Obstacle detection method, apparatus, device, and storage medium

Abstract

The disclosure provides an obstacle detection method, device, equipment and storage medium, relates to the technical field of computers, in particular to the field of automatic driving, obstacle detection and intelligent transportation. The specific implementation scheme is: according to the position of the vehicle, the detection range of the vehicle is divided into a plurality of detection areas; the detection frequency corresponding to each of the plurality of detection areas is obtained; and obstacle detection is performed on each of the plurality of detection areas according to the detection frequency corresponding to each of the plurality of detection areas. In the embodiment of the disclosure, by reasonably arranging the detection frequency corresponding to the detection area, the detection efficiency can be improved, and the delay can be reduced.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to the fields of autonomous driving, obstacle detection, and intelligent transportation. Background Technology

[0002] Current autonomous driving methods employ comprehensive scene perception and computation, such as full-scale perception computation, planning and control (PNC) lane search, etc. Comprehensive perception and computation require high computing power, resulting in significant latency and slow response. Summary of the Invention

[0003] This disclosure provides an obstacle detection method, apparatus, device, and storage medium.

[0004] According to one aspect of this disclosure, an obstacle detection method is provided, the method comprising the following steps:

[0005] Based on the vehicle's location, the detection area for the vehicle is divided into multiple detection zones;

[0006] Obtain the detection frequency corresponding to each of the multiple detection regions;

[0007] Obstacle detection is performed on each of the multiple detection areas according to the corresponding detection frequency.

[0008] According to another aspect of this disclosure, an obstacle detection device is provided, the device comprising:

[0009] The segmentation module is used to divide the detection range of a vehicle into multiple detection zones based on the vehicle's location.

[0010] The acquisition module is used to acquire the detection frequencies corresponding to the multiple detection areas respectively;

[0011] The detection module is used to perform obstacle detection on the multiple detection areas according to the detection frequency corresponding to each of the multiple detection areas.

[0012] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0013] At least one processor; and

[0014] The memory is communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods of any embodiment of the present disclosure.

[0016] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform a method according to any embodiment of this disclosure.

[0017] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements a method according to any embodiment of this disclosure.

[0018] According to another aspect of this disclosure, an autonomous vehicle is provided, including: an electronic device according to an embodiment of this disclosure.

[0019] 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

[0020] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0021] Figure 1 This is a schematic flowchart of an obstacle detection method according to an embodiment of the present disclosure;

[0022] Figure 2 This is a schematic flowchart of an obstacle detection method according to another embodiment of the present disclosure;

[0023] Figure 3 This is a schematic diagram illustrating the division of the detection region based on the region of interest;

[0024] Figure 4 This is a schematic flowchart of an obstacle detection method according to another embodiment of the present disclosure;

[0025] Figure 5a , Figure 5b , Figure 5c and Figure 5d These are schematic diagrams of the detection areas in non-intersection areas;

[0026] Figure 6a , Figure 6b , Figure 6c and Figure 6d These are schematic diagrams showing the detection areas for going straight, turning right, turning left, and making a U-turn, respectively.

[0027] Figure 7a This is a schematic diagram of the sub-regions of the target road segment and the interfering road segment under the intention to go straight;

[0028] Figure 7b This is a schematic diagram of a sub-region of the target road segment under the intention to turn right;

[0029] Figure 8a This is a schematic diagram of the boundary of the first sub-region;

[0030] Figure 8b This is a schematic diagram of the boundary of the second sub-region;

[0031] Figure 8c This is a schematic diagram of the boundary of the third sub-region;

[0032] Figure 9a and Figure 9b This is a schematic diagram illustrating the determination of the detection area based on obstacles;

[0033] Figure 10 This is a schematic diagram of a regional attention architecture;

[0034] Figure 11 This is a schematic diagram of the structure of an obstacle detection device according to an embodiment of the present disclosure;

[0035] Figure 12 This is a schematic diagram of the structure of an obstacle detection device according to another embodiment of the present disclosure;

[0036] Figure 13 This is a block diagram of an electronic device used to implement the obstacle detection method of the embodiments of this disclosure. Detailed Implementation

[0037] 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 of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0038] Figure 1 This is a flowchart illustrating an obstacle detection method according to an embodiment of the present disclosure, which may include:

[0039] S101. Based on the vehicle's location, divide the vehicle's detection range into multiple detection zones;

[0040] S102. Obtain the detection frequency corresponding to each of the multiple detection areas;

[0041] S103. According to the detection frequency corresponding to each of the multiple detection areas, perform obstacle detection on each of the multiple detection areas.

[0042] In this embodiment, the vehicle can collect data through its own sensors and can also communicate with external devices such as cloud or roadside equipment to obtain data. The data acquired by the vehicle may include its location, which can include its geographic location or its relative location on a map. Various types of sensors, such as radar and cameras, can be installed on the vehicle. The sensors on the vehicle can have a certain detection range. Different types of sensors may have different detection ranges. Based on the vehicle's location, the vehicle's detection range, such as its maximum detection range, can be divided into multiple detection areas. The vehicle's maximum detection range may exceed the portion displayed on the electronic map currently being used by the vehicle. There are various methods for dividing the vehicle's detection range into multiple detection areas. For example, detection areas can be divided according to distance, with detection areas farther away having lower detection frequencies. Alternatively, detection areas can be divided according to the type of responsibility, with detection areas where the vehicle may bear primary responsibility having higher detection frequencies than other areas. Furthermore, detection areas can be divided according to the detection capabilities of one or more sensors, with detection areas within the detection range of a certain sensor having higher detection frequencies than other areas.

[0043] In this embodiment, multiple detection areas can be divided into various types or levels of detection areas, and corresponding priorities can be set for different types or levels of detection areas. Each priority has a corresponding frequency. The detection frequency corresponding to a detection area can be obtained based on its priority. If there are many types or levels of detection areas, some different types or levels of detection areas may have the same detection frequency.

[0044] In this embodiment of the disclosure, obstacle detection requires detecting parameters such as the obstacle's type, location, and speed. Detecting obstacles according to different detection frequencies for different areas helps improve detection efficiency and reduce latency by rationally arranging the detection frequencies corresponding to different areas. For example, the detection frequency in non-primary responsibility areas can be reduced, while the detection frequency in primary responsibility areas can be increased, thereby improving safety.

[0045] Figure 2 This is a flowchart illustrating an obstacle detection method according to another embodiment of the present disclosure. The method may include one or more features of the obstacle detection method described in the above embodiments. In one implementation, S101 divides the detection range of the vehicle into multiple detection areas based on the vehicle's position, including:

[0046] S201. Calculate the detection distance of the vehicle based on the road speed limit at the vehicle's location;

[0047] S202. Determine the region of interest based on the vehicle's detection distance;

[0048] S203. Based on the region of interest, divide the vehicle into multiple detection zones.

[0049] In this embodiment of the disclosure, road speed limits may include the maximum or minimum speed limit for vehicles on the road. An example of a formula for calculating the detection distance of a vehicle based on the road speed limit is as follows:

[0050] S = V 2 / 2a Formula 1

[0051] Where S represents the vehicle's detection distance, V represents the current road speed limit, and a represents the vehicle's deceleration, which can be taken as an absolute value. Deceleration refers to the vehicle's ability to rapidly reduce its speed until it comes to a stop. Using deceleration to determine the detection distance is beneficial for enabling timely stopping through normal slowing when a safety issue is detected.

[0052] In this embodiment of the disclosure, the calculated detection distance may be less than the vehicle's actual maximum detection distance. Therefore, the region of interest may be within the vehicle's actual maximum detection range.

[0053] In some examples, the extent of the region of interest (ROI) can be determined based on its shape and the radius (S). For instance, if the ROI is circular, it can be defined with a point on the vehicle as the center and S as the radius or diameter. Similarly, if the ROI is square, it can be defined by extending S along the road direction in all directions from a point on the vehicle as the center. The ROI can also be other shapes, depending on the specific requirements.

[0054] By using this disclosure, regions of interest can be divided according to the current speed limit of the road. Different detection frequencies can be set for different regions. For example, the detection frequency in the region of interest can be set higher than that outside the region of interest, which can improve detection speed, reduce latency, and improve the safety of autonomous driving.

[0055] In one implementation, S202 determines the region of interest based on the detection distance of the vehicle, including:

[0056] The upper and lower boundaries of the region of interest are determined based on the forward and rear detection distances of the vehicle.

[0057] The left and right boundaries of the region of interest are determined based on the forward detection distance of the vehicle.

[0058] In this embodiment of the disclosure, the vehicle detection distance may include a forward detection distance and a rear detection distance. The forward detection distance and the rear detection distance may be the same or different. Assuming the forward detection distance and the rear detection distance are the same, Formula 1 is modified as follows:

[0059] N = M = V 2 / 2a Formula 2

[0060] Where N represents the detection distance in front of the vehicle, M represents the detection distance behind the vehicle, V represents the current speed limit of the road, and a can represent the deceleration of the vehicle.

[0061] In one example, such as Figure 3 As shown, the upper and lower boundaries of the region of interest (ROI) can be determined by M and N, while the left and right boundaries can be determined by N. If the values ​​of N and M are both 70m, the ROI can include a 70m area in front of, behind, to the left and right of the vehicle. This ROI may include lanes within the road and areas outside the road, centered on the vehicle. The ROI is a square area of ​​140m x 140m. Of course, the values ​​of M and N can also be different, for example, N = 1.5M or 2N = M, depending on the specific requirements.

[0062] By employing the embodiments of this disclosure, regions of interest that conform to the road where the vehicle is located and the vehicle's own kinematic characteristics can be quickly determined, thereby enabling more timely and accurate obstacle detection.

[0063] In one implementation, S203 divides the vehicle into multiple detection areas based on the region of interest, including: dividing the area outside the region of interest in the map and / or the area outside the map into a first detection area.

[0064] In the embodiments disclosed herein, such as Figure 3 As shown, if the Region of Interest (ROI) is a square area, the area outside this square area can be designated as the first detection area, also known as the non-ROI. The first detection area can include areas outside the map range, or it can include areas of non-primary responsibility (non-responsibility areas). Generally, these areas are far from the vehicle, and obstacles in these areas have little impact on vehicle operation. Therefore, the detection frequency in these areas can be reduced, thereby reducing the detection frequency of non-ROI areas and reducing unnecessary obstacle detection.

[0065] In one embodiment, S203 divides the vehicle into multiple detection areas based on the region of interest, and further includes dividing the area within the region of interest into a second detection area.

[0066] In the embodiments disclosed herein, such as Figure 3 As shown, the region of interest can be designated as the second detection region. The region of interest is more important than the non-detection region. Therefore, the detection frequency of the second detection region can be set higher than that of the first detection region, allowing more computing power to be allocated to more important areas and improving overall obstacle detection efficiency.

[0067] Figure 4 This is a flowchart illustrating an obstacle detection method according to another embodiment of the present disclosure. The method may include one or more features of the obstacle detection method described in the above embodiments. In one implementation, S101 divides the detection range of the vehicle into multiple detection areas based on the vehicle's position, and further includes:

[0068] S401. Based on the vehicle's location, determine the driving scenario corresponding to the vehicle;

[0069] S402. Obtain a region attention strategy based on the driving scenario and driving intention of the vehicle, wherein the region attention strategy includes at least one of region offset, region shape and region size.

[0070] S403. Based on the attention strategy of this region, divide the area into a third detection region.

[0071] For example, a vehicle's driving scenario can include intersection scenarios and non-intersection scenarios. If the vehicle is not located at an intersection, the driving scenario is a non-intersection scenario; if the vehicle is located at an intersection, the driving scenario is an intersection scenario. The third detection area, determined based on the driving scenario and driving intent, can include the vehicle's primary responsibility area (or simply primary responsibility area). The primary responsibility area indicates the area where the vehicle may bear primary responsibility for a traffic accident.

[0072] In this embodiment, the area attention strategy may include area offset, area shape, and area size. The area attention strategy may correspond to driving intentions, and different driving intentions may correspond to one or more area attention strategies. Specifically, the area offset may include the distance the area needs to be offset, the area shape may include a circle, square, rectangle, irregular shape, etc., and the area size may include parameters such as the area, radius, diameter, length, and width of the area. The area size and area shape are related. For example, when the area shape is circular, the area size includes the radius, diameter, and area of ​​the circle. When the area shape is square, the area size includes the side length and area of ​​the square. When the area shape is rectangular, the area size includes the length, width, and area of ​​the rectangle.

[0073] In this embodiment of the disclosure, the third detection region is divided by the regional attention strategy obtained from the driving scenario and driving intention. This can result in regions with higher safety requirements. Setting the detection frequency of the third detection region to be higher can increase the detection frequency of regions with higher safety requirements, thereby improving detection efficiency and driving safety.

[0074] In one implementation, S402 obtains a region attention strategy based on the driving scenario corresponding to the vehicle and the driving intention corresponding to the vehicle, including:

[0075] In the case that the driving scenario is not an intersection scenario, the lane information of the road segment where the vehicle is traveling is determined according to the driving intention of the vehicle, and the left and right boundaries of the third detection area are determined according to the lane information and area offset of the road segment where the vehicle is traveling; the upper and lower boundaries of the third detection area are determined according to the detection range of the vehicle.

[0076] For example, such as Figure 5a , Figure 5b , Figure 5c and Figure 5d As shown, when the vehicle is traveling on a non-intersection road segment, the third detection area can include a portion of the lane in the vehicle's direction of travel. The upper and lower boundaries of the third detection area on the rightmost three lanes are obtained using N and M determined by Formula 2; for example, the distance between the upper and lower boundaries is N+M. The right boundary of the third detection area can be the boundary of the rightmost lane. Furthermore, if the area offset includes a half-lane offset from the rightmost lane, then extending half a lane to the right of the rightmost lane can be used as the right boundary of the third detection area. The left boundary of the third detection area can be the centerline of the road. Furthermore, if the area offset includes a half-lane offset from the leftmost lane, then extending half a lane to the left of the leftmost lane can be used as the left boundary of the third detection area.

[0077] For example, if the driver's intention is to change lanes to the right, the right boundary of the third detection area can be extended to the right by half a lane. If the driver's intention is to change lanes to the left, the right boundary of the third detection area can be extended to the left by half a lane.

[0078] In this embodiment of the disclosure, by using the lane information and area offset of the road segment where the vehicle's driving direction is located in a non-intersection scenario, a third detection area with higher safety requirements in a non-intersection scenario can be obtained. Setting the detection frequency of the third detection area to a higher level can increase the detection frequency of areas with higher safety requirements in a non-intersection scenario, thereby improving the detection efficiency and driving safety in a non-intersection scenario.

[0079] In one implementation, S402 obtains a region attention strategy based on the driving scenario corresponding to the vehicle and the driving intention corresponding to the vehicle, and further includes:

[0080] In the case of an intersection driving scenario, the target road segment that the vehicle will enter after passing through the intersection is determined based on the vehicle's corresponding driving intention.

[0081] The interfering road section is determined based on the road segment where the vehicle is located, the driving intention, and information on obstacles that may enter the intersection;

[0082] The third detection area is determined to be a sub-region of the road segment where the vehicle is located, the target road segment, and the interfering road segment.

[0083] In this embodiment of the disclosure, the driving intentions of a vehicle in an intersection scenario may be more varied, such as going straight, turning left, turning right, or making a U-turn. The third detection area obtained based on the attention strategy corresponding to different driving intentions may be different. Figure 6a , Figure 6b , Figure 6c and Figure 6d These are schematic diagrams showing the detection areas for going straight, turning right, turning left, and making a U-turn.

[0084] In this embodiment of the disclosure, the target road segment that the vehicle will enter after passing the intersection is related to the driving intention. Figure 7a In this context, the sub-region of the target road segment with a straight-ahead intention is L1. Figure 7b In the diagram, the sub-region of the target road segment under the intention to turn right is L2.

[0085] In this embodiment of the disclosure, the interfering road segment may include a road segment where obstacles may appear at the intersection and collide with vehicles. Different driving intentions may result in different sub-regions of the interfering road segment at the intersection. Taking a crossroads as an example, in... Figure 7a Under the intent to proceed straight, the sub-regions of the interfering road segment may include D11, D12, D13, and D14. Among them, D14 is the area occupied by the sidewalk in the road segment where the vehicle is located. Since the sidewalk may mainly have obstacles such as pedestrians, the sub-region of this interfering road segment may mainly include the smaller area occupied by the sidewalk.

[0086] In this embodiment of the disclosure, the road segment where the vehicle is located, the target road segment, and the interfering road segment can be obtained according to the intersection scene and driving intention, and the sub-regions of the road segment where the vehicle is located, the target road segment, and the interfering road segment can be determined. In this way, a third detection area with higher safety requirements in the intersection scene can be obtained. Setting the detection frequency of the third detection area to a higher level can increase the detection frequency of the area with higher safety requirements in the intersection scene, thereby improving the detection efficiency and driving safety in the intersection scene.

[0087] In one implementation, determining the third detection area as a sub-region of the road segment where the vehicle is located includes:

[0088] Based on the detection distance of the vehicle, determine the first boundary of the first sub-region corresponding to the road segment where the vehicle is located;

[0089] The second boundary of the first sub-region is determined based on the lane information and area offset of the road segment where the vehicle is located;

[0090] Determine the third and / or fourth boundaries of the first sub-region based on the center point of the intersection into which the vehicle will enter.

[0091] For example, see Figure 8a The first boundary B11 of the first sub-region corresponding to the road segment where the vehicle is located can be determined based on the vehicle's detection distance, such as S or N. This first boundary can also be called the lower boundary of the first sub-region, and the distance between this boundary and the vehicle can be equal to the vehicle's detection distance. For example, the rightmost lane can be determined based on the lane information of the road segment where the vehicle is located. The second boundary B12 of the first sub-region can be obtained by shifting half a lane to the right from the rightmost lane based on the area offset. This second boundary can also be called the right boundary of the first sub-region. Furthermore, if the center point of the intersection the vehicle is about to enter is O, the third boundary B13 of the first sub-region can cross the center point O of the intersection to reach the intersection boundary. This third boundary can be called the left boundary of the first sub-region. The left boundary can also be determined based on the road centerline. The fourth boundary B14 of the first sub-region can be determined based on the center point of the intersection or the upper boundary of the intersection. This fourth boundary can also be called the upper boundary of the first sub-region.

[0092] In this embodiment of the disclosure, the first sub-region corresponding to the road segment where the vehicle is located can be quickly determined based on the vehicle's detection distance, lane information of the road segment where the vehicle is located, area offset, center point of the intersection the vehicle is about to enter, etc., which can improve the speed and rationality of intersection division.

[0093] In one implementation, determining the third detection area as a sub-region of the target road segment includes at least one of the following:

[0094] If the target road segment is a road segment that requires driving straight, turning left, or making a U-turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the center point of the intersection, and the boundary of the target road segment.

[0095] If the target road segment is a road segment that requires a right turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the second boundary of the first sub-region, and the boundary of the target road segment.

[0096] For example, such as Figure 8a As shown, the target road segment is the road segment that requires straight travel. The second sub-region L1 corresponding to the target road segment can have two boundaries that cross the center point of the intersection, and the other two boundaries can coincide with the boundary of the target road segment. The length of the overlapping part between the second sub-region and the target road segment can be determined based on the detection distance, for example, N, which is, for example, 2N.

[0097] For example, such as Figure 8bAs shown, when the target road segment is a road segment that requires a right turn, the second sub-region L2 corresponding to the target road segment can have one boundary B21 that coincides with the second boundary B12 of the first sub-region, i.e., the right boundary, and two boundaries B22 and B23 that coincide with the boundary of the target road segment. Furthermore, based on the detection distance, for example N, the length of the overlapping part between the boundary of the second sub-region and the boundary of the target road segment can be determined, for example, as 2N, thereby determining another boundary B24 of the second sub-region.

[0098] In this embodiment of the disclosure, the first sub-region corresponding to the target road segment to be entered after passing the intersection can be quickly determined based on the vehicle's detection distance, lane information of the road segment where the vehicle is located, area offset, center point of the intersection the vehicle is about to enter, etc., which can improve the speed and rationality of intersection division.

[0099] In one implementation, determining the third detection area as a sub-region of the interfering road segment includes:

[0100] The boundary of the third sub-region corresponding to the interfering road segment is determined based on at least one of the vehicle's detection distance, the obstacle information, and the boundary of the interfering road segment.

[0101] For example, such as Figure 8c As shown, in a sub-region D11 of the interfering road segment, one boundary B31 can coincide with a boundary of the first sub-region of the road where the vehicle is located. Two boundaries B32 and B33 can coincide with the boundary of the interfering road segment. The length of sub-region D11 can be twice the vehicle detection distance, for example, 2N, thus determining another boundary B34. The obstacle information in a sub-region D14 of the interfering road segment is mainly pedestrians, and it can be determined that D14 can cover the sidewalk adjacent to the first sub-region. One boundary B41 of D14 can coincide with the boundary of the sidewalk, and other boundaries can coincide with the boundaries of adjacent areas or road segments.

[0102] In this embodiment of the disclosure, the third sub-region corresponding to the road segment that may interfere with the vehicle can be quickly determined based on the detection distance, obstacle information, and the boundary of the interfering road segment, which can improve the speed and rationality of intersection division and further improve driving safety.

[0103] The examples of determining the various sub-regions of the third detection area in this embodiment are not exhaustive. They can be flexibly selected and adjusted according to the actual driving scenario and driving intention, and no restrictions are imposed here.

[0104] In one embodiment, S101 divides the detection range of the vehicle into multiple detection areas based on the vehicle's position, and further includes: dividing a fourth detection area based on the vehicle's position and the position of the obstacle closest to the vehicle.

[0105] In the embodiments disclosed herein, such as Figure 9a As shown, obstacles around a vehicle can be detected using various sensors. Based on the location of one or more obstacles closest to the vehicle, this can include obstacles closest to the front of the vehicle or those closest to the rear. For example, in the vehicle's direction of travel, there is obstacle A to the left front of the vehicle, 100m away in a straight line; obstacle B directly in front of the vehicle, 50m away in a straight line; and obstacle C to the right rear of the vehicle, 30m away in a straight line. In this case, considering the distance to the closest obstacle, the area between obstacle C and obstacle B can be designated as a fourth detection area. Similarly, the area between obstacle B and obstacle C can also be designated as a fourth detection area. For specific examples, such as... Figure 9b As shown, the lower boundary of the fourth detection area can be determined based on the front of the vehicle, the upper boundary based on the nearest obstacle to the vehicle, and the left and right boundaries based on the boundary of the road segment or lane where the vehicle is located. Furthermore, the upper boundary of the fourth detection area can be determined based on the rear of the vehicle, the lower boundary based on the nearest obstacle to the vehicle, and the left and right boundaries based on the boundary of the road segment or lane where the vehicle is located.

[0106] In this embodiment of the disclosure, by dividing the fourth detection area by the location of the obstacle closest to the vehicle, the detection frequency of the fourth detection area can be set to a higher level, thereby increasing the detection frequency of obstacles in areas closer to the vehicle, which is more conducive to improving driving safety.

[0107] In one example, the area around the vehicle can be divided into different levels, with examples of different levels shown in Table 1:

[0108] Table 1

[0109]

[0110] I. See also Figure 3 The non-responsible area corresponding to LEVEL0 can include areas outside the ROI and outside the map, and area filtering can be performed using map road information.

[0111] 2. The non-responsible region corresponding to LEVEL1 can be a region within the ROI. The calculation methods for M and N in the ROI can be found in the following example.

[0112] III. The primary responsibility areas for LEVEL2 can be divided according to scenario, intent, and obstacle speed, as detailed below. Autonomous driving can be categorized by scenario: intersections and non-intersections. Based on driving intent, it can be divided into: starting, going straight, changing lanes, stopping, and turning. Among these, the primary responsibility areas for lane changing, going straight, stopping, and starting in non-intersection scenarios are as follows: Figure 5a , Figure 5b , Figure 5c and Figure 5d As shown.

[0113] The main responsibility areas for going straight and turning right at intersections are as follows: Figure 6a and Figure 6b As shown.

[0114] In intersection scenarios, the primary responsibility areas for left turns and U-turns at intersections are as follows: Figure 6c and Figure 6d As shown.

[0115] The implementation of the attention mechanism in LEVEL2 is as follows: Figure 10 As shown in the diagram. The driving system's intention mainly includes going straight, changing lanes, turning, and making a U-turn; the intention determines the area requiring attention. The scenario mainly includes intersections and non-intersections; the scenario determines the scope of attention. Obstacle information can include obstacle type and speed, and obstacle information and the vehicle's speed determine the size of the attention area for different obstacles. Based on the intention, scenario, and obstacles, a region attention strategy can be derived. The region attention strategy can include various parameters of the region, such as offset, shape, and size.

[0116] A specific method for calculating the forward detection distance of a vehicle is as follows:

[0117]

[0118] M = N

[0119] Where V is the maximum speed limit of the road, N is the forward detection distance (also known as sight distance), and M is the rear detection distance. M and N can also be different. If V = 60 km / h (16.7 m / s), the forward detection distance is M, and the rear detection distance is N, the size of the ROI in LEVEL1 can be determined based on M and N. The size of some sub-regions in LEVEL2 can also be determined; the specific determination method can be found above. Figure 7a , Figure 7b , Figure 8a , Figure 8b and Figure 8c And related descriptions.

[0120] The above factors can be used to comprehensively determine the current master vehicle's LEVEL1-2 focus area. Alternatively, simpler strategies can be employed to determine the area corresponding to LEVEL2. For example, the area could be circular, with a radius greater than the detection distance N or M+N centered at the intersection. Another example is a circular area with a side length of M or N centered at the intersection.

[0121] IV. LEVEL3 is the area between this vehicle and the nearest obstacle, such as... Figure 9a and Figure 9b As shown. Within each lane, the nearest obstacle to the current vehicle is located in the direction in front of or behind the vehicle. By using an inverted index of obstacles and lane IDs, the nearest obstacle to the vehicle can be quickly found using the lane ID, improving processing efficiency.

[0122] Using the scheme of this disclosure embodiment, based on the above classification, high-frequency calculations of, for example, above 20 Hz can be performed on LEVEL3 obstacles; mid-frequency calculations of, for example, 10 Hz can be performed on LEVEL2 / LEVEL1 regions; and low-frequency calculations of, for example, below 5 Hz can be performed on LEVEL0 regions.

[0123] This disclosure introduces an attention mechanism to perform scene perception and decision-making by region and priority. High-frequency identification is performed on specific high-risk areas to improve responsiveness, security, and intelligence. Simultaneously, low-power computational identification is performed on non-core areas, allowing limited computing power to be applied to the system's core capabilities, thereby reducing and stabilizing system computing power and lowering latency.

[0124] Figure 11 This is a schematic diagram of an obstacle detection device according to an embodiment of the present disclosure. The device may include:

[0125] The division module 1101 is used to divide the detection range of the vehicle into multiple detection areas according to the vehicle's location;

[0126] The acquisition module 1102 is used to acquire the detection frequencies corresponding to the multiple detection areas respectively;

[0127] The detection module 1103 is used to perform obstacle detection on the multiple detection areas according to the detection frequency corresponding to each of the multiple detection areas.

[0128] Figure 12 This is a schematic diagram of an obstacle detection device according to another embodiment of the present disclosure. The device may include one or more features of the obstacle detection device described in the above embodiments. In one possible implementation, the dividing module 1101 includes:

[0129] The calculation submodule 1201 is used to calculate the detection distance of the vehicle based on the road speed limit at the vehicle's location;

[0130] The first determining submodule 1202 is used to determine the region of interest based on the detection distance of the vehicle;

[0131] The first division submodule 1203 is used to divide the vehicle into multiple detection areas based on the region of interest.

[0132] In one possible implementation, the first determining submodule 1202 is further configured to:

[0133] The upper and lower boundaries of the region of interest are determined based on the forward and rear detection distances of the vehicle.

[0134] The left and right boundaries of the region of interest are determined based on the forward detection distance of the vehicle.

[0135] In one possible implementation, the first partitioning submodule 1203 is further configured to:

[0136] The area outside the region of interest on the map and / or the area outside the map is designated as the first detection area.

[0137] In one possible implementation, the first partitioning submodule 1203 is further configured to:

[0138] The area within the region of interest is designated as the second detection region.

[0139] In one possible implementation, such as Figure 12 As shown, the partitioning module 1101 also includes:

[0140] The second determining submodule 1204 is used to determine the driving scenario corresponding to the vehicle based on the vehicle's location;

[0141] Attention submodule 1205 is used to obtain a region attention strategy based on the driving scenario and driving intention of the vehicle, the region attention strategy including at least one of region offset, region shape and region size;

[0142] The second segmentation submodule 1206 is used to segment the third detection region according to the attention strategy of the region.

[0143] In one possible implementation, the attention submodule 1205 is further configured to:

[0144] In the case that the driving scenario is not an intersection scenario, the lane information of the road segment where the vehicle is traveling is determined according to the driving intention of the vehicle, and the left and right boundaries of the third detection area are determined according to the lane information and area offset of the road segment where the vehicle is traveling; the upper and lower boundaries of the third detection area are determined according to the detection range of the vehicle.

[0145] In one possible implementation, the attention submodule 1205 is further configured to:

[0146] In the case of an intersection driving scenario, the target road segment that the vehicle will enter after passing through the intersection is determined based on the vehicle's corresponding driving intention.

[0147] The interfering road section is determined based on the road segment where the vehicle is located, the driving intention, and information on obstacles that may enter the intersection;

[0148] The third detection area is determined to be a sub-region of the road segment where the vehicle is located, the target road segment, and the interfering road segment.

[0149] In one possible implementation, the attention submodule is used to determine a sub-region of the road segment where the vehicle is located within the third detection area, including:

[0150] Based on the detection distance of the vehicle, determine the first boundary of the first sub-region corresponding to the road segment where the vehicle is located;

[0151] The second boundary of the first sub-region is determined based on the lane information and area offset of the road segment where the vehicle is located;

[0152] Determine the third and / or fourth boundaries of the first sub-region based on the center point of the intersection into which the vehicle will enter.

[0153] In one possible implementation, the attention submodule is used to determine a sub-region of the third detection region within the target road segment, including at least one of the following:

[0154] If the target road segment is a road segment that requires driving straight, turning left, or making a U-turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the center point of the intersection, and the boundary of the target road segment.

[0155] If the target road segment is a road segment that requires a right turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the second boundary of the first sub-region, and the boundary of the target road segment.

[0156] In one possible implementation, the attention submodule is used to determine the third detection area as a sub-region of the interfering road segment, including:

[0157] The boundary of the third sub-region corresponding to the interfering road segment is determined based on at least one of the vehicle's detection distance, the obstacle information, and the boundary of the interfering road segment.

[0158] In one possible implementation, such as Figure 12 As shown, the partitioning module 1101 also includes:

[0159] The third division submodule 1207 is used to divide the fourth detection area based on the position of the vehicle and the position of the nearest obstacle to the vehicle.

[0160] The specific functions and examples of each module and submodule of the apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0161] The acquisition, storage, and application 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.

[0162] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0163] Figure 13 A schematic block diagram of an example electronic device 1300 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 assistants, 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.

[0164] like Figure 13 As shown, device 1300 includes a computing unit 1301, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1302 or a computer program loaded from storage unit 1308 into random access memory (RAM) 1303. The RAM 1303 may also store various programs and data required for the operation of device 1300. The computing unit 1301, ROM 1302, and RAM 1303 are interconnected via bus 1304. Input / output (I / O) interface 1305 is also connected to bus 1304.

[0165] Multiple components in device 1300 are connected to I / O interface 1305, including: input unit 1306, such as keyboard, mouse, etc.; output unit 1307, such as various types of monitors, speakers, etc.; storage unit 1308, such as disk, optical disk, etc.; and communication unit 1309, such as network card, modem, wireless transceiver, etc. Communication unit 1309 allows device 1300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0166] The computing unit 1301 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1301 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 1301 performs the various methods and processes described above, such as obstacle detection methods. For example, in some embodiments, the obstacle detection method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1308. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1300 via ROM 1302 and / or communication unit 1309. When the computer program is loaded into RAM 1303 and executed by the computing unit 1301, one or more steps of the obstacle detection method described above may be performed. Alternatively, in other embodiments, the computing unit 1301 may be configured to perform an obstacle detection method by any other suitable means (e.g., by means of firmware).

[0167] 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.

[0168] 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.

[0169] 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.

[0170] 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).

[0171] 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.

[0172] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0173] It should be understood that the various forms of processes shown above can be used to rearrange, 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 disclosed in this disclosure can be achieved, and this is not limited herein.

[0174] 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 principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An obstacle detection method, comprising: Based on the vehicle's location, the detection range of the vehicle is divided into multiple detection zones; Obtain the detection frequency corresponding to each of the multiple detection regions; Obstacle detection is performed on the multiple detection areas according to the detection frequencies corresponding to the multiple detection areas; The method involves dividing the vehicle's detection range into multiple detection areas based on its location, including: determining the driving scenario corresponding to the vehicle based on its location; wherein the driving scenario includes intersection scenarios and non-intersection scenarios; obtaining a region attention strategy based on the driving scenario and the driving intention of the vehicle, wherein the region attention strategy includes at least one of region offset, region shape, and region size; wherein the region attention strategy corresponds to the driving intention, with different driving intentions corresponding to one or more region attention strategies, and the region offset includes the distance the region needs to offset; and dividing a third detection area based on the region attention strategy, wherein the third detection area includes the vehicle's primary responsibility area, which is the area where the vehicle may bear primary responsibility for a traffic accident.

2. The method according to claim 1, wherein, Based on the vehicle's location, the detection range of the vehicle is divided into multiple detection zones, including: Calculate the detection distance of the vehicle based on the road speed limit at the vehicle's location; The region of interest is determined based on the detection distance of the vehicle; Based on the region of interest, the vehicle is divided into multiple detection areas.

3. The method according to claim 2, wherein, Based on the detection distance of the vehicle, the region of interest is determined, including: The upper and lower boundaries of the region of interest are determined based on the forward and rearward detection distances of the vehicle. The left and right boundaries of the region of interest are determined based on the forward detection distance of the vehicle.

4. The method according to claim 2 or 3, wherein, Based on the region of interest, the vehicle is divided into multiple detection regions, including: The area outside the region of interest in the map and / or the area outside the map are designated as the first detection area.

5. The method according to claim 2 or 3, wherein, Based on the region of interest, the vehicle is divided into multiple detection regions, which further includes: The region within the region of interest is divided into a second detection region.

6. The method according to claim 1, wherein, The region attention strategy is obtained based on the driving scenario and driving intention corresponding to the vehicle, including: When the driving scenario is not an intersection scenario, the lane information of the road segment where the vehicle's driving direction is located is determined according to the driving intention of the vehicle, and the left and right boundaries of the third detection area are determined according to the lane information and area offset of the road segment where the driving direction is located; the upper and lower boundaries of the third detection area are determined according to the detection range of the vehicle.

7. The method according to claim 1, wherein, The region attention strategy is obtained based on the driving scenario and driving intention corresponding to the vehicle, including: When the driving scenario is an intersection scenario, the target road segment that the vehicle will enter after passing through the intersection is determined according to the driving intention of the vehicle. The interfering road section is determined based on the road section where the vehicle is located, the driving intention, and information on obstacles that may enter the intersection; The third detection area is determined to be a sub-region of the road segment where the vehicle is located, the target road segment, and the interfering road segment.

8. The method according to claim 7, wherein, Determining the sub-region of the road segment where the vehicle is located in the third detection area includes: Based on the detection distance of the vehicle, determine the first boundary of the first sub-region corresponding to the road segment where the vehicle is located; The second boundary of the first sub-region is determined based on the lane information and area offset of the road segment where the vehicle is located; The third and / or fourth boundaries of the first sub-region are determined based on the center point of the intersection into which the vehicle will enter.

9. The method according to claim 8, wherein, The third detection area is defined as a sub-region of the target road segment, including at least one of the following: When the target road segment is a road segment that requires driving straight, turning left, or making a U-turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the center point of the intersection, and the boundary of the target road segment. If the target road segment is a road segment that requires a right turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the second boundary of the first sub-region, and the boundary of the target road segment.

10. The method according to any one of claims 7 to 9, determining the third detection area as a sub-region of the interfering road segment, comprising: The boundary of the third sub-region corresponding to the interfering road segment is determined based on at least one of the vehicle detection distance, the obstacle information, and the boundary of the interfering road segment.

11. The method according to claim 2 or 3, further comprising dividing the detection range of the vehicle into multiple detection zones based on the vehicle's location: The fourth detection area is defined based on the location of the vehicle and the location of the nearest obstacle.

12. An obstacle detection device, comprising: The segmentation module is used to divide the detection range of the vehicle into multiple detection areas based on the vehicle's location. The acquisition module is used to acquire the detection frequencies corresponding to the plurality of detection regions respectively; The detection module is used to perform obstacle detection on the multiple detection areas according to the detection frequencies corresponding to the multiple detection areas respectively; The segmentation module includes: a second determination submodule, used to determine the driving scenario corresponding to the vehicle based on the vehicle's position; wherein the driving scenario includes intersection scenarios and non-intersection scenarios; an attention submodule, used to obtain a region attention strategy based on the driving scenario corresponding to the vehicle and the driving intention corresponding to the vehicle, wherein the region attention strategy includes at least one of region offset, region shape, and region size; wherein the region attention strategy corresponds to the driving intention, different driving intentions correspond to one or more region attention strategies, and the region offset includes the distance the region needs to offset; and a second segmentation submodule, used to segment a third detection region based on the region attention strategy, wherein the third detection region includes the vehicle's primary responsibility region, which is the region where the vehicle may bear primary responsibility for a traffic accident.

13. The apparatus according to claim 12, wherein, The partitioning module includes: The calculation submodule is used to calculate the detection distance of the vehicle based on the road speed limit at the vehicle's location; The first determining submodule is used to determine the region of interest based on the detection distance of the vehicle; The first segmentation submodule is used to segment the vehicle into multiple detection areas based on the region of interest.

14. The apparatus according to claim 13, wherein, The first determining submodule is further configured to: The upper and lower boundaries of the region of interest are determined based on the forward and rearward detection distances of the vehicle. The left and right boundaries of the region of interest are determined based on the forward detection distance of the vehicle.

15. The apparatus according to claim 13 or 14, wherein, The first segmentation submodule is further configured to segment the area outside the region of interest in the map and / or the area outside the map into a first detection region.

16. The apparatus according to claim 13 or 14, wherein, The first division submodule is further configured to divide the region within the region of interest into a second detection region.

17. The apparatus according to claim 12, wherein, The attention submodule is further configured to, when the driving scenario is a non-intersection scenario, determine the lane information of the road segment where the vehicle's driving direction is located based on the driving intention corresponding to the vehicle, and determine the left and right boundaries of the third detection area based on the lane information and area offset of the road segment where the driving direction is located; and determine the upper and lower boundaries of the third detection area based on the detection range of the vehicle.

18. The apparatus according to claim 12, wherein, The attention submodule is also used for: When the driving scenario is an intersection scenario, the target road segment that the vehicle will enter after passing through the intersection is determined according to the driving intention of the vehicle. The interfering road section is determined based on the road section where the vehicle is located, the driving intention, and information on obstacles that may enter the intersection; The third detection area is determined to be a sub-region of the road segment where the vehicle is located, the target road segment, and the interfering road segment.

19. The apparatus according to claim 18, wherein, The attention submodule is used to determine a sub-region of the road segment where the vehicle is located in the third detection area, including: Based on the detection distance of the vehicle, determine the first boundary of the first sub-region corresponding to the road segment where the vehicle is located; The second boundary of the first sub-region is determined based on the lane information and area offset of the road segment where the vehicle is located; The third and / or fourth boundaries of the first sub-region are determined based on the center point of the intersection into which the vehicle will enter.

20. The apparatus according to claim 19, wherein, The attention submodule is used to determine the sub-regions of the third detection region in the target road segment, including at least one of the following: When the target road segment is a road segment that requires driving straight, turning left, or making a U-turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the center point of the intersection, and the boundary of the target road segment. If the target road segment is a road segment that requires a right turn, the boundary of the second sub-region corresponding to the target road segment is determined based on the vehicle's detection distance, the second boundary of the first sub-region, and the boundary of the target road segment.

21. The apparatus according to any one of claims 18 to 20, wherein the attention submodule is used to determine a sub-region of the third detection area in the interfering road segment, comprising: The boundary of the third sub-region corresponding to the interfering road segment is determined based on at least one of the vehicle detection distance, the obstacle information, and the boundary of the interfering road segment.

22. The apparatus according to claim 13 or 14, wherein the dividing module further comprises: The third division submodule is used to divide the fourth detection area based on the position of the vehicle and the position of the nearest obstacle to the vehicle.

23. An electronic device, comprising: 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 1-11.

24. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-11.

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

26. An autonomous vehicle, comprising: The electronic device according to claim 23.

Images