Method and device for determining a dangerous blind area and an unmanned vehicle

Abstract

The present disclosure provides a method and device for determining a dangerous blind area, electronic equipment, storage medium and program product, which relates to the field of artificial intelligence, and specifically relates to the technical field of automatic driving, intelligent transportation and the like. The specific implementation scheme is as follows: obtaining scene data representing a scene where a target vehicle is located and obstacle data representing obstacles around the target vehicle; constructing a blind area of the target vehicle caused by the obstacles according to the obstacle data; determining a dangerous point in the scene which constitutes a danger to the target vehicle according to the scene data; and determining whether the blind area is a dangerous blind area according to the relative relationship between the blind area and the dangerous point. The present disclosure can filter out non-dangerous blind areas, thereby improving the determination accuracy and effectiveness of the dangerous blind area.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, specifically to the fields of autonomous driving and intelligent transportation, and in particular to methods, devices, electronic devices, storage media, computer program products, and unmanned vehicles for determining dangerous blind spots, which can be used in autonomous driving scenarios. Background Technology

[0002] Normally, human drivers will intentionally slow down when approaching blind spots to avoid collisions with vehicles that may suddenly appear from those blind spots. However, with V2X (vehicle-to-everything) technology not yet widely deployed, autonomous driving solutions are still primarily based on single-vehicle intelligence. Under this single-vehicle intelligence approach, blind spot problems persist, and autonomous vehicles remain susceptible to collisions when navigating potentially dangerous blind spots. Current autonomous driving systems are not accurate enough in their assessment of potentially dangerous blind spots. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, storage medium, computer program product, and unmanned vehicle for determining hazardous blind spots.

[0004] According to the first aspect, a method for determining a dangerous blind spot is provided, comprising: acquiring scene data representing the scene in which the target vehicle is located and obstacle data representing obstacles around the target vehicle; constructing a blind spot of the target vehicle caused by the obstacles based on the obstacle data; determining the danger points in the scene that pose a danger to the target vehicle based on the scene data; and determining whether the blind spot is a dangerous blind spot based on the relative relationship between the blind spot and the danger points.

[0005] According to a second aspect, a device for determining a dangerous blind spot is provided, comprising: an acquisition unit configured to acquire scene data representing the scene in which a target vehicle is located and obstacle data representing obstacles around the target vehicle; a construction unit configured to construct a blind spot of the target vehicle caused by the obstacles based on the obstacle data; a first determination unit configured to determine a hazard point in the scene that poses a danger to the target vehicle based on the scene data; and a second determination unit configured to determine whether the blind spot is a dangerous blind spot based on the relative relationship between the blind spot and the hazard point.

[0006] According to a third aspect, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method as described in any implementation of the first aspect.

[0007] According to a fourth aspect, a non-transitory computer-readable storage medium is provided that stores computer instructions for causing a computer to perform the method described in any implementation of the first aspect.

[0008] According to a fifth aspect, a computer program product is provided, comprising: a computer program that, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0009] According to the sixth aspect, an unmanned vehicle is provided, including: electronic devices as described in the third aspect.

[0010] According to the technology disclosed herein, a method for determining dangerous blind spots is provided. For blind spots caused by surrounding obstacles, the method determines whether a blind spot is a dangerous blind spot based on the relative relationship between the dangerous points that may pose a danger to the target vehicle in the scene where the target vehicle is located and the blind spot. Non-dangerous blind spots can be filtered out, thereby improving the accuracy and effectiveness of determining dangerous blind spots.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0012] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0013] Figure 1 This is an exemplary system architecture diagram that can be applied to an embodiment of this disclosure;

[0014] Figure 2 This is a flowchart of an embodiment of the method for determining hazardous blind zones according to this disclosure;

[0015] Figure 3 This is a schematic diagram illustrating the construction of the blind spot of the target vehicle according to this embodiment;

[0016] Figure 4 This is a schematic diagram illustrating an application scenario of the method for determining dangerous blind zones according to this embodiment;

[0017] Figure 5 This is a schematic diagram illustrating the determination of hazardous points in a pedestrian crossing scenario according to this embodiment.

[0018] Figure 6 This is yet another schematic diagram illustrating the determination of hazardous points in a pedestrian crossing scenario according to this embodiment;

[0019] Figure 7 This is a schematic diagram of an intersection scenario according to this embodiment;

[0020] Figure 8 This is a flowchart of yet another embodiment of the method for determining hazardous blind zones according to this disclosure;

[0021] Figure 9 This is a flowchart of yet another embodiment of the method for determining hazardous blind zones according to this disclosure;

[0022] Figure 10 This is a structural diagram of one embodiment of the device for determining dangerous blind spots according to the present disclosure;

[0023] Figure 11 This is a schematic diagram of the structure of a computer system suitable for implementing the embodiments of the present disclosure. Detailed Implementation

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

[0025] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0026] Figure 1 An exemplary architecture 100 is shown for a method and apparatus for determining hazardous blind spots to which this disclosure can be applied.

[0027] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. The communication connections between terminal devices 101, 102, and 103 form a network topology. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0028] Terminal devices 101, 102, and 103 can be hardware or software that supports network connectivity for data interaction and processing. When terminal devices 101, 102, and 103 are hardware, they can be various electronic devices supporting network connectivity, information acquisition, interaction, display, and processing functions, including but not limited to in-vehicle computers, radar, cameras, smartphones, tablets, e-book readers, laptops, and desktop computers. When terminal devices 101, 102, and 103 are software, they can be installed in the aforementioned electronic devices. They can be implemented as multiple software programs or software modules, for example, to provide distributed services, or as a single software program or software module. No specific limitations are imposed here.

[0029] Server 105 can be a server that provides various services, such as a background processing server that determines dangerous blind spots that pose a danger to the target vehicle based on scene data representing the scene in which the target vehicle is located and obstacle data representing the obstacles around the target vehicle provided by terminal devices 101, 102, and 103. As an example, server 105 can be a cloud server.

[0030] It should be noted that a server can be either hardware or software. When the server is hardware...

[0031] In this case, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single 5-server. When the server is software, it can be implemented as multiple software programs or software modules (e.g., used for...).

[0032] Software or software modules that provide distributed services can also be implemented as a single software or software module.

[0033] No specific limitations are specified here.

[0034] It should also be noted that the method for determining the dangerous blind zone provided in the embodiments of this disclosure can

[0035] The execution can be performed by the server, by the terminal device, or by a combination of both. Correspondingly, the various components (e.g., units) of the blind spot determination device can be entirely located in the server, entirely located in the terminal device, or separately located in both the server and the terminal device.

[0036] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative.

[0037] Depending on the implementation requirements, it can have any number of terminal devices, networks, and servers. When the method for determining the hazard blind zone is in place, the electronic devices operating on it do not need to transmit data to other electronic devices.

[0038] During transmission, the system architecture may consist only of electronic devices (such as servers or terminal devices) on which the method for determining dangerous blind spots operates.

[0039] Please refer to Figure 2 , Figure 2 A flowchart of a method for determining a dangerous blind zone provided in this disclosure embodiment is shown, wherein process 200 includes the following steps:

[0040] Step 201: Obtain scene data representing the scene where the target vehicle is located and obstacle data representing the obstacles around the target vehicle.

[0041] In this embodiment, the entity executing the method for determining dangerous blind zones (e.g., Figure 1 Terminal in

[0042] The device or server can acquire scene data representing the scene in which the target vehicle is located and obstacle data representing the obstacles around the target vehicle from a remote location or from a local location via a wired network connection or a wireless network connection.

[0043] Scene data represents the environment in which the target vehicle is located, such as the vehicle's position and nearby scene data like lane lines, pedestrian crossings, and intersections. Obstacle data is...

[0044] Data that characterizes obstacles (e.g., vehicles, pedestrians) around the target vehicle, such as obstacle type, size, outline, speed, and other obstacle data.

[0045] To improve the effectiveness of information processing and reduce the amount of information to be processed, the aforementioned executing entity can acquire scene data and obstacle data within a predetermined distance range from the target vehicle along its direction of travel. The predetermined distance range can be specifically set according to actual conditions; for example, a predetermined distance range of 50 meters.

[0046] As an example, the aforementioned execution entity can determine scene data in the following way: First, determine the real-time location of the target vehicle based on the positioning module corresponding to the target vehicle; then, based on the real-time location, determine the scene data within a certain distance range of the target vehicle from the high-precision map.

[0047] The aforementioned implementing entity can determine obstacle data in the following way: by sensing obstacles around the target vehicle through cameras, lidar, and other sensors corresponding to the target vehicle, and obtaining obstacle data.

[0048] Step 202: Based on the obstacle data, construct the blind spot of the target vehicle caused by the obstacle.

[0049] In this embodiment, the aforementioned execution entity can construct the blind spot of the target vehicle caused by the obstacle based on the obstacle data.

[0050] The presence of surrounding obstacles prevents the target vehicle from observing what lies behind them, creating a blind spot. As an example, based on the location and outline of the obstacle in the data, and referring to the target vehicle's viewing angle, a designated area behind the obstacle, out of sight, is determined as the blind spot caused by the obstacle.

[0051] As another example, based on the matching relationship between the position and contour of the obstacle data and the point cloud data, the obstacle point cloud representing the obstacle is determined; then, referring to the perspective of the target vehicle observing the obstacle, the obstacle point cloud is projected to determine the blind spot of the target vehicle caused by the presence of the obstacle.

[0052] In some optional implementations of this embodiment, the execution entity can perform step 202 as follows:

[0053] First, based on the location of the point cloud generation device on the target vehicle used to generate obstacle point cloud data, the obstacle point cloud data is projected onto the ground to obtain the obstacle projection point cloud.

[0054] Second, construct the blind spot of the target vehicle based on the point cloud projection of obstacles.

[0055] Continue to refer to Figure 3 The diagram 300 illustrates the construction of the blind spot of a target vehicle. A point cloud generation device (e.g., lidar) 302 is installed on the target vehicle 301, and obstacles 303 surround the target vehicle 301. The aforementioned execution entity projects the obstacle point cloud of the obstacles 303 onto the ground from the position of the point cloud generation device 302, obtaining obstacle projection point clouds 304 and 305. Furthermore, the blind spot of the target vehicle can be constructed by combining the obstacle projection point clouds 304 and 305.

[0056] In this implementation, the obstacle point cloud is projected onto the ground from the position of the point cloud generation device to ultimately determine the blind zone caused by the obstacle, thereby improving the accuracy of the determined blind zone.

[0057] In some optional implementations of this embodiment, the execution entity can perform the second step as follows: First, construct the minimum convex hull polygon including the obstacle projection point cloud; then, construct the blind spot of the target vehicle based on the minimum convex hull polygon.

[0058] Continue to refer to Figure 3After determining the obstacle projection point clouds 304 and 305, the minimum convex hull algorithm can be used to determine the minimum convex hull polygon 306 that includes the obstacle projection point clouds; then, the minimum convex hull polygon 306 is determined as the blind spot of the target vehicle caused by the obstacle 303.

[0059] In this implementation, the minimum convex hull algorithm is used to further process the obstacle projection point cloud, which further improves the accuracy of the determined blind zone.

[0060] Step 203: Based on the scene data, identify the danger points in the scene that pose a danger to the target vehicle.

[0061] In this embodiment, the aforementioned execution entity can determine the danger points in the scene that pose a risk to the target vehicle based on the scene data. A danger point is a node in the scene represented by the scene data that may cause danger to the target vehicle's driving process.

[0062] As an example, the aforementioned executing entity can, based on big data analysis of scene data, statistically identify the hazards in each scene and generate a correspondence table representing the relationship between scenes and hazards. After obtaining the scene data, the executing entity determines the hazards in the scene represented by the scene data according to the correspondence table.

[0063] As another example, the aforementioned implementing entity can use a trained hazard identification model to determine the hazards in the scene represented by the scene data. The hazard identification model is trained as follows: First, a training sample set is obtained, where the training samples in the training sample set include scene data and labels representing the hazards in the scene data; then, a machine learning method is used, with the scene data as input and the labels corresponding to the input scene data as the expected output, to train the hazard identification model.

[0064] Step 204: Determine whether the blind spot is a dangerous blind spot based on the relative relationship between the blind spot and the danger point.

[0065] In this embodiment, the aforementioned execution entity can determine whether a blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point.

[0066] There may be multiple obstacles around the target vehicle, thus identifying multiple blind spots. There may also be multiple hazards in the scene in which the target vehicle is located. For each blind spot, the aforementioned executing entity can determine whether the blind spot includes a hazard. If it is determined that the blind spot includes at least one hazard, it is designated as a hazardous blind spot; if it is determined that the blind spot does not contain any hazard, it is designated as a non-hazardous blind spot.

[0067] Dangerous blind spots indicate blind areas with potential dangers, such as pedestrians or other vehicles that could easily dart out of them; non-dangerous blind spots indicate blind areas without potential dangers.

[0068] See also Figure 4 , Figure 4 This is a schematic diagram 400 illustrating an application scenario of the method for determining hazardous blind spots according to this embodiment. Figure 4 In the application scenario, the autonomous vehicle 401 acquires scene data representing the scene in which the target vehicle is located and obstacle data representing the obstacles around the target vehicle in real time during driving. The scene in which the target vehicle is located is a pedestrian crossing scene, and the obstacles around the target vehicle include vehicles 402 and 403. After determining the obstacle data, the autonomous vehicle constructs blind spots 404 and 405 caused by obstacle 402 and obstacle 403, respectively. After determining the scene data, the autonomous vehicle 401 identifies hazard points 406 in the pedestrian crossing scene that pose a danger to the target vehicle. In a pedestrian crossing scene, hazard points are generally located on the pedestrian crossing. Finally, in response to determining that blind spot 404 includes hazard point 406, the autonomous vehicle 401 determines blind spot 404 as a dangerous blind spot; in response to determining that blind spot 405 does not include hazard point 406, the autonomous vehicle 401 determines blind spot 405 as a non-dangerous blind spot.

[0069] This embodiment provides a method for determining dangerous blind spots. For blind spots caused by surrounding obstacles, the method determines whether a blind spot is a dangerous blind spot based on the relative relationship between the dangerous points that may pose a danger to the target vehicle in the scene where the target vehicle is located and the blind spot. Non-dangerous blind spots can be filtered out, thereby improving the accuracy and effectiveness of determining dangerous blind spots.

[0070] In some optional implementations of this embodiment, the execution entity can perform step 203 as follows:

[0071] First, based on the scene data, determine the scene type corresponding to the scene.

[0072] As an example, the aforementioned executing entity or the electronic device communicatively connected to the aforementioned executing entity is provided with a scene type set, which includes all scene types. The aforementioned executing entity determines the matching scene type from the scene type set by processing and analyzing the scene data.

[0073] Second, based on the scenario type, identify the danger points in the scenario that pose a risk to the target vehicle.

[0074] As an example, the aforementioned execution entity includes a mapping table that represents the correspondence between scene types and hazards. After determining the scene type, the execution entity identifies the hazards within the scene corresponding to that scene type based on the mapping table.

[0075] It is understandable that different scenarios belonging to the same scenario type generally share common hazards. In this implementation, hazards in a scenario are determined based on the scenario type. This ensures the accuracy of the identified hazards while reducing the amount of data processed and alleviating data processing pressure.

[0076] In some optional implementations of this embodiment, the scenario type includes a pedestrian crossing scenario. In this implementation, the execution entity can perform the second step as follows:

[0077] First, determine the relative position of the blind spot and the target vehicle in the pedestrian crossing scenario; then, based on the relative position and the center point of the pedestrian crossing in the scenario, determine the danger points in the pedestrian crossing scenario.

[0078] The relative position of the blind spot to the target vehicle includes the location of the blind spot to the left front and right front of the target vehicle in its direction of travel. Based on the position of the target vehicle and the position of the blind spot, their relative position can be easily determined.

[0079] In this implementation, the aforementioned execution entity, based on a high-precision map, can determine the center point of the pedestrian crossing; then, according to the relative orientation, the center point is moved in a preset manner to determine the danger points in the pedestrian crossing scenario.

[0080] In this implementation, in pedestrian crossing scenarios, the relative orientation of blind spots and target vehicles, as well as the center point of the pedestrian crossing, are used to determine the danger points in the pedestrian crossing scenario, thereby improving the accuracy of the danger points in the pedestrian crossing scenario.

[0081] Continue to refer to Figure 5 This diagram illustrates the identification of hazardous points in a pedestrian crossing scenario.

[0082] In some optional implementations of this embodiment, the aforementioned execution entity can determine the danger points in the pedestrian crossing scenario in the following manner:

[0083] First, the ordinate of the center point is determined as the ordinate of the danger point.

[0084] In this implementation, the SL coordinate system is generally used.

[0085] Then, based on the center point, extend forward and backward by a preset distance along the road extension direction to obtain the first extension point and the second extension point.

[0086] The preset distance can be set according to the actual situation; for example, the preset distance is 10 meters. Based on the center point 501, the preset distance is extended forward along the road direction to obtain the first extension point, and the preset distance is extended backward to obtain the second extension point.

[0087] Finally, in response to determining that the blind spot is located to the left front of the target vehicle, the smaller of the abscissas of the left boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

[0088] Among the left boundary point 502 of the road corresponding to the first extension point and the left boundary point 503 of the road corresponding to the second extension point, the abscissa of the left boundary point 502 is smaller. Therefore, the danger point is determined based on the ordinate of the center point 501 and the abscissa of the left boundary point 502.

[0089] This implementation provides a specific method for determining the danger point when the blind spot is located in front of the left side of the target vehicle, improving the consistency and accuracy of the determined danger point with the actual situation.

[0090] Corresponding to the above-mentioned implementation method for determining dangerous points, in some optional implementation methods of this embodiment, the above-mentioned execution subject can perform the above-mentioned step 204 in the following way: traverse the blind spot located in front of the left side of the target vehicle in the pedestrian crossing scene, and in response to determining that there is a dangerous point located in the blind spot among the dangerous points obtained according to the horizontal coordinate of the left boundary of the road, determine that the blind spot is a dangerous blind spot.

[0091] In this implementation, for the blind spot located to the left front of the target vehicle, it is determined whether it includes a hazard point obtained based on the abscissa of the left boundary of the road. If yes, the blind spot is identified as a hazard blind spot; otherwise, it is identified as a non-hazard blind spot. This further improves the accuracy of hazard blind spot identification in pedestrian crossing scenarios.

[0092] Continue to refer to Figure 6 This diagram illustrates another way to identify danger points in a pedestrian crossing scenario.

[0093] In some optional implementations of this embodiment, the execution entity may also perform the following operation: in response to determining that the blind spot is located to the right front of the target vehicle, the smaller of the abscissas of the right boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

[0094] Based on center point 601, extend forward by a predetermined distance along the road direction to obtain a first extension point, and extend backward by a predetermined distance to obtain a second extension point. Among the right boundary point 602 of the road corresponding to the first extension point 602 and the right boundary point 603 of the road corresponding to the second extension point, the abscissa of the right boundary point 602 is smaller. Therefore, the danger point is determined based on the ordinate of center point 601 and the abscissa of right boundary point 602.

[0095] In this implementation, similar to the method for determining the danger point when the blind spot is located to the left front of the target vehicle, a method for determining the danger point when the blind spot is located to the right front of the target vehicle is provided, which improves the consistency and accuracy of the determined danger point with the actual situation.

[0096] Corresponding to the above-mentioned implementation method for determining dangerous points, in some optional implementation methods of this embodiment, the above-mentioned execution subject can perform the above-mentioned step 204 in the following way: traverse the blind spot located in front of the target vehicle on the right side of the pedestrian crossing scene, and in response to determining that there is a dangerous point located in the blind spot among the dangerous points obtained according to the horizontal coordinate of the right boundary of the road, determine that the blind spot is a dangerous blind spot.

[0097] In this implementation, for the blind spot located to the right front of the target vehicle, it is determined whether it includes a hazard point obtained based on the abscissa of the right boundary of the road. If yes, the blind spot is identified as a hazard blind spot; otherwise, it is identified as a non-hazard blind spot. This further improves the accuracy of hazard blind spot identification in pedestrian crossing scenarios.

[0098] Continue to refer to Figure 7 The diagram shows a scene at an intersection.

[0099] In some optional implementations of this embodiment, the scenario type includes an intersection scenario. In this implementation, the aforementioned execution entity can perform the second step as follows:

[0100] Step (1): Based on the target vehicle's trajectory in the intersection scene, determine the target lane segment in other lanes that intersect with the target vehicle's driving lane in the intersection scene, and the detection direction for the danger point.

[0101] The trajectory to be executed is the path of the target vehicle as it travels within the intersection, and it can be determined based on the autonomous driving strategy of the autonomous vehicle. In an intersection scenario, the target vehicle's lane typically intersects with other lanes, and the intersecting lane segments usually include multiple segments. For example, if target vehicle 701 is traveling straight in lane 702 at the intersection, then lane 702 intersects with target lane segment 704 in the left-turn lane 703 of the opposite lane. In addition, it also includes intersecting lanes such as straight-ahead lanes and left-turn lanes in lanes perpendicular to lane 702.

[0102] Each lane can be formed by connecting multiple lane segments, and the intersecting lane segments are the target lane segments.

[0103] Based on the target vehicle's trajectory at the intersection, the side that needs to be focused on during the journey can be determined, thereby determining the detection direction for dangerous points.

[0104] Step (2): Determine the relative orientation between the target lane segment and the target vehicle based on the road nodes in the obscured portion of the target lane segment that is at least partially obscured by the blind spot.

[0105] The target lane segment may be partially or completely obscured by blind spots. When the target lane segment is completely obscured by blind spots, the entire target lane segment is considered the obscured portion. Based on the relative positional relationship between road nodes within the obscured portion and the target vehicle, the relative positional relationship between the target lane segment and the target vehicle can be determined. This relative positional relationship may include, for example, the target lane segment being to the left or right of the target vehicle.

[0106] Step (3) Identify the danger points from the road nodes in the obscured part of the target.

[0107] The obscured portion of the target refers to the obscured part of the target lane segment that corresponds to the relative orientation relationship of the detection direction.

[0108] The detection direction is specifically manifested as directional information, and the relative orientation relationship is also essentially directional information. When the two match, it is a relative orientation relationship that conforms to the detection direction.

[0109] As an example, the aforementioned executing entity can identify all road nodes in the obscured portion of the target as danger points.

[0110] In this implementation, the aforementioned execution entity determines the hazard points in the intersection scenario from the occluded portion of the target lane segment corresponding to the relative orientation relationship that conforms to the detection direction, thereby improving the accuracy of hazard point determination in the intersection scenario.

[0111] In some optional implementations of this embodiment, the execution entity can perform the above step (2) in the following manner: First, determine the target road node closest to the target vehicle from the road nodes in the obscured part; then, determine the relative orientation relationship between the target road node and the target vehicle as the relative orientation relationship between the target lane segment and the target vehicle.

[0112] In real-world scenarios, autonomous vehicles prioritize the nearest hazardous points. This implementation determines the relative orientation of the target road node closest to the target vehicle within the obscured portion as the relative orientation between the target lane segment and the target vehicle, thus improving the accuracy and effectiveness of the relative orientation relationship.

[0113] In some optional implementations of this embodiment, the above-mentioned execution entity can perform the above step (1) in the following manner: in response to determining that the turning type corresponding to the trajectory to be run is left turn, the detection direction is determined to be right.

[0114] In this implementation, the above-mentioned execution entity can perform the above step (3) in the following way: in response to determining the detection direction as the right side, for the target occluded part in each target lane segment located to the right of the target vehicle, the target road node located in the target occluded part that is closest to the target vehicle is determined as the danger point.

[0115] In real-world scenarios, accidents are more likely to occur on the right side of a target vehicle during a left turn. By identifying the nearest road node to the target vehicle within the obscured portion of its right side during a left turn, the accuracy of identifying hazard points in intersection scenarios is further improved.

[0116] Corresponding to the above implementation method of determining danger points in the case of left turn, in some optional implementation methods of this embodiment, the above execution subject can perform the above step 204 in the following way: in response to determining that the turning type corresponding to the trajectory to be run is left turn, traverse each blind spot located on the right side of the target vehicle; in response to determining that there is a danger point located in the blind spot among the danger points located on the right side of the target vehicle, determine that the blind spot is a danger blind spot.

[0117] In this implementation, when turning left, for the blind spot located to the right of the target vehicle, it is determined whether it includes a dangerous point located to the right of the target vehicle. If yes, the blind spot is identified as a dangerous blind spot; otherwise, it is identified as a non-dangerous blind spot. This further improves the accuracy of identifying dangerous blind spots when turning left at intersections.

[0118] In some optional implementations of this embodiment, the above-mentioned execution entity can perform the above step (1) in the following manner: in response to determining that the turning type corresponding to the trajectory to be run is straight, the detection direction is determined to be right and left.

[0119] In this implementation, the above-mentioned execution entity can perform the above step (3) in the following way: in response to determining the detection direction as right and left, for the target occluded part in each target lane segment located on the right and left of the target vehicle, the target road node located in the target occluded part that is closest to the target vehicle is determined as the danger point.

[0120] In real-world scenarios, when a target vehicle is traveling straight, its right and left sides need to be monitored simultaneously. In this straight-ahead situation, the road node closest to the target vehicle within the obscured portions of its right and left sides is identified as a hazard point, further improving the accuracy of hazard point identification in straight-ahead situations at intersections.

[0121] Corresponding to the above implementation method for determining danger points in the straight-ahead scenario, in some optional implementation methods of this embodiment, the above-mentioned execution subject can perform the above step 204 in the following manner: In response to determining that the steering type corresponding to the trajectory to be run is straight-ahead, traverse each blind spot located on the right side of the target vehicle; in response to determining that there is a danger point located in the blind spot among the danger points located on the right side of the target vehicle, determine that the blind spot is a danger blind spot; and traverse each blind spot located on the left side of the target vehicle; in response to determining that there is a danger point located in the blind spot among the danger points located on the left side of the target vehicle, determine that the blind spot is a danger blind spot.

[0122] In this implementation, when proceeding straight ahead, for a blind spot located to the right of the target vehicle, it is determined whether it includes a hazard point located to the right of the target vehicle. If yes, the blind spot is designated as a hazardous blind spot; otherwise, it is designated as a non-hazardous blind spot. For a blind spot located to the left of the target vehicle, it is determined whether it includes a hazard point located to the left of the target vehicle. If yes, the blind spot is designated as a hazardous blind spot; otherwise, it is designated as a non-hazardous blind spot. This further improves the accuracy of hazardous blind spot determination in straight-ahead situations at intersections.

[0123] In some optional implementations of this embodiment, the above-mentioned execution entity can perform the above step (1) in the following manner: in response to determining that the turning type corresponding to the trajectory to be run is a right turn, the detection direction is determined to be left.

[0124] In this implementation, the above-mentioned execution entity can perform the above step (3) in the following way: in response to determining the detection direction as the left, for the target occluded part in each target lane segment located to the left of the target vehicle, the target road node located in the target occluded part that is closest to the target vehicle is identified as a danger point.

[0125] In real-world scenarios, accidents are more likely to occur on the left side of a target vehicle during a right turn. By identifying the nearest road node to the target vehicle within the obscured portion of its left side during a right turn as the hazard point, the accuracy of hazard point identification in intersection scenarios is further improved.

[0126] Corresponding to the above implementation method of determining danger points in the case of right turn, in some optional implementation methods of this embodiment, the above execution subject can perform the above step 204 in the following way: in response to determining that the turning type corresponding to the trajectory to be run is right turn, traverse each blind spot located on the left side of the target vehicle; in response to determining that there is a danger point located in the blind spot among the danger points located on the left side of the target vehicle, determine that the blind spot is a danger blind spot.

[0127] In this implementation, when turning right, for the blind spot located to the left of the target vehicle, it is determined whether it includes a dangerous point located to the left of the target vehicle. If yes, the blind spot is identified as a dangerous blind spot; otherwise, it is identified as a non-dangerous blind spot. This further improves the accuracy of identifying dangerous blind spots when turning right at intersections.

[0128] In some optional implementations of this embodiment, the execution entity may directly perform the following operation: determine the speed limit information of the target vehicle based on the scene type corresponding to the target blind spot in at least one blind spot.

[0129] Among them, the target hazard blind zone is the hazard blind zone corresponding to the hazard point closest to the target vehicle among the hazard points included in each of the at least one target hazard blind zone.

[0130] When there is one blind spot, that blind spot is the target blind spot; when there are multiple blind spots, the blind spot corresponding to the danger point closest to the target vehicle among the danger points included in each of the multiple target blind spots is the target blind spot.

[0131] Based on the scenario type corresponding to the target blind spot, the speed limit information for the target vehicle is determined. As an example, based on big data statistical analysis, the correspondence between each scenario type and speed limit information can be determined, thus enabling the determination of the corresponding speed limit information based on the scenario type.

[0132] In this implementation, the speed limit strategy is executed based on the target blind spot corresponding to the danger point closest to the target vehicle, which improves the accuracy and timeliness of the speed limit strategy execution timing; the speed limit information is determined according to the scene type corresponding to the target blind spot, which improves the fit between the speed limit information and the scene.

[0133] In some optional implementations of this embodiment, the execution entity may also perform the following operation: determine the length of the speed-limited road section for the target vehicle based on the speed data in the obstacle data of the obstacles constituting the target blind spot.

[0134] As an example, speed data is negatively correlated with the length of speed-limited road sections; that is, the lower the speed, the longer the speed-limited road section; and the higher the speed, the shorter the speed-limited road section.

[0135] In this implementation, the length of the speed-limited road section for the target vehicle is further adjusted based on the speed of the obstacle, thereby further improving the driving safety of the target vehicle in the speed-limited road section.

[0136] In some optional implementations of this embodiment, the execution entity may also perform the following operation: control the deceleration process of the target vehicle based on the speed limit information, the length of the speed limit section, and the preset maximum deceleration.

[0137] Specifically, within the speed-limited section corresponding to its length, the target vehicle uses the speed limit information as its desired speed value. Under the constraint of a preset maximum deceleration, the vehicle's speed gradually approaches the speed limit. The preset maximum deceleration can be set according to actual conditions and is not limited here. To further improve the fit between the preset maximum deceleration and the scenario, different preset maximum decelerations can be set for different scenarios.

[0138] In this implementation, different speed limit strategies are applied to target vehicles for different scenario types, which improves traffic efficiency while ensuring safety; in addition, the deceleration process is limited by a preset maximum deceleration, which improves the comfort of passengers in the vehicle during the deceleration process.

[0139] Continue to refer to Figure 8 The illustration shows a schematic flow 800 of another embodiment of the method for determining dangerous blind spots according to the present disclosure, including the following steps:

[0140] Step 801: Obtain scene data representing the scene where the target vehicle is located and obstacle data representing the obstacles around the target vehicle.

[0141] Step 802: Based on the position of the point cloud generation device on the target vehicle used to generate obstacle point cloud data in obstacle data, project the obstacle point cloud data onto the ground to obtain the obstacle projection point cloud.

[0142] Step 803: Construct the minimum convex hull polygon that includes the point cloud of the obstacle projection.

[0143] Step 804: Construct the blind spot of the target vehicle based on the minimum convex hull polygon.

[0144] Step 805: In response to the scene data, determine that the scene is a pedestrian crossing scene, and determine the relative orientation relationship between the blind spot in the pedestrian crossing scene and the target vehicle.

[0145] Step 806: Determine the ordinate of the center point as the ordinate of the danger point.

[0146] Step 807: Based on the center point, extend forward and backward by a preset distance along the road extension direction to obtain the first extension point and the second extension point.

[0147] Step 808: In response to determining that the blind spot is located to the left front of the target vehicle, the smaller of the abscissas of the left boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

[0148] Step 809: Traverse the blind spot located to the left front of the target vehicle in the pedestrian crossing scene. In response to determining the danger points obtained from the horizontal coordinate of the left boundary of the road, if there is a danger point located in the blind spot, then determine the blind spot as a danger blind spot.

[0149] Step 810: In response to determining that the blind spot is located to the right front of the target vehicle, the smaller of the abscissas of the right boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

[0150] Step 811: Traverse the blind spot located to the right front of the target vehicle in the pedestrian crossing scene. In response to determining the danger points obtained from the horizontal coordinate of the right boundary of the road, if there is a danger point located in the blind spot, then determine the blind spot as a danger blind spot.

[0151] As can be seen from this embodiment, with Figure 2 Compared with the corresponding embodiments, the flowchart 800 of the method for determining dangerous blind spots in this embodiment specifically illustrates the process of determining dangerous blind spots in pedestrian crossing scenarios, further improving the accuracy of determining dangerous blind spots in pedestrian crossing scenarios.

[0152] Continue to refer to Figure 9 The illustration shows a schematic flow 900 of another embodiment of the method for determining dangerous blind zones according to the present disclosure, including the following steps:

[0153] Step 901: Obtain scene data representing the scene where the target vehicle is located and obstacle data representing the obstacles around the target vehicle.

[0154] Step 902: Based on the position of the point cloud generation device on the target vehicle used to generate obstacle point cloud data in obstacle data, project the obstacle point cloud data onto the ground to obtain the obstacle projection point cloud.

[0155] Step 903: Construct the minimum convex hull polygon that includes the point cloud of the obstacle projection.

[0156] Step 904: Construct the blind spot of the target vehicle based on the minimum convex hull polygon.

[0157] Step 905: In response to determining the scene as an intersection scene based on the scene data, and based on the target vehicle's trajectory to be run in the intersection scene, determine the target lane segment in other lanes that intersect with the target vehicle's driving lane in the intersection scene.

[0158] Step 906: Determine the target road node closest to the target vehicle from the road nodes in the obscured portion.

[0159] Step 907: Determine the relative orientation relationship between the target road node and the target vehicle as the relative orientation relationship between the target lane segment and the target vehicle.

[0160] Step 908: In response to determining that the turning type corresponding to the trajectory to be run is left turn, the detection direction is determined to be right.

[0161] Step 909: In response to determining the detection direction as right, for the obscured portion of the target in each target lane segment located to the right of the target vehicle, the target road node located in the obscured portion that is closest to the target vehicle is identified as a danger point.

[0162] Step 910: In response to determining that the steering type corresponding to the trajectory to be run is a left turn, traverse each blind spot located on the right side of the target vehicle. In response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the right side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0163] Step 911: In response to determining that the steering type corresponding to the trajectory to be run is straight, the detection direction is determined to be right and left.

[0164] Step 912, in response to determining the detection direction as right and left, for the target occluded portion in each target lane segment located to the right and left of the target vehicle, the target road node located in the target occluded portion that is closest to the target vehicle is identified as a danger point.

[0165] Step 913: In response to determining that the steering type corresponding to the trajectory to be run is straight, traverse each blind spot located on the right side of the target vehicle. In response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the right side of the target vehicle, determine that the blind spot is a dangerous blind spot. Then traverse each blind spot located on the left side of the target vehicle. In response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the left side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0166] Step 914: In response to determining that the turning type corresponding to the trajectory to be run is right turn, the detection direction is determined to be left.

[0167] Step 915: In response to determining the detection direction as left, for the occluded portion of the target in each target lane segment located to the left of the target vehicle, the target road node located in the occluded portion that is closest to the target vehicle is identified as a danger point.

[0168] Step 916: In response to determining that the steering type corresponding to the trajectory to be run is a right turn, traverse each blind spot located on the left side of the target vehicle. In response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the left side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0169] As can be seen from this embodiment, with Figure 2 Compared with the corresponding embodiments, the flowchart 900 of the method for determining dangerous blind spots in this embodiment specifically illustrates the process of determining dangerous blind spots in intersection scenarios, further improving the accuracy of determining dangerous blind spots in intersection scenarios.

[0170] Continue to refer to Figure 10 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of a device for determining dangerous blind spots, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0171] like Figure 10 As shown, the blind spot determination device 1000 includes: an acquisition unit 1001 configured to acquire scene data representing the scene in which the target vehicle is located and obstacle data representing obstacles around the target vehicle; a construction unit 1002 configured to construct a blind spot of the target vehicle caused by the obstacles based on the obstacle data; a first determination unit 1003 configured to determine a hazard point in the scene that poses a danger to the target vehicle based on the scene data; and a second determination unit 1004 configured to determine whether the blind spot is a dangerous blind spot based on the relative relationship between the blind spot and the hazard point.

[0172] In some optional implementations of this embodiment, the construction unit 1002 is further configured to: project the obstacle point cloud data onto the ground according to the position of the point cloud generation device on the target vehicle used to generate obstacle point cloud data in obstacle data, to obtain the obstacle projection point cloud; and construct the blind spot of the target vehicle based on the obstacle projection point cloud.

[0173] In some optional implementations of this embodiment, the construction unit 1002 is further configured to: construct a minimum convex hull polygon including the obstacle projection point cloud; and construct the blind spot of the target vehicle based on the minimum convex hull polygon.

[0174] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: determine the scene type corresponding to the scene based on scene data; and determine the danger points in the scene that pose a danger to the target vehicle based on the scene type.

[0175] In some optional implementations of this embodiment, the scene type includes a pedestrian crossing scene, and the first determining unit 1003 is further configured to: determine the relative orientation relationship between the blind spot and the target vehicle in the pedestrian crossing scene; and determine the danger point in the pedestrian crossing scene based on the relative orientation relationship and the center point of the pedestrian crossing in the pedestrian crossing scene.

[0176] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: determine the ordinate of the center point as the ordinate of the danger point; based on the center point, extend forward and backward by a preset distance along the road extension direction to obtain a first extension point and a second extension point; in response to determining that the blind spot is located to the left front of the target vehicle, determine the smaller abscissa of the left boundary of the road corresponding to the first extension point and the second extension point as the abscissa of the danger point.

[0177] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: in response to determining that the blind spot is located to the right front of the target vehicle, determine the smaller abscissa of the right boundary of the road corresponding to the first extension point and the second extension point as the abscissa of the danger point.

[0178] In some optional implementations of this embodiment, the second determining unit 1004 is further configured to: traverse the blind spot located to the left front of the target vehicle in the pedestrian crossing scene, and in response to determining that there is a dangerous point located in the blind spot among the dangerous points obtained according to the horizontal coordinate of the left boundary of the road, determine that the blind spot is a dangerous blind spot.

[0179] In some optional implementations of this embodiment, the second determining unit 1004 is further configured to: traverse the blind spot located to the right front of the target vehicle in the pedestrian crossing scene, and in response to determining that there is a dangerous point located in the blind spot among the dangerous points obtained according to the horizontal coordinate of the right boundary of the road, determine that the blind spot is a dangerous blind spot.

[0180] In some optional implementations of this embodiment, the scene type includes an intersection scene, and the first determining unit 1003 is further configured to: determine the target lane segment in other lanes that intersect with the target vehicle's driving lane in the intersection scene, and the detection direction for the danger point, based on the target vehicle's trajectory to be run in the intersection scene; determine the relative orientation relationship between the target lane segment and the target vehicle based on the road nodes in the obscured portion of the target lane segment that is at least partially obscured by blind spots; and determine the danger point from the road nodes in the obscured portion of the target lane segment, wherein the obscured portion of the target lane segment is the obscured portion of the target lane segment that conforms to the relative orientation relationship of the detection direction.

[0181] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: determine the target road node closest to the target vehicle from the road nodes in the obscured portion; and determine the relative orientation relationship between the target road node and the target vehicle as the relative orientation relationship between the target lane segment and the target vehicle.

[0182] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is a left turn, determine that the detection direction is to the right; in response to determining that the detection direction is to the right, for the target occluded part in each target lane segment located to the right of the target vehicle, determine the target road node located in the target occluded part that is closest to the target vehicle as a danger point.

[0183] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is straight, determine that the detection direction is right and left; in response to determining that the detection direction is right and left, for the target occluded part in each target lane segment located to the right and left of the target vehicle, determine the target road node located in the target occluded part that is closest to the target vehicle as a danger point.

[0184] In some optional implementations of this embodiment, the first determining unit 1003 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is a right turn, determine that the detection direction is left; in response to determining that the detection direction is left, for the target occluded part in each target lane segment located to the left of the target vehicle, determine the target road node located in the target occluded part that is closest to the target vehicle as a danger point.

[0185] In some optional implementations of this embodiment, the second determining unit 1004 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is a left turn, traverse each blind spot located on the right side of the target vehicle; in response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the right side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0186] In some optional implementations of this embodiment, the second determining unit 1004 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is straight, traverse each blind spot located on the right side of the target vehicle; in response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the right side of the target vehicle, determine that the blind spot is a dangerous blind spot; and traverse each blind spot located on the left side of the target vehicle; in response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the left side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0187] In some optional implementations of this embodiment, the second determining unit 1004 is further configured to: in response to determining that the steering type corresponding to the trajectory to be run is a right turn, traverse each blind spot located on the left side of the target vehicle; in response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the left side of the target vehicle, determine that the blind spot is a dangerous blind spot.

[0188] In some optional implementations of this embodiment, the above-mentioned device further includes: a third determining unit (not shown in the figure), configured to: determine the speed limit information of the target vehicle according to the scene type corresponding to the target blind spot in at least one blind spot, wherein the target blind spot is the blind spot corresponding to the closest danger point to the target vehicle among the danger points included in each of the at least one target blind spot.

[0189] In some optional implementations of this embodiment, the above-mentioned device further includes: a fourth determining unit (not shown in the figure), configured to: determine the length of the speed-limited road section of the target vehicle based on the speed data in the obstacle data of the obstacles constituting the target blind spot.

[0190] In some optional implementations of this embodiment, the above-mentioned device further includes a control unit (not shown in the figure), configured to control the deceleration process of the target vehicle based on speed limit information, the length of the speed limit section, and a preset maximum deceleration.

[0191] In this embodiment, a device for determining dangerous blind spots is provided. For blind spots caused by surrounding obstacles, the device determines whether a blind spot is a dangerous blind spot based on the relative relationship between the dangerous points that may pose a danger to the target vehicle in the scene where the target vehicle is located and the blind spot. Non-dangerous blind spots can be filtered out, thereby improving the accuracy and effectiveness of determining dangerous blind spots.

[0192] According to embodiments of this disclosure, this disclosure also provides an electronic device, the electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to implement the method for determining dangerous blind zones described in any of the above embodiments when executed.

[0193] According to embodiments of this disclosure, this disclosure also provides a readable storage medium storing computer instructions that, when executed by a computer, enable the method for determining dangerous blind zones described in any of the above embodiments.

[0194] This disclosure provides a computer program product that, when executed by a processor, can implement the method for determining dangerous blind zones described in any of the above embodiments.

[0195] According to embodiments of this disclosure, this disclosure also provides an unmanned vehicle, including: an electronic device, the electronic device including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the method for determining dangerous blind spots described in any of the above embodiments to be implemented when the at least one processor executes them.

[0196] Figure 11 A schematic block diagram of an example electronic device 1100 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0197] like Figure 11As shown, device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1102 or a computer program loaded from storage unit 1108 into random access memory (RAM) 1103. The RAM 1103 may also store various programs and data required for the operation of device 1100. The computing unit 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. Input / output (I / O) interface 1105 is also connected to bus 1104.

[0198] Multiple components in device 1100 are connected to I / O interface 1105, including: input unit 1106, such as keyboard, mouse, etc.; output unit 1107, such as various types of monitors, speakers, etc.; storage unit 1108, such as disk, optical disk, etc.; and communication unit 1109, such as network card, modem, wireless transceiver, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0199] The computing unit 1101 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above, such as the method for determining dangerous blind zones. For example, in some embodiments, the method for determining dangerous blind zones may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1100 via ROM 1102 and / or communication unit 1109. When the computer program is loaded into RAM 1103 and executed by the computing unit 1101, one or more steps of the method for determining dangerous blind zones described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform a method for determining dangerous blind zones by any other suitable means (e.g., by means of firmware).

[0200] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0201] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0202] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0203] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0204] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0205] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, also known as cloud computing servers or cloud hosts, which are hosting products within the cloud computing service system to address the management difficulties and weak business scalability inherent in traditional physical hosts and Virtual Private Servers (VPS) services; they can also be servers for distributed systems or servers integrated with blockchain technology.

[0206] The autonomous driving module of the unmanned vehicle includes at least one processor and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed, enable the at least one processor to implement the method for determining blind spots described in embodiments 200 and 400. The autonomous driving module may also include components such as traffic light recognition cameras, front, rear, left, and right surround-view cameras, multi-line LiDAR, positioning modules (e.g., BeiDou, GPS), and inertial navigation units to perform positioning and environmental information collection to achieve autonomous driving functions.

[0207] According to the technical solution of the present disclosure, a method for determining dangerous blind spots is provided. For blind spots caused by surrounding obstacles, the method determines whether a blind spot is a dangerous blind spot based on the relative relationship between the dangerous points that may pose a danger to the target vehicle in the scene where the target vehicle is located and the blind spot. Non-dangerous blind spots can be filtered out, thereby improving the accuracy and effectiveness of determining dangerous blind spots.

[0208] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution provided in this disclosure can be achieved, and this is not limited herein.

[0209] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining a hazard blind spot, comprising: Scene data representing the scene where the target vehicle is located is determined from the high-precision map, and obstacle data representing the obstacles around the target vehicle is obtained; Based on the obstacle data, construct the blind spot of the target vehicle caused by the obstacle; Based on the scene data, determine the scene type corresponding to the scene; Based on the scenario type, identify the danger points in the scenario that pose a danger to the target vehicle. The danger points are road nodes in the scenario represented by the scenario data that cause danger to the driving process of the target vehicle. Based on the relative relationship between the blind zone and the danger point, determine whether the blind zone is a dangerous blind zone.

2. The method according to claim 1, wherein, The step of constructing the blind spot of the target vehicle caused by the obstacle based on the obstacle data includes: Based on the position of the point cloud generation device on the target vehicle used to generate obstacle point cloud data in the obstacle data, the obstacle point cloud data is projected onto the ground to obtain the obstacle projection point cloud; The blind spot of the target vehicle is constructed based on the obstacle projection point cloud.

3. The method according to claim 2, wherein, The step of constructing the blind spot of the target vehicle based on the obstacle projection point cloud includes: Construct a minimum convex hull polygon that includes the projected point cloud of the obstacle; The blind spot of the target vehicle is constructed based on the minimum convex hull polygon.

4. The method according to claim 1, wherein, The scenario types include pedestrian crossing scenarios, and The step of determining the danger points that pose a danger to the target vehicle in the scenario based on the scenario type includes: Determine the relative orientation between the blind spot in the pedestrian crossing scenario and the target vehicle; Based on the relative orientation and the center point of the pedestrian crossing in the pedestrian crossing scenario, the danger points in the pedestrian crossing scenario are determined.

5. The method according to claim 4, wherein, The step of determining the danger points in the pedestrian crossing scene based on the relative orientation relationship and the center point of the pedestrian crossing scene includes: The ordinate of the center point is determined as the ordinate of the danger point; Based on the center point, extend forward and backward by a preset distance along the road extension direction to obtain the first extension point and the second extension point; In response to determining that the blind spot is located to the left front of the target vehicle, the smaller of the abscissas of the left boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

6. The method according to claim 5, wherein, The step of determining the danger points in the pedestrian crossing scene based on the relative orientation relationship and the center point of the pedestrian crossing scene further includes: In response to determining that the blind spot is located to the right front of the target vehicle, the smaller of the abscissas of the right boundary of the road corresponding to the first extension point and the second extension point is determined as the abscissa of the danger point.

7. The method according to claim 5, wherein, The step of determining whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point includes: Traverse the blind spot located to the left front of the target vehicle in the pedestrian crossing scenario, and in response to determine that there is a dangerous point located in the blind spot among the dangerous points obtained from the horizontal coordinate of the left boundary of the road, determine that the blind spot is a dangerous blind spot.

8. The method according to claim 6, wherein, The step of determining whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point includes: Traverse the blind spot located to the right front of the target vehicle in the pedestrian crossing scenario, and in response to determining that there is a dangerous point located within the blind spot among the dangerous points obtained based on the horizontal coordinate of the right boundary of the road, determine that the blind spot is a dangerous blind spot.

9. The method according to claim 1, wherein, The scenario types include intersection scenarios, and The step of determining the danger points that pose a danger to the target vehicle in the scenario based on the scenario type includes: Based on the trajectory of the target vehicle in the intersection scenario, determine the target lane segment in other lanes that intersect with the target vehicle's driving lane in the intersection scenario, and the detection direction for the danger point; The relative orientation relationship between the target lane segment and the target vehicle is determined based on the road nodes in the obscured portion of the target lane segment that is at least partially obscured by the blind spot. The danger point is determined from the road nodes in the obscured portion of the target, wherein the obscured portion of the target is the obscured portion in the target lane segment corresponding to the relative orientation relationship that conforms to the detection direction.

10. The method according to claim 9, wherein, Determining the relative orientation between the target lane segment and the target vehicle based on road nodes in the obscured portion of the target lane segment that is at least partially obscured by the blind spot includes: From the road nodes in the obscured portion, determine the target road node that is closest to the target vehicle; The relative orientation relationship between the target road node and the target vehicle is determined as the relative orientation relationship between the target lane segment and the target vehicle.

11. The method according to claim 9, wherein, The step of determining the detection direction for the danger point based on the trajectory of the target vehicle in the intersection scenario includes: In response to determining that the steering type corresponding to the trajectory to be run is a left turn, the detection direction is determined to be to the right; and The process of determining the danger point from road nodes in the obscured portion of the target includes: In response to determining that the detection direction is to the right, for the obscured portion of the target in each target lane segment located to the right of the target vehicle, the target road node located in the obscured portion that is closest to the target vehicle is identified as the danger point.

12. The method according to claim 9, wherein, The step of determining the detection direction for the danger point based on the trajectory of the target vehicle in the intersection scenario includes: In response to determining that the steering type corresponding to the trajectory to be run is straight, the detection direction is determined to be right and left; and The process of determining the danger point from road nodes in the obscured portion of the target includes: In response to determining that the detection direction is right and left, for the obscured portion of the target in each target lane segment located to the right and left of the target vehicle, the target road node located in the obscured portion that is closest to the target vehicle is identified as the danger point.

13. The method according to claim 9, wherein, The step of determining the detection direction for the danger point based on the trajectory of the target vehicle in the intersection scenario includes: In response to determining that the steering type corresponding to the trajectory to be run is a right turn, the detection direction is determined to be left; and The process of determining the danger point from road nodes in the obscured portion of the target includes: In response to determining that the detection direction is to the left, for the obscured portion of the target in each target lane segment located to the left of the target vehicle, the target road node located in the obscured portion that is closest to the target vehicle is identified as the danger point.

14. The method according to claim 11, wherein, The step of determining whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point includes: In response to determining that the steering type corresponding to the trajectory to be run is a left turn, each blind spot located on the right side of the target vehicle is traversed. In response to determining that there is a dangerous point located in the blind spot among the dangerous points located on the right side of the target vehicle, the blind spot is determined to be a dangerous blind spot.

15. The method according to claim 12, wherein, The step of determining whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point includes: In response to determining that the steering type corresponding to the trajectory to be run is straight, each blind spot located on the right side of the target vehicle is traversed. In response to determining that there is a dangerous point located within the blind spot among the dangerous points located on the right side of the target vehicle, the blind spot is determined to be a dangerous blind spot. Then, each blind spot located on the left side of the target vehicle is traversed. In response to determining that there is a dangerous point located within the blind spot among the dangerous points located on the left side of the target vehicle, the blind spot is determined to be a dangerous blind spot.

16. The method according to claim 13, wherein, The step of determining whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point includes: In response to determining that the steering type corresponding to the trajectory to be run is a right turn, each blind spot located on the left side of the target vehicle is traversed. In response to determining that there is a dangerous point located in the blind spot among the dangerous points on the left side of the target vehicle, the blind spot is determined to be a dangerous blind spot.

17. The method according to claim 1, wherein, Also includes: Based on the scene type corresponding to the target blind spot in at least one blind spot, the speed limit information of the target vehicle is determined, wherein the target blind spot is the blind spot corresponding to the hazard point closest to the target vehicle among the hazard points included in each of the at least one target blind spot.

18. The method according to claim 17, wherein, Also includes: Based on the speed data in the obstacle data of the obstacles that constitute the target blind spot, the length of the speed-limited road section for the target vehicle is determined.

19. The method according to claim 18, wherein, Also includes: The deceleration process of the target vehicle is controlled based on the speed limit information, the length of the speed limit section, and the preset maximum deceleration.

20. A device for determining a hazard blind spot, comprising: The acquisition unit is configured to determine scene data representing the scene where the target vehicle is located from the high-precision map, and acquire obstacle data representing the obstacles around the target vehicle. The construction unit is configured to construct the blind spot of the target vehicle caused by the obstacle based on the obstacle data; The first determining unit is configured to determine the scene type corresponding to the scene based on the scene data; Based on the scenario type, identify the danger points in the scenario that pose a danger to the target vehicle. The danger points are road nodes in the scenario represented by the scenario data that cause danger to the driving process of the target vehicle. The second determining unit is configured to determine whether the blind zone is a dangerous blind zone based on the relative relationship between the blind zone and the danger point.

21. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-19.

22. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-19.

23. A computer program product, comprising: A computer program that, when executed by a processor, implements the method according to any one of claims 1-19.

24. An unmanned vehicle, comprising: The electronic device as claimed in claim 21.

Images