Vehicle control method, device, vehicle, and program product

Abstract

The present disclosure relates to a vehicle control method, device, vehicle and program product, and relates to the technical field of intelligent driving of automobiles. The method specifically comprises: establishing a current obstacle map of a vehicle, wherein the obstacle map is obtained according to map information where the vehicle is located and a perception result of the vehicle; determining a visual field blind area of a vehicle sensor according to the obstacle map and a perception observation point of the vehicle, wherein the perception observation point is an installation position of the vehicle sensor on the vehicle; and performing deceleration control on the vehicle in response to the vehicle pre-arriving at an intersection being located in the visual field blind area. Thus, by determining the visual field blind area of the vehicle sensor, the vehicle can be decelerated in advance in response to the vehicle pre-arriving at the intersection being located in the visual field blind area, so as to reduce the risk of ghost probe and improve the safety during the driving of the vehicle.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent driving technology, and in particular to a vehicle control method, device, vehicle, and program product. Background Technology

[0002] Vehicles often encounter situations like pedestrians suddenly appearing at intersections while driving. If the vehicle is traveling too fast, there may be safety risks. Related technologies often involve slowing down in advance at intersections, but this has problems such as slow speed when passing through intersections, poor experience, and a heavy mechanical feel when slowing down at intersections. Summary of the Invention

[0003] This disclosure provides a vehicle control method, apparatus, vehicle, computer-readable storage medium, computer program product, and chip.

[0004] The technical solution disclosed herein is as follows:

[0005] According to a first aspect of the present disclosure, a vehicle control method is provided, the method comprising: establishing a current obstacle map of the vehicle, wherein the obstacle map is obtained based on map information of the vehicle's location and the vehicle's perception results; determining a blind spot of a vehicle sensor based on the obstacle map and the vehicle's perception observation points, wherein the perception observation points are the installation locations of the vehicle sensors on the vehicle; and performing deceleration control on the vehicle in response to the vehicle's expected arrival at an intersection being within the blind spot.

[0006] In the above embodiments, by establishing an obstacle map of the vehicle's current location, and based on the obstacle map and the vehicle's perception observation points, the blind spots of the vehicle's sensors are determined. The perception observation points are the installation locations of the vehicle sensors on the vehicle. In response to the vehicle's expected arrival at an intersection being within the blind spot, the vehicle is decelerated. Thus, by determining the blind spots of the vehicle sensors, this disclosure can decelerate the vehicle in advance in response to the vehicle's expected arrival at an intersection being within the blind spot, reducing the risk of vehicles suddenly appearing out of sight and improving the safety of the vehicle during driving.

[0007] Optionally, in some embodiments of this disclosure, determining the blind spot based on the obstacle map and the vehicle's perception observation points includes: filtering obstacles based on semantic information of obstacles in the obstacle map to obtain target obstacles that pose a risk of visual obstruction; and determining the blind spot based on the perception observation points and the target obstacles.

[0008] Optionally, in some embodiments of this disclosure, determining the visual blind spot based on the sensing observation point and the target obstacle includes: determining the type of the target obstacle based on the semantic information of the target obstacle; determining the visual occlusion point pair of the target obstacle based on the type of the target obstacle; and obtaining the visual blind spot based on the sensing observation point and the visual occlusion point pair.

[0009] In the above embodiments, the type of the target obstacle is determined based on the semantic information of the target obstacle; the visual field occlusion point pairs of the target obstacle are determined based on the type of the target obstacle; and the visual field blind spot is obtained based on the perception observation point and the visual field occlusion point pairs. Thus, this disclosure can adopt different visual field blind spot determination methods for different types of target obstacles, which improves the flexibility and accuracy of determining the visual field blind spot and lays the foundation for accurate vehicle control in the future.

[0010] Optionally, in some embodiments of this disclosure, determining the visual obstruction point pair of the target obstacle according to the type of the target obstacle includes: in response to the target obstacle being a first type of obstacle, determining a first edge point and a tail edge point from the set of edge points of the target obstacle; identifying the visual obstruction point pair of the target obstacle from the set of edge points based on the first edge point and the tail edge point; wherein the first edge point and the tail edge point are determined based on the traversal direction of the set of edge points.

[0011] Optionally, in some embodiments of this disclosure, identifying visual obstruction point pairs from the set of edge points based on the first edge point and the last edge point includes: generating a first-to-last connecting line based on the first edge point and the last edge point; traversing and filtering the edge points in the set of edge points to obtain the distance between the currently traversed edge point and the first-to-last connecting line; determining the currently traversed edge point as a visual obstruction point in response to the distance being greater than or equal to a first preset distance threshold; and determining two adjacent visual obstruction points as a visual obstruction point pair in response to the end of traversal.

[0012] Optionally, in some embodiments of this disclosure, after determining that the currently traversed edge point is a visual obstruction point, the method further includes: generating a new set of edge points from the currently traversed edge point and the remaining untraversed edge points, wherein the currently traversed edge point is the first edge point of the new set of edge points; and continuing to identify the next visual obstruction point pair of the target obstacle based on the first edge point and the last edge point in the new set of edge points, until the traversal ends.

[0013] In the above embodiments, in response to the target obstacle being a first type of obstacle, a first edge point and a last edge point are determined from the set of edge points of the target obstacle. Based on the first edge point and the last edge point, visual obstruction point pairs of the target obstacle are identified from the set of edge points. The first edge point and the last edge point are determined based on the traversal direction of the set of edge points. Thus, for the first type of obstacle, this disclosure can identify visual obstruction point pairs of the target obstacle from the set of edge points based on the first edge point and the last edge point of the obstacle, thereby improving the efficiency and adaptability of determining visual obstruction point pairs of the target obstacle and enhancing the robustness of environmental perception.

[0014] Optionally, in some embodiments of this disclosure, determining the visual obstruction point pair of the target obstacle according to the type of the target obstacle includes: in response to the target obstacle being a second type of obstacle, determining the corner point of the target obstacle, wherein the corner point of the target obstacle is calculated by a preset target detection algorithm; and determining the visual obstruction point pair from the corner point based on the perception observation point.

[0015] Optionally, in some embodiments of this disclosure, determining the visual field occlusion point pair from the corner points based on the sensing observation point includes: connecting the corner point and the sensing observation point to generate a line connecting the corner point and the sensing observation point; determining the angle between the line connecting the corner point and the sensing observation point and a preset reference line; and selecting the intersection point corresponding to the largest angle and the corner point corresponding to the smallest angle as the visual field occlusion point pair.

[0016] In the above embodiments, in response to the target obstacle being a second type of obstacle, the corner points of the target obstacle are determined, and based on the sensing observation points, the visual field occlusion point pairs are determined from the corner points. Thus, for the second type of obstacle, this disclosure can dynamically determine the visual field occlusion point pairs from the corner points based on the sensing observation points, ensuring the accuracy of determining the visual field occlusion point pairs of the target obstacle.

[0017] Optionally, in some embodiments of this disclosure, obtaining the visual blind zone based on the sensing observation point and the visual field occlusion point pair includes: generating a first blind zone boundary line based on a first visual field occlusion point in the sensing observation point and the visual field occlusion point pair; generating a second blind zone boundary line based on a second visual field occlusion point in the sensing observation point and the visual field occlusion point pair; and determining the visual blind zone based on the first blind zone boundary line and the second blind zone boundary line of the visual field occlusion point pair.

[0018] Optionally, in some embodiments of this disclosure, the number of visual field occlusion point pairs is multiple, and determining the visual field blind zone based on the first blind zone boundary line and the second blind zone boundary line of the visual field occlusion point pairs includes: determining a corresponding candidate visual field blind zone based on the first blind zone boundary line and the second blind zone boundary line of each visual field occlusion point pair; and combining multiple candidate visual field blind zones to obtain the visual field blind zone.

[0019] Optionally, in some embodiments of this disclosure, when the vehicle is about to arrive at an intersection within the blind spot, deceleration control of the vehicle includes: in response to the vehicle being about to arrive at an intersection within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being less than a second preset distance threshold, deceleration control of the vehicle is performed.

[0020] Optionally, in some embodiments of this disclosure, when the vehicle is about to arrive at the intersection and is within the blind spot, the deceleration control of the vehicle further includes: monitoring the distance between the vehicle and the center point of the intersection; and decelerating the vehicle in response to the distance between the vehicle and the center point of the intersection being less than or equal to a set distance threshold.

[0021] Optionally, in some embodiments of this disclosure, the method further includes at least one of the following operations: in response to the vehicle's expected arrival at an intersection not being within the blind spot, or in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second preset distance threshold, sending a speed-maintaining control signal to the vehicle's controller.

[0022] In the above embodiments, in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first or second blind spot boundary line is less than a second preset distance threshold, the vehicle can be slowed down in advance to reduce the risk of a vehicle suddenly appearing out of sight, improve the vehicle's safety factor, and ensure the safety of the vehicle during driving. Furthermore, in response to the expected arrival at an intersection not being within the blind spot, or in response to the expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first or second blind spot boundary line is greater than or equal to the second preset distance threshold, the vehicle does not need to slow down, which helps to improve the vehicle's average cruising speed. The vehicle can be accelerated / decelerated according to different actual situations, avoiding the problem of a heavy mechanical feel and improving the user experience.

[0023] According to a second aspect of the present disclosure, a vehicle control device is provided, the device comprising: a first determining module, configured to determine a current obstacle map of the vehicle, wherein the obstacle map is obtained based on map information of the vehicle and the vehicle's perception results; a second determining module, configured to determine a blind spot of a vehicle sensor based on the obstacle map and the vehicle's perception observation points, wherein the perception observation points are the installation positions of the vehicle sensors on the vehicle; and a control module, configured to decelerate the vehicle in response to the vehicle's expected arrival at an intersection being within the blind spot.

[0024] Optionally, in some embodiments of this disclosure, the second determining module is configured to: filter the obstacles based on the semantic information of the obstacles in the obstacle map to obtain target obstacles that pose a risk of visual obstruction; and determine the blind spot based on the perception observation point and the target obstacle.

[0025] Optionally, in some embodiments of this disclosure, the second determining module is configured to: in response to the target obstacle being a first type of obstacle, determine a first edge point and a tail edge point from the set of edge points of the target obstacle; and identify a pair of visual obstruction points of the target obstacle from the set of edge points based on the first edge point and the tail edge point; wherein the first edge point and the tail edge point are determined based on the traversal direction of the set of edge points.

[0026] Optionally, in some embodiments of this disclosure, the second determining module is configured to: determine the corner points of the target obstacle in response to the target obstacle being a second type of obstacle; wherein the corner points of the target obstacle are calculated by a preset target detection algorithm; and determine the visual obstruction point pair from the corner points based on the perception observation point.

[0027] According to a third aspect of the present disclosure, a vehicle is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the vehicle control method provided as in the first aspect of the present disclosure.

[0028] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform a vehicle control method as provided in the first aspect of the present disclosure.

[0029] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the vehicle control method provided in the first aspect of the present disclosure.

[0030] According to a sixth aspect of the present disclosure, a chip is provided, the chip including an interface circuit and a processing circuit coupled to each other, the interface circuit being used to input or output signals, and the processing circuit being configured to implement the steps of the vehicle control method provided in the first aspect of the present disclosure.

[0031] The technical solutions provided by the embodiments of this disclosure bring at least the following beneficial effects:

[0032] An embodiment of the present disclosure of a vehicle control method establishes a current obstacle map for the vehicle and determines the blind spot of the vehicle's sensors based on the obstacle map and the vehicle's perception observation points. The perception observation points are the installation positions of the vehicle sensors on the vehicle. In response to the vehicle's expected arrival at an intersection being within the blind spot, the method performs deceleration control on the vehicle. Thus, by determining the blind spot of the vehicle's sensors, the present disclosure can perform deceleration control on the vehicle in advance in response to the vehicle's expected arrival at an intersection being within the blind spot, reducing the risk of vehicles suddenly appearing out of sight and improving the safety of the vehicle during driving.

[0033] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0034] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0035] Figure 1 This is a schematic flowchart illustrating a vehicle control method according to an exemplary embodiment.

[0036] Figure 2 This is a flowchart illustrating another vehicle control method according to an exemplary embodiment.

[0037] Figure 3 This is a flowchart illustrating another vehicle control method according to an exemplary embodiment.

[0038] Figure 4(a) is a schematic diagram of a pair of view occlusion points according to an exemplary embodiment.

[0039] Figure 4(b) is a schematic diagram illustrating another pair of field-of-view occlusion points according to an exemplary embodiment.

[0040] Figure 5 This is a flowchart illustrating another vehicle control method according to an exemplary embodiment.

[0041] Figure 6 This is a schematic diagram illustrating another pair of viewpoint occlusion points according to an exemplary embodiment.

[0042] Figure 7(a) is a schematic diagram of a blind spot according to an exemplary embodiment.

[0043] Figure 7(b) is a schematic diagram illustrating another blind spot according to an exemplary embodiment.

[0044] Figure 8 This is a flowchart illustrating another vehicle control method according to an exemplary embodiment.

[0045] Figure 9 This is a schematic flowchart illustrating a vehicle control method according to an exemplary embodiment.

[0046] Figure 10 This is a block diagram illustrating a vehicle control device according to an exemplary embodiment.

[0047] Figure 11 This is a block diagram illustrating a vehicle according to an exemplary embodiment.

[0048] Figure 12 This is a block diagram illustrating a chip according to an exemplary embodiment.

[0049] Figure 13 This is a block diagram illustrating another chip according to an exemplary embodiment. Detailed Implementation

[0050] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0051] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0052] Figure 1 This is a schematic flowchart of a vehicle control method provided in an embodiment of the present disclosure.

[0053] like Figure 1 As shown, the vehicle control method includes the following steps:

[0054] S101, Establish the current obstacle map of the vehicle, wherein the obstacle map is obtained based on the map information of the vehicle's location and the vehicle's perception results.

[0055] It should be noted that the map information of the vehicle's location refers to the map information of the area where the vehicle is driving or parked. As an optional method, the map information can be obtained from navigation software.

[0056] In this embodiment of the disclosure, the vehicle's perception results and map information of the vehicle's location can be obtained, and an obstacle map can be determined based on the map information of the vehicle's location and the vehicle's perception results.

[0057] Optionally, real-time data of the vehicle's surrounding environment can be collected using vehicle sensors to obtain the vehicle's perception results. For example, real-time data of the vehicle's surrounding environment can be collected using vehicle sensors such as cameras, lidar, and ultrasonic radar, and the collected real-time data of the vehicle's surrounding environment can be fused to obtain the vehicle's perception results.

[0058] In this embodiment of the disclosure, after obtaining the perception results of the vehicle, obstacles can be identified based on the perception results of the vehicle, and obstacles can be marked on the map information where the vehicle is located to establish the current obstacle map of the vehicle.

[0059] The obstacle map can be used to describe dynamic and static obstacles in the environment in real time.

[0060] S102, Based on the obstacle map and the vehicle's perception observation points, determine the blind spots of the vehicle's sensors, where the perception observation points are the installation locations of the vehicle sensors on the vehicle.

[0061] Among them, the vehicle's perception observation point (the installation position of the sensor on the vehicle) can be used to determine the field of view and blind spot of the vehicle sensor. The perception observation point is the fixed installation position of the vehicle sensor on the vehicle, which is usually represented by three-dimensional coordinates.

[0062] In this embodiment of the disclosure, obstacles can be filtered based on the semantic information of obstacles in the obstacle map to obtain target obstacles that pose a risk of visual obstruction, and blind spots in the field of vision can be determined based on the perception observation point and the target obstacle.

[0063] S103 responds to the vehicle's impending arrival at an intersection being within the blind spot by slowing down the vehicle.

[0064] In this embodiment of the disclosure, the distance between the vehicle's position and the center point of the intersection can be obtained. If the distance between the vehicle's position and the center point of the intersection is less than or equal to a preset distance threshold, and there is a blind spot at the intersection to be reached, then it is determined that the intersection to be reached by the vehicle is within the blind spot.

[0065] It should be noted that this disclosure does not limit the setting of the distance threshold; for example, the distance threshold can be set to 15m.

[0066] For example, if the distance threshold is 15m, and the distance between the vehicle's position and the center point of the intersection to be reached is 10m, that is, the distance between the vehicle's position and the center point of the intersection to be reached (10m) is less than the distance threshold (15m), and there is a blind spot at the intersection to be reached, then it is determined that the intersection to be reached by the vehicle is within the blind spot.

[0067] In this embodiment of the disclosure, in response to the vehicle being in the blind spot of the intersection it is about to reach, a slow deceleration control signal is sent to the vehicle controller to decelerate the vehicle and achieve smooth deceleration.

[0068] Optionally, by controlling the vehicle's deceleration, after the vehicle passes through the intersection, a smooth acceleration control signal can be sent to the vehicle controller to accelerate the vehicle smoothly until the vehicle reaches the cruising speed.

[0069] An embodiment of the present disclosure provides a vehicle control method that establishes a current obstacle map for the vehicle. The obstacle map is obtained based on map information of the vehicle's location and the vehicle's perception results. Based on the obstacle map and the vehicle's perception observation points, the blind spots of the vehicle's sensors are determined. The perception observation points are the installation locations of the vehicle sensors on the vehicle. In response to the vehicle's expected arrival at an intersection being within the blind spot, the vehicle is decelerated. Thus, by determining the blind spots of the vehicle's sensors, the present disclosure can decelerate the vehicle in advance in response to the vehicle's expected arrival at an intersection being within the blind spot, reducing the risk of vehicles suddenly appearing out of sight and improving the safety of the vehicle during driving.

[0070] Figure 2 This is a schematic flowchart of a vehicle control method according to an embodiment of the present disclosure. Based on the above embodiment, it further incorporates... Figure 2 The process of determining blind spots based on obstacle maps and vehicle perception observation points is explained, including the following steps:

[0071] S201, based on the semantic information of obstacles in the obstacle map, the obstacles are filtered to obtain target obstacles that pose a risk of obstructing the line of sight.

[0072] Optionally, semantic information of obstacles and corresponding line-of-sight occlusion risk labels can be obtained. An obstacle mapping table can be constructed based on the semantic information of obstacles and corresponding line-of-sight occlusion risk labels. The pre-constructed obstacle mapping table can be queried based on the semantic information of obstacles in the obstacle map to filter obstacles and obtain target obstacles with line-of-sight occlusion risk.

[0073] For example, if the semantic information of obstacle 1 is a wall, the semantic information of obstacle 2 is a pillar, and the semantic information of obstacle 3 is a car, by querying the pre-built obstacle mapping table, it is determined that the line-of-sight occlusion risk labels corresponding to the semantic information of obstacle 1, obstacle 2, and obstacle 3 are all line-of-sight occlusion risk objects. Therefore, obstacle 1, obstacle 2, and obstacle 3 are all target obstacles with line-of-sight occlusion risk objects.

[0074] In this embodiment of the disclosure, in order to improve the accuracy of determining target obstacles, after obtaining target obstacles that pose a risk of obstructing the line of sight, the screened target obstacles can be verified based on the size information of the obstacles, so as to retain the target obstacles that pass the verification.

[0075] Among them, the target obstacles that pass the verification can be understood as obstacles that will cause blind spots in the field of vision, while the target obstacles that fail the verification can be understood as obstacles that will not cause blind spots in the field of vision.

[0076] Optionally, preset obstacle size threshold information can be obtained, and the selected target obstacles can be verified based on the obstacle size threshold information and obstacle size information to retain the target obstacles that pass the verification.

[0077] For example, if the obstacle is a vehicle, and the preset vehicle height threshold is 1.5m, four-wheeled vehicles such as cars, trucks, and commercial vehicles with a height greater than 1.5m can be selected as target obstacles that pass the verification, while non-motorized vehicles such as bicycles and electric vehicles with a height less than or equal to 1.5m are considered target obstacles that fail the verification.

[0078] For example, if obstacle 3 is a car and obstacle 4 is a bicycle, and the preset vehicle height threshold is 1.5m, then the height information of obstacle 3 is greater than 1.5m and the height information of obstacle 4 is less than 1.5m. Therefore, obstacle 3 passes the verification and obstacle 4 fails the verification.

[0079] S202, determine the blind spot based on the sensing observation point and the target obstacle.

[0080] In this embodiment of the disclosure, the type of the target obstacle can be determined based on the semantic information of the target obstacle, the visual field occlusion point pair of the target obstacle can be determined based on the type of the target obstacle, and the visual field blind spot can be obtained based on the perception observation point and the visual field occlusion point pair.

[0081] The target obstacles include static and dynamic types. Static obstacles can be understood as immovable obstacles, such as walls, pillars, and curbs. Dynamic obstacles can be understood as movable obstacles, such as pedestrians, bicycles, electric vehicles, and cars.

[0082] As one possible implementation, such as Figure 3 As shown, based on the above embodiments, the specific process of determining the visual obstruction point pairs of the target obstacle according to the type of the target obstacle includes the following steps:

[0083] S301, in response to the target obstacle being a first type of obstacle, determine the first edge point and the last edge point from the set of edge points of the target obstacle.

[0084] Among them, the first type of obstacle can be a static type of obstacle.

[0085] The set of edge points of the target obstacle can be determined by the outline information of the target obstacle.

[0086] Optionally, the traversal direction of the edge point set can be obtained, and the first edge point and the last edge point can be determined from the edge point set of the target obstacle based on the traversal direction.

[0087] For example, in the traversal direction of the edge point set, the first edge point in the target obstacle's edge point set can be taken as the first edge point, and the last edge point in the target obstacle's edge point set can be taken as the last edge point.

[0088] S302, based on the first edge point and the last edge point, identify the visual obstruction point pairs of the target obstacle from the set of edge points, wherein the first edge point and the last edge point are determined based on the traversal direction of the set of edge points.

[0089] In this embodiment of the disclosure, a first-to-last connecting line can be generated based on the first edge point and the last edge point. The edge points in the edge point set are traversed and filtered to obtain the distance between the currently traversed edge point and the first-to-last connecting line. In response to the distance being greater than or equal to a first set distance threshold, the currently traversed edge point is determined to be a visual field occlusion point. In response to the end of traversal, two adjacent visual field occlusion points are determined to be a visual field occlusion point pair.

[0090] It should be noted that this disclosure does not limit the setting of the first set distance threshold. For example, the first set distance threshold can be set to 0.15m.

[0091] The first pair of visual obstruction points includes the first edge point and the first edge point traversed at a distance greater than or equal to a first set distance threshold.

[0092] For example, as shown in Figure 4(a), the target obstacle is an arc-shaped wall. If the first set threshold is 0.15m, for the first edge point 1 and the last edge point n, the line L1 connecting the first edge point 1 and the last edge point n is used as the first and last connecting line. By traversing and filtering the edge points in the edge point set (i.e. the edge points between the first edge point 1 and the last edge point n, and including the first edge point 1 and the last edge point n), the first edge point with a distance greater than 0.15m is edge point m. Then edge point m is a visual obstruction point, and the first visual obstruction point pair is the first edge point 1 and edge point m.

[0093] In this embodiment of the disclosure, after determining that the currently traversed edge point is a visual obstruction point, the method further includes: generating a new set of edge points from the currently traversed edge point and the remaining untraversed edge points, wherein the currently traversed edge point is the first edge point of the new set of edge points, and based on the first edge point and the last edge point in the new set of edge points, continuing to identify the next visual obstruction point pair of the target obstacle until the traversal ends.

[0094] For example, as shown in Figure 4(b), if the current traversed edge point is edge point m, a new set of edge points is generated based on edge point m and the remaining untraversed edge points (i.e., edge points between edge point m and the tail edge point n, including the tail edge point n). Edge point m is the first edge point of the new set of edge points. The line L2 connecting the first edge point m and the tail edge point n is used as the first-to-last connection line. The edge points in the set of edge points are traversed and filtered. In response to the distance between the current traversed edge point x and the first-to-last connection line being greater than 0.15m, the current traversed edge point x is determined to be a visual occlusion point. The above steps are repeated until the traversal ends. After the traversal ends, the tail edge point n has been traversed. Two adjacent visual occlusion points (edge ​​point m and edge point x) are determined to be a visual occlusion point pair.

[0095] In this embodiment of the disclosure, in response to the target obstacle being a first type of obstacle, a first edge point and a last edge point are determined from the set of edge points of the target obstacle. Based on the first edge point and the last edge point, visual obstruction point pairs of the target obstacle are identified from the set of edge points. The first edge point and the last edge point are determined based on the traversal direction of the set of edge points. Thus, for the first type of obstacle, this disclosure can identify visual obstruction point pairs of the target obstacle from the set of edge points based on the first edge point and the last edge point of the obstacle, thereby improving the efficiency and adaptability of determining visual obstruction point pairs of the target obstacle and enhancing the robustness of environmental perception.

[0096] As one possible implementation, such as Figure 5 As shown, based on the above embodiments, the specific process of determining the visual obstruction point pairs of the target obstacle according to the type of the target obstacle includes the following steps:

[0097] S501, in response to the target obstacle being a second type of obstacle, determine the corner points of the target obstacle, wherein the corner points of the target obstacle are calculated by a preset target detection algorithm.

[0098] The second type of obstacle can be a dynamic type of obstacle.

[0099] For example, if the target obstacle is a car, the vertices of the car's 3D detection box can be calculated using a preset target detection algorithm. The 3D detection box is then rotated and translated to obtain the updated vertices. The updated vertices of the 3D detection box are then projected onto the ground plane. Four corner points are selected from the multiple vertices projected onto the ground plane as the corner points of the car to describe the car's outline on the ground.

[0100] S502, based on the sensing observation points, determines the pairs of occlusion points from the corner points.

[0101] In this embodiment, corner points and sensing observation points can be connected to generate a line connecting the corner points and sensing observation points. The angle between the line connecting the corner points and sensing observation points and a preset reference line is determined. The corner points corresponding to the largest and smallest angles are selected as pairs of visual obstruction points. The preset reference line refers to the lateral extension line drawn from the sensing observation point to both sides in the vehicle width direction in the vehicle coordinate system.

[0102] For example, such as Figure 6 As shown, if the target obstacle is a car, for the sensing observation point A and the four corner points of the car (corner point 1, corner point 2, corner point 3, and corner point 4), connect corner point 1 to sensing observation point A to generate line L3 between corner point 1 and sensing observation point A; connect corner point 2 to sensing observation point A to generate line L4 between corner point 2 and sensing observation point A; connect corner point 3 to sensing observation point A to generate line L5 between corner point 3 and sensing observation point A; and connect corner point 4 to sensing observation point A to generate line L6 between corner point 4 and sensing observation point A. By comparing lines L3, L4, L5, and L6 with a preset reference line (…),… Figure 6 The included angles of the marked angles are compared, and the corner point corresponding to the largest included angle is determined as corner point 1 and the corner point corresponding to the smallest included angle is determined as corner point 4. Corner point 1 and corner point 4 are used as the visual field occlusion point pair.

[0103] In this embodiment of the disclosure, in response to the target obstacle being a second type of obstacle, the corner points of the target obstacle are determined. The corner points of the target obstacle are calculated by a preset target detection algorithm. Based on the perception observation points, visual obstruction point pairs are determined from the corner points. Thus, for the second type of obstacle, this disclosure can dynamically determine visual obstruction point pairs from the corner points based on the perception observation points, ensuring the accuracy of determining the visual obstruction point pairs of the target obstacle.

[0104] In this embodiment of the present disclosure, a first blind zone boundary line can be generated based on the first visual field occlusion point in the pair of sensing observation points and visual field occlusion points, and a second blind zone boundary line can be generated based on the second visual field occlusion point in the pair of sensing observation points and visual field occlusion points. The visual field blind zone is determined based on the first blind zone boundary line and the second blind zone boundary line of the pair of visual field occlusion points.

[0105] In this embodiment of the disclosure, when there are multiple pairs of visual field occlusion points, a corresponding candidate visual field blind zone is determined based on the first blind zone boundary line and the second blind zone boundary line of each pair of visual field occlusion points. Multiple candidate visual field blind zones are combined to obtain a visual field blind zone.

[0106] For example, as shown in Figure 7(a), for the perception observation point A and the first visual field occlusion point pair, a first blind zone boundary line S1 is generated based on the first visual field occlusion point (first edge point 1) in the perception observation point A and the visual field occlusion point pair. A second blind zone boundary line S2 is generated based on the second visual field occlusion point (edge ​​point m) in the perception observation point A and the visual field occlusion point pair. Based on the first blind zone boundary line S1 and the second blind zone boundary line S2, the visual field blind zone is determined. That is, the visual field blind zone includes the area occluded by the target obstacle and located at the first blind zone boundary line S1 and the second blind zone boundary line S2. The area between the boundary lines S2 and S3; for another pair of visual field obstruction points, a first blind zone boundary line S1 is generated based on the first visual field obstruction point (edge ​​point m) in the pair of perception observation point A and visual field obstruction points, and a second blind zone boundary line S2 is generated based on the second visual field obstruction point (edge ​​point x) in the pair of perception observation point A and visual field obstruction points. Based on the first blind zone boundary line S1 and the second blind zone boundary line S2, the visual field blind zone is determined, that is, the visual field blind zone includes the area obstructed by the target obstacle and located between the first blind zone boundary line S1 and the second blind zone boundary line S3.

[0107] For example, as shown in Figure 7(b), for a perception observation point A, a first blind zone boundary line S1 is generated based on the first visual field occlusion point (corner point 1) in the pair between the perception observation point A and the visual field occlusion point. A second blind zone boundary line S2 is generated based on the second visual field occlusion point (corner point 4) in the pair between the perception observation point A and the visual field occlusion point. Based on the first blind zone boundary line S1 and the second blind zone boundary line S2, the visual field blind zone is determined. That is, the visual field blind zone is the area that is occluded by the target obstacle and is located between the first blind zone boundary line S1 and the second blind zone boundary line S2.

[0108] A vehicle control method according to an embodiment of this disclosure filters obstacles based on semantic information of obstacles in an obstacle map to obtain target obstacles that pose a risk of visual obstruction. Based on the perception observation point and the target obstacle, a blind spot is determined. Thus, this disclosure can adopt different blind spot determination methods for static and dynamic target obstacles, improving the flexibility and accuracy of blind spot determination and laying the foundation for accurate vehicle control in the future.

[0109] Figure 8 This is a schematic flowchart illustrating a vehicle control method provided in an embodiment of this disclosure. The reference numerals in the flowchart do not represent any timing limitations.

[0110] like Figure 8 As shown, the vehicle control method includes the following steps:

[0111] S801, acquires the perception results of the vehicle.

[0112] Optionally, real-time data of the vehicle's surrounding environment can be collected collaboratively using vehicle sensors, and the vehicle's perception results can be obtained by summarizing the real-time data.

[0113] For example, real-time data such as image data and video stream data of the vehicle's surrounding environment can be collected through vehicle sensors, as well as real-time data such as the vehicle's own position, speed, and heading information. By summarizing the real-time data, the vehicle's perception results can be obtained.

[0114] S802, based on the map information where the vehicle is located and the vehicle's perception results, establishes the current obstacle map for the vehicle.

[0115] In this embodiment of the disclosure, map information can be downloaded from the server based on the vehicle's current location information, map information can be read from the cache, and if map information is not read from the cache, map information can be downloaded from the server based on the vehicle's current location information.

[0116] In this embodiment of the disclosure, the type and outline information of obstacles can be determined based on the vehicle's perception results. Based on the type and corresponding outline information of the obstacles, the obstacles are marked on the map information where the vehicle is located, and a current obstacle map of the vehicle is established.

[0117] The types of obstacles include static obstacles and dynamic obstacles. Static obstacles can be understood as immovable obstacles, such as walls, pillars, and curbs. Dynamic obstacles can be understood as movable obstacles, such as pedestrians, bicycles, electric vehicles, and cars.

[0118] Optionally, after obtaining the perception results of the vehicle, obstacle detection can be performed on the perception results of the vehicle to obtain the semantic information of the obstacle, and the type and outline information of the obstacle can be determined based on the semantic information of the obstacle.

[0119] For example, if the semantic information of obstacle 1 is a wall, the semantic information of obstacle 2 is a pillar, and the semantic information of obstacle 3 is a car, based on the semantic information of obstacle 1, obstacle 2, and obstacle 3, the pre-built mapping table between obstacle semantic information and obstacle type is queried to determine that obstacle 1 and obstacle 2 are static obstacles and obstacle 3 is dynamic obstacles.

[0120] S803: Based on the semantic information of obstacles in the obstacle map, the obstacles are filtered to obtain target obstacles that pose a risk of obstructing the line of sight.

[0121] S804 determines the blind spot based on the sensing observation point and the target obstacle.

[0122] It should be noted that, due to the constant movement of the vehicle and the changing surrounding environment, the blind spot needs to be updated every frame. By continuously updating and building the current obstacle map of the vehicle, the blind spot of the vehicle's sensors is updated and determined based on the current obstacle map and the vehicle's perception observation points.

[0123] In this embodiment of the disclosure, after obtaining the blind spot, the blind spot can be sent to the vehicle planning terminal.

[0124] S805, in response to the vehicle being about to arrive at an intersection within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being less than a second set distance threshold, decelerates the vehicle.

[0125] It should be noted that this disclosure does not limit the setting of the second set distance threshold. For example, the second set distance threshold can be set to 5m.

[0126] In this embodiment of the present disclosure, after receiving the blind spot, the vehicle planning terminal can determine whether the intersection the vehicle is about to reach is within the blind spot. In response to the intersection the vehicle is about to reach being within the blind spot, a slow deceleration control signal can be sent to the controller of the vehicle planning terminal to decelerate the vehicle and achieve smooth deceleration of the vehicle.

[0127] In this embodiment of the disclosure, after the vehicle is decelerated, in response to the vehicle passing through the intersection, a control signal to accelerate to the cruising speed is sent to the controller at the vehicle planning end, so that the vehicle can smoothly accelerate to the cruising speed.

[0128] S806 monitors the distance between the vehicle and the center point of the intersection, and decelerates the vehicle when the distance between the vehicle and the center point of the intersection is less than or equal to a set distance threshold.

[0129] S807, in response to the vehicle's expected arrival at an intersection not being within the blind spot, or in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second preset distance threshold, sends a speed-maintaining control signal to the vehicle's controller.

[0130] In this embodiment of the disclosure, a speed-maintaining control signal can be sent to the controller at the vehicle planning end to keep the vehicle at its current speed without deceleration.

[0131] A vehicle control method according to an embodiment of this disclosure acquires the vehicle's perception results, establishes a current obstacle map based on the map information where the vehicle is located and the vehicle's perception results, filters obstacles based on the semantic information of obstacles in the obstacle map to obtain target obstacles that pose a risk of visual obstruction, determines blind spots based on the perception observation point and the target obstacles, and, in response to the vehicle's expected arrival at an intersection being within the blind spot and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being less than a second set distance threshold, performs deceleration control on the vehicle, monitors the distance between the vehicle and the center point of the intersection, and, in response to the distance between the vehicle and the center point of the intersection being less than or equal to a set distance threshold, decelerates the vehicle. A speed-maintaining control signal is sent to the vehicle's controller in response to a predetermined distance threshold. This is done when the vehicle is not within the blind spot of its expected arrival at the intersection, or when the vehicle is within the blind spot and the distance between the intersection's center point and either the first or second blind spot boundary line is greater than or equal to a second predetermined distance threshold. Specifically, this disclosure responds to the vehicle being within the blind spot and the distance between the intersection's center point and either the first or second blind spot boundary line is less than the second predetermined distance threshold. By monitoring the distance between the vehicle and the intersection's center point, and responding when the distance is less than or equal to the predetermined distance threshold... It can decelerate the vehicle in advance, reducing the risk of unexpected pedestrians appearing out of sight, improving the vehicle's safety factor, and ensuring safety during vehicle operation. In response to the vehicle's expected arrival at an intersection not being in the blind spot, or in response to the vehicle's expected arrival at an intersection being in the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second set distance threshold, a speed-maintaining control signal is sent to the vehicle's controller. The vehicle does not need to decelerate, which helps to improve the vehicle's average cruising speed. It can control the vehicle's acceleration / deceleration according to different actual conditions, avoiding the problem of heavy mechanical feel and improving the user experience.

[0132] Figure 9 This is a schematic flowchart illustrating a vehicle control method provided in an embodiment of this disclosure. The reference numerals in the flowchart do not represent any timing limitations.

[0133] like Figure 9 As shown, the vehicle control method includes the following steps:

[0134] S901, acquires the perception results of the vehicle.

[0135] S902, based on the map information where the vehicle is located and the vehicle's perception results, establishes the current obstacle map for the vehicle.

[0136] S903, based on the semantic information of obstacles in the obstacle map, filters the obstacles to obtain target obstacles that pose a risk of visual obstruction.

[0137] S904, determine the type of the target obstacle based on the semantic information of the target obstacle.

[0138] S905, in response to the target obstacle being a first type of obstacle, determine the first edge point and the last edge point from the set of edge points of the target obstacle.

[0139] S906, based on the first edge point and the last edge point, identify the visual obstruction point pairs of the target obstacle from the set of edge points, wherein the first edge point and the last edge point are determined based on the traversal direction of the set of edge points.

[0140] S907, in response to the target obstacle being a second type of obstacle, determine the corner points of the target obstacle, wherein the corner points of the target obstacle are calculated by a preset target detection algorithm.

[0141] S908 determines pairs of occlusion points from corner points based on sensing observation points.

[0142] S909, generate the first blind zone boundary line based on the first visual obstruction point in the pair between the sensing observation point and the visual obstruction point.

[0143] S9010, based on the alignment of the sensing observation point and the field of view occlusion point with the second field of view occlusion point, generate the second blind zone boundary line.

[0144] S9011, determine the blind zone based on the first blind zone boundary line and the second blind zone boundary line of the visual field occlusion point pair.

[0145] S9012, in response to the vehicle being about to arrive at an intersection within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being less than a second set distance threshold, deceleration control is applied to the vehicle.

[0146] S9013 monitors the distance between a vehicle and the center point of an intersection, and decelerates the vehicle when the distance between the vehicle and the center point of the intersection is less than or equal to a set distance threshold.

[0147] S9014, in response to the vehicle's expected arrival at an intersection not being within the blind spot, or in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second preset distance threshold, sends a speed-maintaining control signal to the vehicle's controller.

[0148] In summary, the vehicle control method of this disclosure can adopt different blind spot determination methods for different types of target obstacles, improving the flexibility and accuracy of blind spot determination. When the vehicle is about to arrive at an intersection within the blind spot, and the distance between the center point of the intersection and either the first or second blind spot boundary line is less than a second set distance threshold, by monitoring the distance between the vehicle and the center point of the intersection, and when the distance is less than or equal to the set distance threshold, the vehicle can be slowed down in advance, reducing the risk of unexpected pedestrian appearances and improving vehicle safety. When the vehicle is about to arrive at an intersection not within the blind spot, or when the vehicle is about to arrive at an intersection within the blind spot, and the distance between the center point of the intersection and either the first or second blind spot boundary line is greater than or equal to the second set distance threshold, a speed-maintaining control signal is sent to the vehicle's controller. The vehicle does not need to slow down, which helps to improve the vehicle's average cruising speed. The method can control acceleration / deceleration of the vehicle according to different actual conditions, avoiding a heavy mechanical feel and improving the user experience.

[0149] Figure 10 This is a block diagram illustrating a vehicle control device according to an exemplary embodiment.

[0150] like Figure 10 As shown, the vehicle control device 1000 includes: a first determining module 110, a second determining module 120, and a control module 130.

[0151] The first determining module 110 is used to establish a current obstacle map of the vehicle, wherein the obstacle map is obtained based on the map information of the vehicle and the perception results of the vehicle;

[0152] The second determining module 120 is used to determine the blind spot of the vehicle sensor based on the obstacle map and the vehicle's perception observation point, wherein the perception observation point is the installation position of the vehicle sensor on the vehicle.

[0153] The control module 130 is used to decelerate the vehicle in response to the vehicle being located within the blind spot of the intersection it is about to reach.

[0154] Furthermore, the second determining module 120 is used to: filter the obstacles according to the semantic information of the obstacles in the obstacle map to obtain target obstacles that pose a risk of visual obstruction; and determine the blind spot according to the perception observation point and the target obstacle.

[0155] Furthermore, the second determining module 120 is configured to: determine the type of the target obstacle based on the semantic information of the target obstacle; determine the visual field occlusion point pair of the target obstacle based on the type of the target obstacle; and obtain the visual field blind spot based on the perception observation point and the visual field occlusion point pair.

[0156] Furthermore, the second determining module 120 is configured to: in response to the target obstacle being a first type of obstacle, determine a first edge point and a tail edge point from the set of edge points of the target obstacle; and identify a pair of visual obstruction points of the target obstacle from the set of edge points based on the first edge point and the tail edge point; wherein the first edge point and the tail edge point are determined based on the traversal direction of the set of edge points.

[0157] Furthermore, the second determining module 120 is configured to: generate a first-to-last connecting line based on the first edge point and the last edge point; traverse and filter the edge points in the edge point set to obtain the distance between the currently traversed edge point and the first-to-last connecting line; determine the currently traversed edge point as a view occlusion point in response to the distance being greater than or equal to a first set distance threshold; and determine two adjacent view occlusion points as a view occlusion point pair in response to the end of traversal.

[0158] Furthermore, after determining that the currently traversed edge point is a visual obstruction point, the device 1000 is further configured to: generate a new set of edge points from the currently traversed edge point and the remaining untraversed edge points, wherein the currently traversed edge point is the first edge point of the new set of edge points; and continue to identify the next pair of visual obstructions of the target obstacle based on the first edge point and the last edge point in the new set of edge points, until the traversal ends.

[0159] Furthermore, the second determining module 120 is configured to: in response to the target obstacle being a second type of obstacle, determine the corner points of the target obstacle, wherein the corner points of the target obstacle are calculated by a preset target detection algorithm; and determine the visual obstruction point pair from the corner points based on the perception observation point.

[0160] Furthermore, the second determining module 120 is used to: connect the corner point and the sensing observation point to generate a line connecting the corner point and the sensing observation point; determine the angle between the line connecting the corner point and the sensing observation point and the preset reference line; and select the corner point corresponding to the largest angle and the corner point corresponding to the smallest angle as the field of view occlusion point pair.

[0161] Furthermore, the number of the visual field occlusion point pairs is multiple, and the second determining module 120 is used to: determine the corresponding candidate visual field blind zone based on the first blind zone boundary line and the second blind zone boundary line of each visual field occlusion point pair; and combine multiple candidate visual field blind zones to obtain the visual field blind zone.

[0162] Furthermore, the second determining module 120 is configured to: generate a first blind zone boundary line based on a first visual field occlusion point in the pair of the sensing observation point and the visual field occlusion point; generate a second blind zone boundary line based on a second visual field occlusion point in the pair of the sensing observation point and the visual field occlusion point; and determine the visual field blind zone based on the first blind zone boundary line and the second blind zone boundary line of the pair of visual field occlusion points.

[0163] Furthermore, the control module 130 is configured to: in response to the vehicle being about to reach an intersection within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being less than a second preset distance threshold, decelerate the vehicle.

[0164] Furthermore, the control module 130 is used to: monitor the distance between the vehicle and the center point of the intersection; and, in response to the distance between the vehicle and the center point of the intersection being less than or equal to a set distance threshold, to control the vehicle to decelerate.

[0165] Furthermore, the device 1000 is configured to: in response to the vehicle's expected arrival at an intersection not being within the blind spot, or in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second preset distance threshold, send a speed-maintaining control signal to the vehicle's controller.

[0166] A vehicle control device according to an embodiment of this disclosure establishes a current obstacle map for the vehicle. The obstacle map is obtained based on the map information of the vehicle's location and the vehicle's perception results. Based on the obstacle map and the vehicle's perception observation points, the blind spots of the vehicle's sensors are determined. The perception observation points are the installation positions of the vehicle sensors on the vehicle. In response to the vehicle's expected arrival at an intersection being within the blind spot, the vehicle is decelerated. Thus, by determining the blind spots of the vehicle's sensors, this disclosure can decelerate the vehicle in advance in response to the vehicle's expected arrival at an intersection being within the blind spot, reducing the risk of vehicles suddenly appearing out of sight and improving the safety of the vehicle during driving.

[0167] To implement the above embodiments, this disclosure also provides a vehicle, such as... Figure 11 As shown, the vehicle 2000 includes: a processor 210; and one or more memories 220 for storing executable instructions of the processor 210; wherein the processor 210 is configured to execute the vehicle control method described in the above embodiments. The processor 210 and the memories 220 are connected via a communication bus.

[0168] To implement the above embodiments, this disclosure also provides a computer-readable storage medium including instructions, such as a memory 220 including instructions, which can be executed by the processor 210 of the device 1000 to perform the above methods. Optionally, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0169] To implement the above embodiments, this disclosure also provides a chip 3000, such as... Figure 12 As shown, the chip includes an interface circuit and a processing circuit that are coupled to each other. The interface circuit is used to input or output signals, and the processing circuit is configured to implement the steps of the vehicle control method described in the above embodiments.

[0170] Figure 12 This is a schematic diagram of the structure of a chip according to an embodiment of this disclosure. See also... Figure 12 The diagram shown is a schematic representation of the structure of chip 3000, but it is not limited to this.

[0171] The chip 3000 includes a processing circuit 310 and an interface circuit 320. The interface circuit 320 is used to read instructions and send instructions to the processing circuit 310 so that the processing circuit 310 executes the steps of the vehicle control method described in the above embodiments.

[0172] Optionally, such as Figure 13 As shown, Figure 13This is a schematic diagram of another chip structure proposed in an embodiment of this disclosure. The chip 3000 may further include: a memory 330 for storing instructions, and an interface circuit 320 for reading the instructions stored in the memory 330.

[0173] Optionally, the interface circuit 320 is connected to the memory 330. The interface circuit 320 can be used to receive signals from the memory 330 or other devices, and can also be used to send signals to the memory 330 or other devices. For example, the interface circuit 320 can read instructions stored in the memory 330 and send those instructions to the processing circuit 310.

[0174] Optionally, the number of memories 330 can be one or more. The number of interface circuits 320 can also be one or more.

[0175] In some embodiments, the interface circuit 320 performs at least one of the communication steps such as sending and / or receiving in the above method, and the processing circuit 310 performs other steps.

[0176] In some embodiments, the terms interface circuit, interface, transceiver pin, transceiver, etc., can be used interchangeably.

[0177] Alternatively, all or part of the memory 330 may be located outside the chip 3000.

[0178] To implement the above embodiments, this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the vehicle control method described in the above embodiments.

[0179] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0180] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A vehicle control method, characterized in that, The method includes: Establish a current obstacle map for the vehicle, wherein the obstacle map is obtained based on map information of the vehicle's location and the vehicle's perception results; Based on the semantic information of obstacles in the obstacle map, the obstacles are filtered to obtain target obstacles that pose a risk of obstructing the line of sight; Based on the semantic information of the target obstacle, the type of the target obstacle is determined, wherein the type of the target obstacle includes a static type or a dynamic type; Based on the type of the target obstacle, determine the visual obstruction point pairs of the target obstacle, wherein the determination method for visual obstruction point pairs differs for different types of target obstacles; Based on the vehicle's perception observation points and the field of view occlusion point pair, the blind spot of the vehicle sensor's field of view is determined, wherein the perception observation points are the installation positions of the vehicle sensor on the vehicle. In response to the vehicle's expected arrival at the intersection being within the blind spot, the vehicle is decelerated. The step of determining the pair of visual obstruction points of the target obstacle based on the type of the target obstacle includes: In response to the target obstacle being a first type of obstacle, the visual obstruction point pair is determined based on the first edge point and the last edge point in the edge point set of the target obstacle, wherein the first type of obstacle is a static type of obstacle.

2. The method according to claim 1, characterized in that, Determining the pair of obstructing points based on the first and last edge points in the set of edge points of the target obstacle includes: Determine the first edge point and the last edge point from the set of edge points of the target obstacle; Based on the first edge point and the last edge point, identify the visual obstruction point pairs of the target obstacle from the set of edge points; The first edge point and the last edge point are determined based on the traversal direction of the edge point set.

3. The method according to claim 2, characterized in that, The step of identifying pairs of visual obstruction points of the target obstacle from the set of edge points based on the first edge point and the last edge point includes: Generate a start-end connection line based on the start edge point and end edge point; The edge points in the set of edge points are traversed and filtered to obtain the distance between the currently traversed edge point and the beginning and end connecting line; In response to the distance being greater than or equal to a first preset distance threshold, the currently traversed edge point is determined to be a visual occlusion point; In response to the end of the traversal, two adjacent occlusion points are determined to be a pair of occlusion points.

4. The method according to claim 3, characterized in that, After determining that the currently traversed edge point is a view occlusion point, the method further includes: The currently traversed edge point and the remaining untraversed edge points are used to generate a new set of edge points, where the currently traversed edge point is the first edge point of the new set of edge points. Based on the first and last edge points in the new set of edge points, continue to identify the next pair of visual obstruction points of the target obstacle until the traversal is complete.

5. The method according to claim 1, characterized in that, The step of determining the pair of visual obstruction points of the target obstacle based on the type of the target obstacle further includes: In response to the target obstacle being a second type of obstacle, the corner points of the target obstacle are determined; wherein, the corner points of the target obstacle are calculated by a preset target detection algorithm; Based on the perception observation points, the pairs of visual obstruction points are determined from the corner points.

6. The method according to claim 5, characterized in that, The step of determining the pair of visual obstruction points from the corner points based on the sensing observation points includes: Connect the corner point to the sensing observation point to generate a line between the corner point and the sensing observation point; Determine the angle between the line connecting the corner point and the sensing observation point and the preset reference line; The corner points corresponding to the largest included angle and the corner points corresponding to the smallest included angle are selected as the field of view occlusion point pairs.

7. The method according to any one of claims 1-6, characterized in that, The process of obtaining the blind spot based on the sensing observation point and the visual field occlusion point pair includes: A first blind zone boundary line is generated based on the first visual obstruction point in the pair between the perception observation point and the visual obstruction point. A second blind zone boundary line is generated by aligning the second visual obstruction point with the perception observation point and the visual obstruction point. The visual blind zone is determined based on the first blind zone boundary line and the second blind zone boundary line of the visual field occlusion point pair.

8. The method according to claim 7, characterized in that, The number of the visual field obstruction point pairs is multiple, and the determination of the visual field blind zone based on the first blind zone boundary line and the second blind zone boundary line of the visual field obstruction point pairs includes: Based on the first blind zone boundary line and the second blind zone boundary line of each pair of visual field occlusion points, the corresponding candidate visual field blind zone is determined. The candidate blind zones are combined to obtain the blind zone.

9. The method according to claim 7, characterized in that, The response to the vehicle's expected arrival at the intersection being within the blind spot, including deceleration control of the vehicle, includes: In response to the vehicle being about to arrive at an intersection being within the blind spot, and the distance between the center point of the intersection and either the boundary line of the first blind spot or the boundary line of the second blind spot being less than a second set distance threshold, the vehicle is decelerated.

10. The method according to claim 9, characterized in that, The method further includes at least one of the following operations: In response to the vehicle's expected arrival at an intersection not being within the blind spot, or in response to the vehicle's expected arrival at an intersection being within the blind spot, and the distance between the center point of the intersection and either the first blind spot boundary line or the second blind spot boundary line being greater than or equal to a second preset distance threshold, a speed maintenance control signal is sent to the vehicle's controller.

11. The method according to claim 1, characterized in that, The method of responding to the vehicle's expected arrival at the intersection being within the blind spot by decelerating the vehicle further includes: The distance between the vehicle and the center point of the intersection is monitored; In response to the distance between the vehicle and the center point of the intersection being less than or equal to a set distance threshold, the vehicle is decelerated.

12. A vehicle control device, characterized in that, The device includes: The first determining module is used to establish a current obstacle map for the vehicle, wherein the obstacle map is obtained based on map information of the vehicle's location and the vehicle's perception results; The second determining module is used to determine the blind spot of the vehicle sensor based on the obstacle map and the vehicle's perception observation point, wherein the perception observation point is the installation position of the vehicle sensor on the vehicle. The control module is used to decelerate the vehicle in response to the vehicle being located within the blind spot at the intersection it is about to reach. The second determining module is further configured to: Based on the semantic information of obstacles in the obstacle map, the obstacles are filtered to obtain target obstacles that pose a risk of obstructing the line of sight; Based on the semantic information of the target obstacle, the type of the target obstacle is determined, wherein the type of the target obstacle includes a static type or a dynamic type; Based on the type of the target obstacle, determine the visual obstruction point pairs of the target obstacle, wherein the determination method for visual obstruction point pairs differs for different types of target obstacles; The blind spot is determined based on the vehicle's perception observation points and the pair of visual field obstruction points; The second determining module is further configured to: In response to the target obstacle being a first type of obstacle, the visual obstruction point pair is determined based on the first edge point and the last edge point in the edge point set of the target obstacle, wherein the first type of obstacle is a static type of obstacle.

13. The apparatus according to claim 12, characterized in that, Determining the pair of visual obstruction points based on the first edge point and the last edge point in the set of edge points of the target obstacle includes: Determine the first edge point and the last edge point from the set of edge points of the target obstacle; Based on the first edge point and the last edge point, identify the visual obstruction point pairs of the target obstacle from the set of edge points; The first edge point and the last edge point are determined based on the traversal direction of the edge point set.

14. The apparatus according to claim 12, characterized in that, The second determining module is further configured to: In response to the target obstacle being a second type of obstacle, the corner points of the target obstacle are determined; wherein, the corner points of the target obstacle are calculated by a preset target detection algorithm; Based on the sensing observation points, the pairs of visual obstruction points are determined from the corner points.

15. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the method as described in any one of claims 1-11.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1-11.

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

18. A chip, characterized in that, The chip includes an interface circuit and a processing circuit coupled to each other. The interface circuit is used to input or output signals, and the processing circuit is configured to implement the steps of the method according to any one of claims 1-11.

Images