the unobservable object information of a first sub-blind area, it indicates that the target object just enters the blind area, and a new tracking target can be established for the second sub-blind area where the target object enters, and the unobservable object information is not updated temporarily. If a target object is observed in the first point cloud data and the second point cloud data, and exists in the unobservable object information of a first sub-blind area, and does not exist in the observable object information of any second sub-blind area, it can be considered that the target object leaves the blind area, and the target object is removed from the unobservable object information of the second sub-blind area corresponding to the target object. If a target object is observed in the first point cloud data and the second point cloud data, and does not exist in any observable object information and unobservable object information, it indicates that the target object is outside the blind area, and the target object is not processed. If a target object is not observed in the first point cloud data and the second point cloud data, and does not exist in any observable object information, but exists in one or more unobservable object information, it indicates that the target object is still in the blind area, and an implicit tracking method is used to record the blind area where the target object is located.

[0104] As shown in the following table, a judgment method of the tracking method used for the target object tracking is shown. Figure 3

[0105]

[0106]

[0107]

[0108] ​​The tracking of the target object can be performed through the first object recognition result, the second object recognition result, the observable object information in the initial object information, and the unobservable object information in the target object information in the first blind area, so as to determine which target objects should exist in the unobservable object information of each second sub-blind area, update the unobservable object information in the initial object information, and finally obtain the target object information in the second blind area.

[0109] In S130, the driving state of the target vehicle is controlled based on the second blind area information.

[0110] According to the target object information in the second blind area, the type, quantity, size, direction, and other information of the target object existing in each second sub-blind area of the second blind area can be obtained, and the driving state of the target vehicle can be controlled according to the above information to reduce the risk of automatic driving.

[0111] In some possible embodiments, the driving state of the target vehicle can be controlled by the following steps:

[0112] Based on the unobservable object information of the target object information in the second blind area, the type and quantity of the unobservable object in each second sub-blind area are determined.

[0113] Based on the type and quantity of the unobservable object in each second sub-blind area, the corresponding danger level of each second sub-blind area is determined.

[0114] Based on the corresponding danger level of each second sub-blind area, the driving state of the target vehicle is controlled.

[0115] The danger level can be related to the type and quantity of the unobservable object, for example, the more pedestrians and bicycles (types of road vulnerable users) in the second sub-blind area, the higher the danger level, and the more unobservable objects in the second sub-blind area, the higher the danger level. Further, the danger level can also be related to the position information of the second sub-blind area, for example, the closer the second sub-blind area to the future trajectory of the target vehicle, the higher the danger level.

[0116] The higher the danger level of a certain second sub-blind area, the more likely an accident occurs near the second sub-blind area, such as a sudden rush out of the vehicle or a pedestrian. The target vehicle can be controlled to slow down or adjust the driving route of the target vehicle at a position with a high danger level to avoid accidents, thereby improving the safety of automatic driving of the target vehicle.

[0117] The above embodiment utilizes the position information of the second blind area, the position information of the first blind area, and the object information in the first blind area, so that the second blind area can inherit the object information in the first blind area of the historical frame to obtain the initial object information in the second blind area. Then, the initial object information is updated based on the target object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area to obtain the target object information in the second blind area, so as to determine the obstacle in the blind area and control the driving state of the vehicle according to the target object information.

[0118] Corresponding to the above vehicle control method, the disclosure also discloses a vehicle control device. Each module in the device can implement each step in the positioning method of each embodiment described above and can achieve the same beneficial effects, so the same parts will not be described here. Specifically, as shown in Figure 4 The vehicle control device comprises:

[0119] The acquisition module 410 is configured to acquire first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in a historical frame before a current frame, a first object recognition result corresponding to the first point cloud data, and second point cloud data collected in the current frame.

[0120] The determination module 420 is configured to determine second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information, and the first object recognition result.

[0121] The control module 430 is configured to control a driving state of the target vehicle based on the second blind area information.

[0122] In a possible implementation, the first blind area information comprises position information of the first blind area in the first point cloud data, and the second blind area information comprises position information of the second blind area in the second point cloud data.

[0123] The determination module 420 is specifically configured to:

[0124] determine a second object recognition result corresponding to the second point cloud data and the position information of the second blind area.

[0125] In a possible implementation, the first blind area information further comprises object information in the first blind area, the second blind area information further comprises target object information in the second blind area, and the determination module 420, when determining the second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information, and the first object recognition result, is configured to:

[0126] determine initial object information in the second blind area based on the position information of the second blind area, the position information of the first blind area, and object information in the first blind area;

[0127] update the initial object information based on the object information in the first blind area, the second object recognition result, the first object recognition result, and the position information of the second blind area, to obtain target object information in the second blind area.

[0128] In a possible implementation, the determining module 420 is specifically configured to:

[0129] determine information of an obstacle within a set range from the target vehicle in the second point cloud data based on the second point cloud data;

[0130] determine blind area position information corresponding to the second point cloud data based on the beam information emitted by the radar and the determined information of the obstacle.

[0131] In a possible implementation, the first blind area includes at least one first sub-blind area, and the second blind area includes at least one second sub-blind area.

[0132] The determining module 430 is specifically configured to:

[0133] determine an association relationship between each second sub-blind area and each first sub-blind area based on the position information of the first blind area and the position information of the second blind area;

[0134] determine initial object information in each second sub-blind area based on the determined association relationship and target object information in each first sub-blind area.

[0135] In a possible implementation, when determining the association relationship between each second sub-blind area and each first sub-blind area based on the position information of the first blind area and the position information of the second blind area, the determining module 420 is configured to:

[0136] determine an overlapping area between each second sub-blind area and each first sub-blind area based on the position information of the first blind area and the position information of the second blind area;

[0137] For any second sub-blind area, determine the association relationship between the second sub-blind area and each first sub-blind area based on an area of the overlapping area between the second sub-blind area and each first sub-blind area.

[0138] In a possible implementation, the determining module 420 is configured to:

[0139] For any second sub-blind area, determine initial object information in the second sub-blind area based on object information in each first sub-blind area and a correlation between the second sub-blind area and each first sub-blind area.

[0140] In a possible implementation, the target object information includes observable object information and unobservable object information, and the initial object information includes observable object information and unobservable object information.

[0141] The determining module 420 is configured to update the initial object information based on the target object information in the first blind area, the second object recognition result, the first object recognition result, and position information of the second blind area, to obtain target object information in the second blind area.

[0142] Update observable object information in the initial object information based on the first object recognition result, the second object recognition result, and the position information of the second blind area.

[0143] Update unobservable object information in the initial object information based on the first object recognition result, the second object recognition result, observable object information in the initial object information, and unobservable object information in the target object information in the first blind area.

[0144] Determine that the updated initial object information is the target object information in the second blind area.

[0145] In a possible implementation, the control module 430 is specifically configured to:

[0146] Determine types and quantities of unobservable objects in each second sub-blind area based on unobservable object information of the target object information in the second blind area.

[0147] Determine a danger level corresponding to each second sub-blind area based on the types and quantities of unobservable objects in each second sub-blind area.

[0148] Control a driving state of the target vehicle based on the danger level corresponding to each second sub-blind area.

[0149] Corresponding to the vehicle control method, the present embodiment also provides an electronic device 500, as shown in Figure 5 FIG. 1 is a structural schematic diagram of an electronic device 500 provided by the present embodiment, which includes:

[0150] The processor 51, the memory 52, and the bus 53; the memory 52 is used to store execution instructions, including the internal memory 521 and the external memory 522; the internal memory 521 here is also called the internal memory, used to temporarily store the operation data in the processor 51 and the data exchanged with the external memory 522 such as a hard disk, the processor 51 exchanges data with the external memory 522 through the internal memory 521, when the electronic device 500 is running, the processor 51 and the memory 52 communicate through the bus 53, so that the processor 51 executes the following instructions:

[0151] Obtain first blind area information corresponding to first point cloud data collected by a radar installed on a target vehicle in historical frames before a current frame, first object recognition results corresponding to the first point cloud data, and second point cloud data collected in the current frame;

[0152] Determine second blind area information corresponding to the second point cloud data based on the second point cloud data, the first blind area information, and the first object recognition results;

[0153] Control a driving state of the target vehicle based on the second blind area information.

[0154] The embodiment of the disclosure also provides a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program is run by a processor, the steps of the vehicle control method described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer readable storage medium.

[0155] The embodiment of the disclosure also provides a computer program product, including a computer readable storage medium storing program codes, and the program codes include instructions for executing the steps of the vehicle control method described in the above method embodiment. For details, refer to the above method embodiment, which will not be repeated here.

[0156] The computer program product can be specifically implemented by hardware, software, or a combination thereof. In an optional embodiment, the computer program product is specifically embodied as a computer storage medium, and in another optional embodiment, the computer program product is specifically embodied as a software product, such as a software development kit (Software Development Kit, SDK), etc.

[0157] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here. In several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and another division can be made in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some communication interfaces, devices or units, and can be electrical, mechanical or other forms.

[0158] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0159] In addition, each functional unit in each embodiment of the present disclosure can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0160] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present disclosure essentially or the part of the prior art or the part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present disclosure. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, and various program code storage media.

[0161] Finally, it should be noted that the above-described embodiments are merely specific embodiments of the present disclosure, used to illustrate the technical solutions of the present disclosure, and are not intended to limit the present disclosure. The protection scope of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can make modifications or easy changes to the technical solutions described in the foregoing embodiments, or easily think of changes or equivalent replacements for some of the technical features; and these modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A vehicle control method, characterized in that, include: Acquire the first blind zone information corresponding to the first point cloud data collected by the radar installed on the target vehicle in historical frames before the current frame, the first object recognition result corresponding to the first point cloud data, and the second point cloud data collected in the current frame; Based on the second point cloud data, the first blind zone information, and the first object recognition result, the second blind zone information corresponding to the second point cloud data is determined. Based on the second blind spot information, the driving status of the target vehicle is controlled; The first blind zone information includes the location information of the first blind zone in the first point cloud data, and the second blind zone information includes the location information of the second blind zone in the second point cloud data; The step of determining the second blind zone information corresponding to the second point cloud data based on the second point cloud data, the first blind zone information, and the first object recognition result includes: Determine the second object recognition result corresponding to the second point cloud data and the location information of the second blind zone; The first blind zone information also includes object information within the first blind zone; the second blind zone information also includes target object information within the second blind zone; The step of determining the second blind zone information corresponding to the second point cloud data based on the second point cloud data, the first blind zone information, and the first object recognition result includes: Based on the location information of the second blind zone, the location information of the first blind zone, and the object information within the first blind zone, the initial object information within the second blind zone is determined; Based on the object information within the first blind zone, the second object recognition result, the first object recognition result, and the location information of the second blind zone, the initial object information is updated to obtain the target object information within the second blind zone.

2. The method according to claim 1, characterized in that, The location information of the second blind zone corresponding to the second point cloud data is determined by the following steps: Based on the second point cloud data, information about obstacles within a set range from the target vehicle in the second point cloud data is determined; Based on the radar-transmitted beam information and the identified obstacle information, the blind spot location information corresponding to the second point cloud data is determined.

3. The method according to claim 1 or 2, characterized in that, The first blind zone includes at least one first sub-blind zone, and the second blind zone includes at least one second sub-blind zone; The step of determining the initial object information within the second blind zone based on the location information of the second blind zone, the location information of the first blind zone, and the object information within the first blind zone includes: Based on the location information of the first blind zone and the location information of the second blind zone, the association relationship between each second sub-blind zone and each first sub-blind zone is determined respectively; Based on the established association and the object information within each of the first sub-blind zones, the initial object information within each of the second sub-blind zones is determined.

4. The method according to claim 3, characterized in that, The step of determining the association between each second sub-blind zone and each first sub-blind zone based on the location information of the first blind zone and the location information of the second blind zone includes: Based on the location information of the first blind zone and the location information of the second blind zone, the overlapping area between each second sub-blind zone and each first sub-blind zone is determined; For any second sub-blind zone, the correlation between the second sub-blind zone and each first sub-blind zone is determined based on the area of ​​the overlapping region between the second sub-blind zone and each first sub-blind zone.

5. The method according to claim 3 or 4, characterized in that, The determination of initial object information in each second sub-blind zone based on the established association relationships and target object information within each first sub-blind zone includes: For any second sub-blind zone, based on the object information in each of the first sub-blind zones and the correlation between the second sub-blind zone and each of the first sub-blind zones, the initial object information in the second sub-blind zone is determined.

6. The method according to claim 1, characterized in that, The target object information includes observable object information and unobservable object information; the initial object information includes observable object information and unobservable object information. The step of updating the initial object information based on object information within the first blind zone, the second object recognition result, the first object recognition result, and the location information of the second blind zone to obtain target object information within the second blind zone includes: Based on the first object recognition result, the second object recognition result, and the location information of the second blind zone, update the observable object information in the initial object information; Based on the first object recognition result, the second object recognition result, the observable object information in the initial object information, and the unobservable object information in the object information within the first blind zone, update the unobservable object information in the initial object information; The updated initial object information is determined to be the target object information within the second blind zone.

7. The method according to claim 6, characterized in that, The step of controlling the driving state of the target vehicle based on the second blind spot information includes: Based on the unobservable object information of the target object information in the second blind zone, determine the type and quantity of unobservable objects in each of the second sub-blind zones; Based on the type and number of unobservable objects within each second sub-blind zone, the hazard level corresponding to each second sub-blind zone is determined; The driving status of the target vehicle is controlled based on the danger level corresponding to each of the second sub-blind zones.

8. A vehicle control device, characterized in that, include: The acquisition module is used to acquire the first blind zone information corresponding to the first point cloud data collected by the radar installed on the target vehicle in the previous historical frames before the current frame, the first object recognition result corresponding to the first point cloud data, and the second point cloud data collected in the current frame. The determination module is used to determine the second blind zone information corresponding to the second point cloud data based on the second point cloud data, the first blind zone information, and the first object recognition result; The control module is used to control the driving state of the target vehicle based on the second blind spot information; The first blind zone information includes the location information of the first blind zone in the first point cloud data, and the second blind zone information includes the location information of the second blind zone in the second point cloud data; The determining module is specifically used for: Determine the second object recognition result corresponding to the second point cloud data and the location information of the second blind zone; The first blind zone information also includes object information within the first blind zone; the second blind zone information also includes target object information within the second blind zone; The determining module is specifically used for: Based on the location information of the second blind zone, the location information of the first blind zone, and the object information within the first blind zone, the initial object information within the second blind zone is determined; Based on the object information within the first blind zone, the second object recognition result, the first object recognition result, and the location information of the second blind zone, the initial object information is updated to obtain the target object information within the second blind zone.

9. An electronic device, characterized in that, include: The processor and the memory, the memory storing machine-readable instructions executable by the processor, the processor executing the machine-readable instructions stored in the memory, wherein when the machine-readable instructions are executed by the processor, the processor performs the steps of the vehicle control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a computer device, performs the steps of the vehicle control method as described in any one of claims 1 to 7.

Images