Method, device and equipment for determining position relationship between vehicle and lane line

Abstract

A method, device and equipment for determining the positional relationship between a vehicle and a lane line are disclosed, and relate to the technical field of intelligent driving. The method comprises: determining an image frame collected when a first vehicle travels in a first lane; determining, based on the image frame, a plurality of spatial position regions in which a second vehicle deviates from a second lane to the first lane; and determining, based on the plurality of spatial position regions, the positional relationship of the second vehicle relative to the first lane in a preset direction. Since the disclosure has high accuracy in identifying the plurality of spatial position regions in which the second vehicle deviates from the second lane to the first lane from the image frame, the positional relationship of the second vehicle relative to the first lane in the preset direction can be more accurately determined according to the plurality of spatial position regions, thereby ensuring the safety of autonomous driving.

Description

Technical Field

[0001] This disclosure relates to the field of driving technology, and in particular to a method, apparatus and equipment for determining the positional relationship between a vehicle and a lane line during driving. Background Technology

[0002] Currently, in Advanced Driving Assistance Systems (ADAS) based on vision, when a vehicle (such as an autonomous vehicle) is driving in a target lane (such as its own lane), it can acquire image frames through its own installed sensors. Then, it identifies the overall position parameters (such as the vehicle's center point, length, width, and orientation) of other vehicles about to enter or partially entering its own lane from adjacent lanes from these image frames. Based on these overall position parameters, it determines the positional relationship of other vehicles relative to its own lane in a preset direction (e.g., perpendicular to the lane line). However, the current recognition accuracy of the overall vehicle position parameters is relatively low, resulting in inaccurate determination of the positional relationship of other vehicles relative to the own lane in the preset direction, which cannot ensure the safety of autonomous driving. Summary of the Invention

[0003] Currently, when calculating the positional relationship of other vehicles relative to the driving lane in a preset direction, the recognition object of the image frame is the overall vehicle parameters of other vehicles. However, the recognition accuracy of the overall vehicle position parameters is low. Therefore, the positional relationship of other vehicles relative to the driving lane in the preset direction calculated based on the recognized overall vehicle position parameters is not accurate enough, which cannot ensure the safety of autonomous driving.

[0004] To address the aforementioned technical problems, this disclosure provides a method, apparatus, electronic device, and storage medium for determining the positional relationship between a vehicle and lane lines during driving, thereby resolving the issue of inaccurate determination of the positional relationship between the vehicle and lane lines.

[0005] The first aspect of this disclosure provides a method for determining the positional relationship between a vehicle and lane lines during driving, comprising:

[0006] Identify the image frames captured when the first vehicle is traveling in the first lane;

[0007] Based on the image frames, multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane are determined;

[0008] Based on the multiple spatial location regions, the positional relationship of the second vehicle relative to the first lane in a preset direction is determined.

[0009] A second aspect of this disclosure provides an apparatus for determining the positional relationship between a vehicle and lane lines during driving, comprising:

[0010] The first determining module is used to determine the image frames captured when the first vehicle is traveling in the first lane;

[0011] The second determining module determines multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane based on the image frame;

[0012] The third determining module determines the positional relationship of the second vehicle relative to the first lane in a preset direction based on the multiple spatial location regions.

[0013] A third aspect of this disclosure provides a computer-readable storage medium storing a computer program for performing the method described in the first aspect for determining the positional relationship between a vehicle and lane lines during driving.

[0014] A fourth aspect of this disclosure provides an electronic device, the electronic device comprising:

[0015] processor;

[0016] Memory used to store the processor's executable instructions;

[0017] The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described in the first aspect for determining the positional relationship between a vehicle and lane lines during driving.

[0018] In this embodiment, since the positional relationship of the second vehicle relative to the first lane in the preset direction is determined based on multiple spatial position regions where the second vehicle shifts from the second lane to the first lane, the accuracy of identifying these multiple spatial position regions from the image frame directly affects the accuracy of the positional relationship of the second vehicle relative to the first lane in the preset direction. These multiple spatial position regions where the second vehicle shifts from the second lane to the first lane are defined portions of the second vehicle's spatial position regions. Identifying these portions from the acquired image frames has high accuracy, far exceeding the accuracy of identifying the overall vehicle position parameters. Therefore, determining the positional relationship of the second vehicle relative to the first lane in the preset direction based on multiple spatial position regions is more accurate and beneficial for improving the safety of intelligent driving. Attached Figure Description

[0019] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0020] Figure 1 This is a schematic diagram illustrating an application scenario of an exemplary embodiment of this disclosure.

[0021] Figure 2 This is a flowchart illustrating an exemplary embodiment of the present disclosure of a method for determining the positional relationship between a vehicle and a lane line during driving.

[0022] Figure 3 This is a schematic diagram of image frames captured by a first vehicle while it is in motion, provided in an exemplary embodiment of this disclosure.

[0023] Figure 4 This is a flowchart illustrating a method for determining the positional relationship between a vehicle and lane lines during driving, provided by another exemplary embodiment of this disclosure.

[0024] Figure 5 This is a flowchart illustrating a method for determining the positional relationship between a vehicle and lane lines during driving, provided in yet another exemplary embodiment of this disclosure.

[0025] Figure 6A This is a flowchart illustrating a method for determining the positional relationship between a vehicle and lane lines during driving, provided in yet another exemplary embodiment of this disclosure.

[0026] Figure 6B This is a flowchart illustrating a method for determining a first positional relationship between a wheel frame position region and a first lane in a preset direction, provided by an exemplary embodiment of this disclosure.

[0027] Figure 7 This is a schematic diagram of the composition of a device for determining the positional relationship between a vehicle and a lane line during driving, provided by an exemplary embodiment of this disclosure.

[0028] Figure 8 This is a schematic diagram of the composition of a device for determining the positional relationship between a vehicle and a lane line during driving, provided by another exemplary embodiment of this disclosure.

[0029] Figure 9 This is a schematic diagram of the composition of a device for determining the positional relationship between a vehicle and a lane line during driving, provided in another exemplary embodiment of this disclosure.

[0030] Figure 10 This is a schematic diagram of the composition of a device for determining the positional relationship between a vehicle and a lane line during driving, provided in another exemplary embodiment of this disclosure.

[0031] Figure 11 This is a schematic diagram of the composition of a device for determining the positional relationship between a vehicle and a lane line during driving, provided in another exemplary embodiment of this disclosure.

[0032] Figure 12 This is a structural diagram of an electronic device provided in an exemplary embodiment of this disclosure. Detailed Implementation

[0033] To explain this disclosure, exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0034] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "multiple" means two or more. "A and / or B" includes the following three combinations: only A, only B, and a combination of A and B.

[0035] Application Overview

[0036] In this disclosure, the vehicle may be referred to as a self-driving vehicle, and the lane in which the self-driving vehicle travels may be referred to as a self-driving lane. In some examples, the self-driving vehicle may be a vehicle in an intelligent driving state. For example, the self-driving vehicle may be a vehicle in a normal assisted driving state or a vehicle in an advanced assisted driving state.

[0037] In an ADAS system, a vehicle in intelligent driving mode, while traveling in its lane, can acquire image frames of its driving environment through its sensors during the visual perception phase. From these image frames, it can identify the overall position parameters of other vehicles (such as at least one vehicle other than its own). Then, based on these overall position parameters, it determines the positional relationship of the other vehicles relative to its lane in a preset direction. Finally, during the decision-making phase, this positional relationship is used as the basis for the vehicle's intelligent driving decisions, and corresponding actions are executed. For example, the vehicle can perform actions such as braking, deceleration, not braking, or lane changing based on the positional relationship of the other vehicles relative to its lane in a preset direction.

[0038] For example, while the vehicle is traveling in its own lane, other vehicles may enter the lane from other lanes adjacent to it. The image frames captured by the vehicle may include other vehicles that have partially entered or are about to enter the lane from other lanes; these other vehicles may also be referred to as intruding vehicles. The positional relationship of the intruding vehicle relative to the lane in a preset direction includes the distance of the intruding vehicle relative to the lane line closest to it in the preset direction (e.g., a direction perpendicular to the lane line of the lane), which may also be referred to as the intruding vehicle's line crossing amount. The line crossing amount of the intruding vehicle can be determined as a positive or negative value based on the positional relationship between the intruding vehicle and the lane.

[0039] In some examples, the line crossing amount of an intruding vehicle that has partially entered the lane can be considered a positive value, and the line crossing amount of an intruding vehicle about to enter the lane can be considered a negative value. Alternatively, the line crossing amount of an intruding vehicle that has partially entered the lane can be considered a negative value, and the line crossing amount of an intruding vehicle about to enter the lane can be considered a positive value. This disclosure does not limit this approach; the following embodiments are exemplified by considering the line crossing amount of an intruding vehicle that has partially entered the lane as a positive value and the line crossing amount of an intruding vehicle about to enter the lane as a negative value.

[0040] Exemplary scenario

[0041] Figure 1 This is a schematic diagram illustrating an application scenario provided by an exemplary embodiment of this disclosure. For example... Figure 1 As shown, the lane indicated by the arrow is lane 10, where the vehicle is currently traveling. Lane 10 has lane lines 101 and 102. The lanes adjacent to lane 10 are lanes 11 and 12. Lane 11 has lane lines 111 and 101, and lane 12 has lane lines 102 and 121.

[0042] For example, such as Figure 1 As shown, vehicle 13 is about to enter lane 10 from lane 11. The positional relationship of vehicle 13 relative to lane 10 in a preset direction can be the distance 131 from the nearest point of vehicle 13 to lane line 101. In some examples, this distance 131 can be referred to as the line crossing amount 131 of vehicle 13. Since vehicle 13 is about to enter lane 10, the line crossing amount 131 of vehicle 13 is a negative value.

[0043] like Figure 1As shown, vehicle 14 has partially entered lane 10. The positional relationship of vehicle 14 relative to lane 10 in a preset direction can be the distance 141 from the nearest point of vehicle 14 to lane line 101. In some examples, this distance 141 can be referred to as the line crossing amount 141 of vehicle 14. Since vehicle 14 has partially entered lane 10, the line crossing amount of vehicle 14 is a positive value.

[0044] like Figure 1 As shown, vehicle 15 has partially entered lane 10. The positional relationship of vehicle 15 relative to lane 10 in a preset direction can be the distance 151 from the nearest point of vehicle 15 to lane line 102. In some examples, this distance 151 may also be referred to as the line crossing amount 151 of vehicle 15. Since vehicle 15 has partially entered lane 10, the line crossing amount of vehicle 15 is a positive value.

[0045] like Figure 1 As shown, taking the image frame acquired by the vehicle as an example, including vehicle 14, the overall position parameters of vehicle 14 can be identified from the image frame, and the positional relationship of vehicle 14 relative to lane 10 in a preset direction can be determined based on the overall position parameters of vehicle 14.

[0046] Because the accuracy of identifying the overall position parameters of vehicle 14 from image frames is relatively low, the accuracy of determining the positional relationship of vehicle 14 relative to lane 10 in the preset direction based on the overall position parameters of vehicle 14 is also low, affecting the safety of intelligent driving.

[0047] To address the aforementioned technical issues, this disclosure provides a method for determining the positional relationship between a vehicle and lane lines during driving. This method identifies multiple spatial location regions of a second vehicle from acquired image frames and determines the positional relationship of the second vehicle relative to a first lane in a preset direction based on these multiple spatial location regions. Because this disclosure achieves high accuracy in identifying multiple spatial location regions from image frames, it can accurately determine the positional relationship of the second vehicle relative to the first lane in a preset direction based on these multiple spatial location regions, thus ensuring the safety of autonomous driving.

[0048] Exemplary methods

[0049] Figure 2 This is a flowchart illustrating a method for determining the positional relationship between a vehicle and lane lines during driving, provided in an exemplary embodiment of this disclosure. This embodiment can be applied to vehicles; to distinguish this vehicle from other vehicles, this embodiment uses the term "self-vehicle" or "first vehicle" in its description. Figure 2 As shown, it may include the following steps 201 to 203.

[0050] Step 201: Determine the image frames captured when the first vehicle is traveling in the first lane.

[0051] For example, the first vehicle may be referred to as the "self-vehicle". The first lane is the lane currently being traveled by the self-vehicle, and may also be referred to as the "self-lane".

[0052] In some embodiments, image frames can be acquired by multiple image detection devices distributed at different locations on the vehicle. These image detection devices include cameras, and the type of image detection device for acquiring image frames is not limited in this disclosure.

[0053] For example, when the vehicle is in motion, the captured image frame may include at least one object from the vehicle's driving environment, such as the vehicle's speed, the location of the cameras, and the camera's accuracy. For instance, the image frame may include at least one object from the following categories: other vehicles, pedestrians, cyclists, green belts, lane lines, traffic lights, and turn arrows. This disclosure does not limit the type of object included in the image frame; the objects included in the captured image frame may differ depending on the vehicle's driving environment.

[0054] Step 202: Based on the image frames, determine multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane.

[0055] For example, the second vehicle can be any vehicle other than the first vehicle that has partially entered or is about to enter the first lane. The second lane is the lane in which the second vehicle travels, and the lane in which the second vehicle travels can be an adjacent lane to the first lane. For example, refer to... Figure 1 As shown, when the second vehicle is vehicle 13 or vehicle 14, the second lane is lane 11 to the left of lane 10. When the second vehicle is vehicle 15, the second lane is lane 12 to the right of lane 10.

[0056] In some examples, the image frame acquired by the first vehicle may or may not include the first lane. For instance, before performing step 202 above, the inclusion relationship between the image frame and the first lane can be determined. If the inclusion relationship indicates that the image frame includes the first lane, then step 202 is performed.

[0057] For example, the image frame can be identified using a lane recognition model. If the first lane is identified, it is determined that the image frame includes the first lane, and step 202 is continued. If the first lane cannot be identified from the image frame, it is determined that the image frame does not include the first lane, and step 202 is not executed. In some examples, if the first lane cannot be identified from the image frame, the image frame is discarded, and image frames are continuously acquired until an image frame containing the first lane is acquired, and step 202 is continued.

[0058] The embodiments of this disclosure, by determining multiple spatial location regions of the second vehicle shifting from the second lane to the first lane based on the image frame when the image frame includes the first lane, do not require processing for image frames that do not include the first lane. In this way, the efficiency of determining the positional relationship of the second vehicle relative to the first lane in a preset direction can be improved.

[0059] For example, multiple spatial location regions can refer to at least two location regions in the second vehicle in the image coordinate system used to represent the positional relationship between the second vehicle and the first lane. In some examples, the multiple spatial location regions of the second vehicle may include the wheel frame location region of the second vehicle, the wheel contact point location region of the second vehicle, and the segmentation location region of the second vehicle. This disclosure does not limit the specific locations of the multiple spatial location regions in the second vehicle; the following embodiments use the inclusion of the wheel frame location region, the wheel contact point location region, and the segmentation location region as examples for illustrative purposes.

[0060] In some implementations, multiple neural network models can be used to identify multiple spatial location regions from an image frame where the second vehicle has shifted from the second lane to the first lane. These multiple neural network models can be pre-trained neural network models designed to identify the corresponding spatial location regions.

[0061] Step 203: Based on multiple spatial location regions, determine the positional relationship of the second vehicle relative to the first lane in a preset direction.

[0062] For example, the preset direction can be perpendicular to the direction of the first lane. The positional relationship of the second vehicle relative to the first lane in the preset direction can be the distance between a spatial location area and the first lane in the preset direction. Taking the lane line of the first lane as a vertical lane line as an example, the positional relationship of the second vehicle relative to the first lane in the preset direction can be the lateral distance of the spatial location area of ​​the second vehicle relative to the lane line of the first lane in the horizontal direction.

[0063] In some implementations, determining the positional relationship of the second vehicle relative to the first lane in a preset direction based on multiple spatial location regions may include calculating the distance between each spatial location region and the first lane in the preset direction, and determining the positional relationship of the second vehicle relative to the first lane in the preset direction based on the obtained distance values.

[0064] In this embodiment, multiple spatial location regions where the second vehicle shifts from the second lane to the first lane are identified from an image frame, and the positional relationship of the second vehicle relative to the first lane in a preset direction is determined based on these multiple spatial location regions. Since identifying multiple spatial location regions from an image frame represents only a portion of the vehicle's position, the identification accuracy is higher compared to identifying the overall position parameters of the vehicle. Consequently, the positional relationship between the second vehicle and the first lane determined based on these multiple spatial locations is also more accurate, thus improving the safety of intelligent driving.

[0065] like Figure 3 As shown above, in the above Figure 2 Based on the illustrated embodiment, step 202 may include steps 2021 to 2023 as follows.

[0066] Step 2021: Determine the wheel frame location region of the second vehicle from the image frame using the wheel frame recognition model.

[0067] For example, the wheel frame recognition model can be a pre-trained neural network model for recognizing wheel frame location regions from image frames. The wheel frame location regions of the second vehicle can include front wheel frame location regions and / or rear wheel frame location regions, where the front wheel frame location region includes the area where the front wheels of the second vehicle are located, and the rear wheel frame location region includes the area where the rear wheels of the second vehicle are located.

[0068] In some implementations, when the first vehicle acquires an image frame including the first lane in real time, the input image frame can be processed by a wheel frame recognition model to obtain the wheel frame position area of ​​the second vehicle.

[0069] like Figure 4 As shown, if the second vehicle is vehicle 401 about to enter lane 400, then the wheel rim position area of ​​vehicle 401 includes the front wheel rim position area 402 and the rear wheel rim position area 403 of vehicle 401. The front wheel rim position area 402 consists of multiple position points on the front wheel rim, which may include position points on the lower edge of the front wheel rim (e.g., ...). Figure 4 The midpoint of the lower edge of the front wheel rim is located at position 4021. The rear wheel rim position area 403 consists of multiple position points on the rear wheel rim, which may include position points on the lower edge of the rear wheel rim (e.g., ...). Figure 4 The midpoint of the lower edge of the rear wheel rim is 4031.

[0070] Step 2022: Determine the wheel contact location area of ​​the second vehicle from the image frame using the wheel contact identification model.

[0071] For example, the wheel contact recognition model can be a pre-trained neural network model for identifying wheel contact location regions from image frames. The wheel contact location regions of the second vehicle can include front wheel contact location regions and / or rear wheel contact location regions. The front wheel contact location region consists of multiple front wheel contact location points, and the rear wheel contact location region consists of multiple rear wheel contact location points.

[0072] In some implementations, when the first vehicle acquires an image frame including the first lane in real time, the image frame can be processed by a wheel ground contact recognition model to obtain the wheel ground contact location area of ​​the second vehicle.

[0073] like Figure 4 As shown, if the second vehicle is vehicle 401 about to enter lane 400, then the wheel contact area of ​​the second vehicle includes the front wheel contact area of ​​vehicle 401 (e.g., Figure 4 The front grounding wire of vehicle 401 in contact with the ground and the grounding area of ​​the rear wheels (such as...) Figure 4 The rear grounding wire of vehicle 401 (the rear wheel of the vehicle in contact with the ground). This front grounding wire consists of multiple front wheel grounding points, which may include... Figure 4 The midpoint of the front grounding wire is shown at position 404. The rear grounding wire consists of multiple rear wheel grounding points, which may include... Figure 4 The midpoint of the rear grounding wire is shown at position 405.

[0074] Step 2023: Determine the segmentation location region of the second vehicle from the image frame using the segmentation location recognition model.

[0075] For example, the segmented location region of the second vehicle can be the location region in the image frame that segments the vehicle body and the driving area of ​​the second vehicle. (Reference) Figure 4 As shown, if the second vehicle is vehicle 401 about to enter lane 400, then the segmented position region of the second vehicle can be the intersection line 406 between the vehicle body of vehicle 401 and the driving area of ​​vehicle 401 (e.g., the lane area in the image frame). Figure 4 (The bold black solid line shown). This intersection line 406 can be composed of multiple dividing points.

[0076] In some embodiments, the segmentation location recognition model may be a pre-trained neural network model used to identify the intersection location region between the region of the intruding vehicle in the image frame and the drivable area of ​​the intruding vehicle.

[0077] In some implementations, when the first vehicle acquires an image frame including the first lane in real time, the image frame can be processed by a segmentation location recognition model to obtain the segmentation location region of the second vehicle.

[0078] In this embodiment, the wheel frame position region, wheel ground contact position region, and segmentation position region are determined from image frames acquired from the first vehicle using a wheel ground contact recognition model, a wheel frame recognition model, and a segmentation position recognition model, respectively. Identifying the wheel frame position region, wheel ground contact position region, and segmentation position region using different models improves the recognition accuracy of multiple spatial position regions. Furthermore, these wheel frame position region, wheel ground contact position region, and segmentation position region more accurately represent the position of the second vehicle compared to the overall vehicle position parameters in the prior art.

[0079] like Figure 5 As shown above, in the above Figure 2 Based on the illustrated embodiment, step 203 may include steps 2031 to 2033 as follows.

[0080] Step 2031: Determine the lane lines of the first lane based on the image frame.

[0081] For example, the lane line of the first lane may be composed of multiple lane position points located on the first lane, and the coordinates of the multiple lane position points may be coordinate data in the image coordinate system.

[0082] In some implementations, determining the lane lines of the first lane based on an image frame can be achieved by identifying the lane lines of the first lane from the image frame using a lane line recognition model.

[0083] Step 2032: Determine the lane line parametric equations based on the lane lines of the first lane.

[0084] For example, the lane line parameter equations can be lane line parameter equations in a first vehicle coordinate system.

[0085] In some implementations, multiple lane line fitting points can be selected from the lane lines of the first lane, and the lane line parameter equation can be obtained by fitting multiple lane line fitting points.

[0086] For example, if multiple lane fitting points in the lane line of the first lane are data in the image coordinate system and the lane line parameter equation is an equation in the first vehicle coordinate system, then before fitting multiple lane fitting points, it is necessary to transform the coordinates of multiple lane line fitting points from the image coordinate system to the first vehicle coordinate system, and then obtain the lane line parameter equation by performing coordinate fitting on multiple lane line fitting points in the first vehicle coordinate system.

[0087] Step 2033: Based on multiple spatial location regions and lane line parameter equations, determine the positional relationship of the second vehicle relative to the first lane in a preset direction.

[0088] In some implementations, step 2033 may include: calculating the corresponding position of each spatial location region on the lane line of the first lane using lane line parametric equations, and calculating the positional relationship of the second vehicle relative to the first lane in a preset direction based on each spatial location region and its corresponding position on the lane line of the first lane.

[0089] In this embodiment, since the lane line parameter equation is determined based on the lane line of the first vehicle identified from the real-time acquired image frame, it can more accurately reflect the lane line of the first vehicle at the current moment. Therefore, the positional relationship of the second vehicle relative to the first lane in the preset direction, determined based on multiple spatial location regions and the lane line parameter equation, is also more accurate.

[0090] like Figure 6A As shown above, in the above Figure 5 Based on the illustrated embodiment, step 2033 may include steps 601 to 604 as follows.

[0091] Step 601: Based on the wheel frame position region and lane line parameter equations in multiple spatial position regions, determine the first positional relationship of the wheel frame position region relative to the first lane in a preset direction.

[0092] In some implementations, the wheel frame position region may include a first wheel frame position and a second wheel frame position. For example, both the first wheel frame position and the second wheel frame position are positions in an image coordinate system.

[0093] In some examples, reference Figure 4 As shown, if the second vehicle is vehicle 401 about to enter lane 400, then the position of the first wheel rim can be... Figure 4 The midpoint of the lower edge of the front wheel rim of the vehicle 401 shown is 4021; the position of the second wheel rim can be... Figure 4 The lower edge of the rear wheel rim of the vehicle 401 shown is located at position 4031.

[0094] In some embodiments, such as Figure 6B As shown, step 601 above may include steps 6011 to 6013.

[0095] Step 6011: Based on the first wheel frame position in the wheel frame position region and the lane line parameter equation, determine the first distance of the first wheel frame position relative to the first lane in a preset direction.

[0096] In some examples, a reference position corresponding to the position of the first wheel frame can be determined first on the lane line of the first lane based on the lane line parametric equation, and then the first distance of the position of the first wheel frame relative to the first lane in a preset direction can be determined based on the reference position and the position of the first wheel frame.

[0097] For example, the reference position corresponding to the first wheel rim position can be the position on the lane line of the first lane that is closest to the first wheel rim position. This reference position can be the position in the first vehicle coordinate system.

[0098] In some examples, since the position of the first wheel frame is in the image coordinate system, a coordinate system transformation is needed to determine the reference position using the lane line parametric equations. Therefore, when determining the reference position corresponding to the first wheel frame position, the first wheel frame position can be first converted to coordinates in the first vehicle coordinate system; then, the lane line parametric equations are used to calculate the coordinates of the first wheel frame position in the first vehicle coordinate system to obtain the reference position corresponding to the first wheel frame position. For example, a first proportional relationship can exist between the image coordinate system and the first vehicle coordinate system. Based on this first proportional relationship, the position coordinates of the first wheel frame position can be transformed to convert the coordinates in the image coordinate system to coordinates in the first vehicle coordinate system.

[0099] For example, after obtaining the position coordinates of the first wheel frame in the first vehicle coordinate system, the position coordinates of the first wheel frame can be substituted into the lane line parameter equation to calculate the coordinates of the reference position corresponding to the position of the first wheel frame.

[0100] In some implementations, after determining the reference position, the coordinates of the reference position can be subtracted from the coordinates of the first wheel frame position to obtain the first distance of the first wheel frame position relative to the first lane in a preset direction.

[0101] For example, the vertical coordinate of the reference position can be subtracted from the vertical coordinate of the position of the first wheel frame in the first vehicle coordinate system, and the difference in the resulting vertical coordinate can be determined as the first distance of the position of the first wheel frame relative to the first lane in a preset direction.

[0102] refer to Figure 4 As shown, taking the second vehicle as vehicle 401 about to enter lane 400, and the first wheel rim position as the midpoint 4021 of the lower edge of the front wheel rim of vehicle 401, the closest point to the first wheel rim position can be determined on the lane line 4000 of lane 400 as 4001. That is, the reference position corresponding to the first wheel rim position 4021 is position 4001 on the lane line 4000.

[0103] For example, since vehicle 401 is about to enter lane 400, the distance D1 of the first wheel rim position relative to lane 400 in a preset direction is negative.

[0104] In this embodiment, the first distance of the first wheel frame relative to the first lane in a preset direction is determined by a reference position on the lane line corresponding to the position of the first wheel frame and the position of the first wheel frame. Therefore, when the accuracy of the identified first wheel frame position and the accuracy of the lane line parameter equation are both high, the accuracy of the reference position on the lane line corresponding to the position of the first wheel frame is also high. Consequently, the first distance of the first wheel frame relative to the first lane in the preset direction can be determined relatively accurately.

[0105] Step 6012: Based on the second wheel frame position in the wheel frame position region and the lane line parameter equation, determine the second distance of the second wheel frame position relative to the first lane in a preset direction.

[0106] In some embodiments, a reference position corresponding to the position of the second wheel frame can be determined first on the lane line of the first lane based on the lane line parameter equation; then, based on the reference position corresponding to the position of the second wheel frame and the position of the second wheel frame, a second distance of the position of the second wheel frame relative to the first lane in a preset direction can be determined.

[0107] It should be noted that the method for determining the second distance in step 6012 is similar to the method for determining the first distance in step 6011, and will not be repeated here.

[0108] refer to Figure 4 As shown, if the second vehicle is vehicle 401 that is about to enter lane 400, and the position of the second wheel rim is the midpoint of the lower edge of the rear wheel rim of vehicle 401 at position 4031, the position point closest to the position of the first wheel rim can be determined on the lane line 4000 of the first lane 400 as 4002 (the reference position 4002 corresponding to the position of the second wheel rim).

[0109] For example, since vehicle 401 is about to enter lane 400, the distance D2 of the second wheel rim position relative to the lane in the preset direction is also negative.

[0110] Step 6013: Determine the first positional relationship based on the relationship between the first distance and the second distance.

[0111] For example, the first positional relationship is a first target distance of the wheel frame position region relative to the first lane in a preset direction. This first target distance is determined based on a first distance and a second distance. For instance, the larger of the first distance and the second distance can be determined as the first target distance.

[0112] In some examples, if the first distance is greater than the second distance, then the first distance is determined as the first target distance; if the first distance is less than the second distance, then the second distance is determined as the first target distance. For example, see reference... Figure 4As shown, since -D1 is greater than -D2, -D1 is determined as the first target distance of the wheel frame position area relative to the first lane in the preset direction.

[0113] In this embodiment, a first distance is determined by the position of the first wheel rim and the wheel line parameter equation, and a second distance is determined by the position of the second wheel rim and the wheel line parameter equation. When the accuracy of the first wheel rim position, the second wheel rim position, and the wheel line parameter equation is high, the accuracy of both the first and second distances is also high. Furthermore, when determining the first positional relationship based on the magnitude relationship between the first and second distances, since there is no superposition between the errors of the first and second distances, the accuracy of the first positional relationship determined based on both the first and second distances is also high. Therefore, based on the first positional relationship, the positional relationship of the second vehicle relative to the first lane in a preset direction can be accurately determined.

[0114] Step 602: Based on the wheel contact point region and lane line parameter equations in multiple spatial location regions, determine the second positional relationship of the wheel contact point region relative to the first lane in a preset direction.

[0115] In some implementations, the second positional relationship is a second target distance of the wheel contact point region relative to the first lane in a preset direction. The wheel contact point region may include a first wheel contact point and a second wheel contact point. Exemplarily, both the first wheel contact point and the second wheel contact point are positions in an image coordinate system.

[0116] In some implementations, step 602 may include: determining a third distance in a preset direction between the first wheel contact position and the first lane line parameter equation in the wheel contact position region; determining a fourth distance in a preset direction between the second wheel contact position and the first lane line parameter equation in the wheel contact position region; and determining a second positional relationship based on the magnitude relationship between the third distance and the fourth distance.

[0117] In some examples, a reference position corresponding to the first wheel contact point can be determined on the lane line of the first lane based on the lane line parametric equation; based on this reference position and the first wheel contact point, a third distance of the first wheel contact point relative to the first lane in a preset direction can be determined.

[0118] In some examples, a reference position corresponding to the second wheel contact point can be determined on the lane line of the first lane based on the lane line parametric equation; based on this reference position and the second wheel contact point, a fourth distance of the second wheel contact point relative to the first lane in a preset direction can be determined.

[0119] It should be noted that the methods for determining the third and fourth distances are similar to those for determining the first distance in step 6011, and will not be repeated here.

[0120] refer to Figure 4 As shown, if the second vehicle is vehicle 401 that is about to enter lane 400, the first wheel contact point can be the midpoint of the front contact line in sight 404, and the second wheel contact point can be the midpoint of the rear contact line in sight 405.

[0121] In some examples, because the recognition accuracy of the wheel frame recognition model used to identify the wheel frame position and the wheel grounding recognition model used to identify the wheel grounding position is different, the midpoint position 404 of the visible front grounding line of sight may or may not coincide with the midpoint position 4021 of the lower frame of the aforementioned front wheel frame; the midpoint position 405 of the visible rear grounding line of sight may or may not coincide with the midpoint position 4031 of the lower frame of the aforementioned rear wheel frame. Figure 4 The following example illustrates the situation: the midpoint of the ground visible line of sight 404 coincides with the midpoint of the lower edge of the front wheel rim 4021, and the midpoint of the ground visible line of sight 405 coincides with the midpoint of the lower edge of the rear wheel rim 4031.

[0122] refer to Figure 4 As shown, since the midpoint of the visible front ground contact line 404 is the same as the midpoint of the lower frame of the front wheel rim 4021, and the midpoint of the visible rear ground contact line 405 is the same as the midpoint of the lower frame of the rear wheel rim 4031, the reference position corresponding to the first wheel ground contact position is also position 4001, the reference position corresponding to the second wheel ground contact position is also position 4002, the third distance of the first wheel ground contact position relative to the first lane in the preset direction is also D1, and the fourth distance of the second wheel ground contact position relative to the first lane in the preset direction is also D2.

[0123] For example, the second positional relationship is determined in a similar way to the first positional relationship. In some examples, the larger of the third and fourth distances can be determined as the second positional relationship (also referred to as the second target distance).

[0124] In this embodiment, a third distance is determined by the ground contact position of the first wheel rim and the wheel line parameter equation, and a fourth distance is determined by the ground contact position of the second wheel and the wheel line parameter equation. Therefore, when the accuracy of the first wheel ground contact position, the second wheel ground contact position, and the wheel line parameter equation is high, the accuracy of both the third and fourth distances is also high. Simultaneously, when determining the second positional relationship based on the magnitude relationship between the third and fourth distances, since there is no superposition of errors in the third distance, when the accuracy of both the third and fourth distances is high, the accuracy of the second positional relationship determined based on the first and second distances is also high. Thus, based on the second positional relationship, the positional relationship of the second vehicle relative to the first lane in a preset direction can be accurately determined.

[0125] Step 603: Based on the segmented position region and lane line parameter equations among multiple spatial position regions, determine the third positional relationship of the segmented position region relative to the first lane in a preset direction.

[0126] In some implementations, the third positional relationship is the third target distance of the segmented position region relative to the first lane in a preset direction. The segmented position region may include at least two segmented positions. For example, both of the at least two segmented positions are positions in an image coordinate system.

[0127] In some examples, reference Figure 4 As shown, if the second vehicle is vehicle 401 that is about to enter lane 400, then at least two dividing points can be at least two dividing point points on the intersection line 406.

[0128] In some implementations, step 603 may include: determining at least two fifth distances relative to the first lane in a preset direction for at least two segmented locations in the segmented location region and lane line parameter equations; and determining a third positional relationship based on the magnitude relationship between the at least two fifth distances.

[0129] In some examples, at least two reference positions corresponding to at least two segmentation positions can be determined first on the lane line of the first lane based on the lane line parametric equation; then, based on the at least two reference positions and the at least two segmentation positions, at least two fifth distances relative to the first lane in a preset direction are determined. This disclosure does not limit the specific locations of the at least two segmentation positions; the at least two segmentation positions can be any at least two locations on the intersection line 406, and the at least two reference positions corresponding to the at least two segmentation positions can also be any at least two locations on the lane line.

[0130] It should be noted that the method for determining the fifth distance is similar to the method for determining the first distance in step 6011, and will not be repeated here.

[0131] For example, the third positional relationship is determined in the same way as the first and / or second positional relationships. In some examples, the larger of at least two fifth distances can be determined as the third positional relationship (also referred to as the third target distance).

[0132] In this embodiment, at least two fifth distances are determined using at least two segmentation positions and wheel line parameter equations. When the accuracy of the at least two segmentation positions and wheel line parameter equations is high, the accuracy of the at least two fifth distances is also high. When determining the first positional relationship based on the magnitude relationship of the at least two fifth distances, since there is no superposition of errors between the at least two fifth distances, and when the at least two fifth distances are both high, the third positional relationship can be determined more accurately based on the at least two fifth distances. Therefore, based on the third positional relationship, the positional relationship of the second vehicle relative to the first lane in a preset direction can be accurately determined.

[0133] Step 604: Based on the first positional relationship, the second positional relationship, and the third positional relationship, determine the positional relationship of the second vehicle relative to the first lane in a preset direction.

[0134] In some implementations, step 604 may include: processing a first target distance representing a first positional relationship, a second target distance representing a second positional relationship, and a third target distance representing a third positional relationship, and determining the processing result as the positional relationship of the second vehicle relative to the first lane in a preset direction.

[0135] In some examples, the average value of the first target distance, the second target distance, and the third target distance can be calculated, and the average value can be determined as the positional relationship of the second vehicle relative to the first lane in a preset direction.

[0136] In this embodiment, the positional relationship of the second vehicle relative to the first lane in a preset direction is determined by a first positional relationship, a second positional relationship, and a third positional relationship. Since the first positional relationship corresponds to the wheel rim position area, the second positional relationship corresponds to the wheel contact position area, and the third positional relationship corresponds to the segmented position area, the positional relationship of the second vehicle relative to the first lane in the preset direction comprehensively considers the first positional relationship corresponding to the wheel rim position area, the second positional relationship corresponding to the wheel contact position area, and the third positional relationship corresponding to the segmented position area. Because the identified wheel rim position area, wheel contact position area, and segmented position area have high precision, the determined first, second, and third positional relationships also have high precision, thereby improving the accuracy of determining the positional relationship of the second vehicle relative to the first lane in the preset direction and contributing to improved safety in intelligent driving.

[0137] Exemplary device

[0138] Figure 7 This is a schematic diagram illustrating the structural composition of a device for determining the positional relationship between a vehicle and lane lines during driving, provided by an exemplary embodiment of this disclosure. Figure 7 As shown, the device 70 for determining the positional relationship between a vehicle and a lane line during driving includes a first determining module 701, a second determining module 702, and a third determining module 703.

[0139] The first determining module 701 is used to determine the image frames collected when the first vehicle is traveling in the first lane.

[0140] The second determining module 702 is used to determine multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane based on image frames.

[0141] The third determining module 703 is used to determine the positional relationship of the second vehicle relative to the first lane in a preset direction based on multiple spatial location regions.

[0142] In some embodiments of this disclosure, such as Figure 8 As shown above, in the above Figure 7 Based on the embodiment shown, the second determining module 702 may include: a first determining unit 7021, a second determining unit 7022 and a third determining unit 7023.

[0143] The first determining unit 7021 is used to determine the wheel frame position region of the second vehicle from the image frame through the wheel frame recognition model.

[0144] The second determining module 7022 is used to determine the wheel grounding location area of ​​the second vehicle from the image frame through the wheel grounding recognition model.

[0145] The third determining unit 7023 is used to determine the segmentation location region of the second vehicle from the image frame through the segmentation location recognition model.

[0146] Among them, multiple spatial location areas include the wheel frame location area, the wheel ground contact location area, and the segmentation location area.

[0147] In some embodiments of this disclosure, such as Figure 9 As shown above, in the above Figure 7 Based on the embodiment shown, the third determining module 703 may include: a fourth determining unit 7031, a fifth determining unit 7032 and a sixth determining unit 7033.

[0148] The fourth determining unit 7031 is used to determine the lane lines of the first lane based on the image frame.

[0149] The fifth determining unit 7032 is used to determine the lane line parametric equation based on the lane line of the first lane.

[0150] The sixth determining unit 7033 is used to determine the positional relationship of the second vehicle relative to the first lane in a preset direction based on multiple spatial location regions and lane line parameter equations.

[0151] In some embodiments of this disclosure, such as Figure 10 As shown above, in the above Figure 9 Based on the illustrated embodiment, the sixth determining unit 7033 may include:

[0152] The first sub-determining unit 1001, the second sub-determining unit 1002, the third sub-determining unit 1003, and the fourth sub-determining unit 1004.

[0153] The first sub-determining unit 1001 is used to determine the first positional relationship of the wheel frame position area relative to the first lane in a preset direction based on the wheel frame position area and lane line parameter equation in multiple spatial position areas.

[0154] The second sub-determining unit 1002 is used to determine the second positional relationship of the wheel contact position area relative to the first lane in a preset direction based on the wheel contact position area and lane line parameter equation among multiple spatial position areas.

[0155] The third sub-determination unit 1003 is used to determine the third positional relationship of the segmented position region relative to the first lane in a preset direction based on the segmented position region and lane line parameter equation among multiple spatial position regions.

[0156] The fourth sub-determination unit 1004 is used to determine the positional relationship of the second vehicle relative to the first lane in a preset direction based on the first positional relationship, the second positional relationship and the third positional relationship.

[0157] Specifically, the first sub-determining unit 1001 is used to determine a first distance in a preset direction between the first wheel frame position and the first lane line parameter equation in the wheel frame position region; determine a second distance in a preset direction between the second wheel frame position and the first lane line parameter equation in the wheel frame position region; and determine a first positional relationship based on the magnitude relationship between the first distance and the second distance.

[0158] The first sub-determining unit 1001 is specifically used to determine a reference position corresponding to the position of the first wheel frame on the lane line of the first lane based on the lane line parameter equation; and to determine a first distance of the first wheel frame relative to the first lane in a preset direction based on the reference position and the position of the first wheel frame.

[0159] The second sub-determining unit 1002 is specifically used to determine the third distance of the first wheel contact position relative to the first lane in a preset direction based on the first wheel contact position and lane line parameter equation in the wheel contact position region; to determine the fourth distance of the second wheel contact position relative to the first lane in a preset direction based on the second wheel contact position and lane line parameter equation in the wheel contact position region; and to determine the second positional relationship based on the magnitude relationship between the third distance and the fourth distance.

[0160] The third sub-determination unit 1003 is specifically used to determine at least two fifth distances relative to the first lane in a preset direction based on at least two segmented positions in the segmented position region and lane line parameter equations; the at least two segmented positions include at least two intersection positions on the intersection line between the drivable area of ​​the second vehicle and the area where the second vehicle is located; and to determine a third positional relationship based on the magnitude relationship between the at least two fifth distances.

[0161] In some embodiments of this disclosure, such as Figure 11 As shown above, in the above Figure 7 Based on the embodiment shown, the second determining module 702 may include a seventh determining unit 7024 and an eighth determining unit 7025.

[0162] The seventh determining unit 7024 is used to determine the inclusion relationship between the image frame and the first lane.

[0163] The eighth determining unit 7025 is used to determine multiple spatial location regions of the second vehicle shifting from the second lane to the first lane based on the image frame, in response to the inclusion relationship being that the image frame contains the first lane.

[0164] Regarding the device for determining the positional relationship between the vehicle and the lane line during driving in the above embodiments, the specific methods of operation of each module and the corresponding beneficial effects have been described in detail in the corresponding embodiment section of the back-end implementation method of the aforementioned mid-chip system. Please refer to the corresponding execution operation methods and beneficial technical effects in the above exemplary method section, which will not be repeated here.

[0165] Exemplary electronic devices

[0166] Figure 12 This is a structural diagram of an electronic device provided in an exemplary embodiment of this disclosure. The electronic device can be the aforementioned device for determining the positional relationship between a vehicle and lane lines during driving, or it can be a terminal or server including the device for determining the positional relationship between a vehicle and lane lines during driving. Figure 12 As shown, the electronic device 120 includes a processor 1201 and a memory 1202.

[0167] The memory 1202 stores instructions executable by the processor 1201, which in turn implements the functions of each module in the device for determining the positional relationship between a vehicle and lane lines during driving, as described in the above embodiments. The memory 1202 stores at least one instruction, which is loaded and executed by the processor 1201 to implement the method for determining the positional relationship between a vehicle and lane lines during driving, as provided in the above embodiments.

[0168] The processor 1201 may include one or more central processing units (CPUs) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 120 to perform desired functions. The individual CPU in the processor 1201 may be a single-core processor or a multi-core processor.

[0169] The memory 1202 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1201 may execute one or more computer program instructions to implement the methods and / or other desired functions for determining the positional relationship between a vehicle and lane lines during driving, as described in the various embodiments of this disclosure above.

[0170] In some examples, the electronic device 120 may also include an input device 1203 and an output device 1204, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0171] The input device 1203 may also include, for example, a keyboard, a mouse, etc.

[0172] The output device 1204 can output various information to the outside, including, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0173] Of course, for the sake of simplicity, Figure 12Only some of the components of the electronic device 120 relevant to this disclosure are shown, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device 120 may include any other suitable components depending on the specific application.

[0174] Exemplary computer program products and computer-readable storage media

[0175] In addition to the methods and devices described above, embodiments of this disclosure may also provide a computer program product, including computer program instructions that, when executed by a processor, cause the processor to perform steps in the methods for determining the positional relationship between a vehicle and lane lines during driving, as described in the various embodiments of this disclosure in the "Exemplary Methods" section above.

[0176] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of embodiments of this disclosure. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0177] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform steps in the methods for determining the positional relationship between a vehicle and lane lines during driving, as described in the various embodiments of this disclosure in the "Exemplary Methods" section above.

[0178] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, but is not limited to, systems, apparatuses, or devices that are electrical, magnetic, optical, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0179] The basic principles of this disclosure have been described above with reference to specific embodiments. However, the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0180] Various modifications and variations can be made to this disclosure without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, this disclosure is also intended to include such modifications and variations.

Claims

1. A method for determining the positional relationship between a vehicle and lane lines during driving, comprising: Identify the image frames captured when the first vehicle is traveling in the first lane; Based on the image frame, multiple spatial location regions where the second vehicle shifts from the second lane to the first lane are determined; wherein, the multiple spatial location regions refer to at least two location regions in the image coordinate system that represent the positional relationship between the second vehicle and the first lane; Based on the multiple spatial location regions, the positional relationship of the second vehicle relative to the first lane in a preset direction is determined; Specifically, determining multiple spatial location regions where the second vehicle deviates from the second lane to the first lane based on the image frame includes: extracting multiple spatial location regions where the second vehicle deviates from the second lane to the first lane from the image frame using multiple neural network models; the multiple spatial location regions include the wheel frame location region of the second vehicle, the wheel contact location region of the second vehicle, and the segmented location region of the second vehicle.

2. The method according to claim 1, wherein, The step of extracting multiple spatial location regions from the image frame, respectively, by using multiple neural network models, from which the second vehicle has shifted from the second lane to the first lane, includes: The wheel frame location region of the second vehicle is determined from the image frame using a wheel frame recognition model; The wheel contact location area of ​​the second vehicle is determined from the image frame using a wheel contact identification model; The segmentation location region of the second vehicle is determined from the image frame using a segmentation location recognition model.

3. The method according to claim 1, wherein, Determining the positional relationship of the second vehicle relative to the first lane in a preset direction based on the multiple spatial location regions includes: Based on the image frame, the lane lines of the first lane are determined; Based on the lane lines of the first lane, determine the lane line parametric equations; Based on the multiple spatial location regions and the lane line parameter equations, the positional relationship of the second vehicle relative to the first lane in a preset direction is determined.

4. The method according to claim 3, wherein, Determining the positional relationship of the second vehicle relative to the first lane in a preset direction based on the multiple spatial location regions and the lane line parameter equation includes: Based on the wheel frame position region among the multiple spatial position regions and the lane line parameter equation, a first positional relationship between the wheel frame position region and the first lane in the preset direction is determined. Based on the wheel contact point region among the plurality of spatial location regions and the lane line parameter equation, a second positional relationship between the wheel contact point region and the first lane in the preset direction is determined. Based on the segmented location regions among the multiple spatial location regions and the lane line parameter equation, the third positional relationship of the segmented location regions relative to the first lane in the preset direction is determined; Based on the first positional relationship, the second positional relationship, and the third positional relationship, the positional relationship of the second vehicle relative to the first lane in the preset direction is determined.

5. The method according to claim 4, wherein, The step of determining the first positional relationship of the wheel frame position region relative to the first lane in a preset direction based on the wheel frame position region among the plurality of spatial position regions and the lane line parameter equation includes: Based on the first wheel frame position in the wheel frame position region and the lane line parameter equation, determine the first distance of the first wheel frame position relative to the first lane in the preset direction; Based on the second wheel frame position in the wheel frame position region and the lane line parameter equation, determine the second distance of the second wheel frame position relative to the first lane in the preset direction; The first positional relationship is determined based on the magnitude relationship between the first distance and the second distance.

6. The method according to claim 5, wherein, Determining the first distance of the first wheel rim position relative to the first lane in the preset direction based on the first wheel rim position in the wheel rim position region and the lane line parameter equation includes: Based on the lane line parameter equation, a reference position corresponding to the position of the first wheel frame is determined on the lane line of the first lane. Based on the reference position and the position of the first wheel rim, a first distance is determined relative to the first lane in the preset direction.

7. The method according to claim 4, wherein, The determination of the second positional relationship of the wheel contact point region relative to the first lane in a preset direction, based on the wheel contact point region among multiple spatial location regions and the lane line parameter equation, includes: Based on the first wheel contact point in the wheel contact point area and the lane line parameter equation, a third distance of the first wheel contact point relative to the first lane in the preset direction is determined; Based on the second wheel contact point in the wheel contact point area and the lane line parameter equation, a fourth distance of the second wheel contact point relative to the first lane in the preset direction is determined; The second positional relationship is determined based on the magnitude relationship between the third distance and the fourth distance.

8. The method according to claim 4, wherein, The step of determining the third positional relationship of the segmented position region relative to the first lane in a preset direction based on the segmented position region among the plurality of spatial position regions and the lane line parameter equation includes: Based on at least two segmentation locations in the segmentation location region and the lane line parameter equation, at least two fifth distances of the at least two segmentation locations relative to the first lane in the preset direction are determined respectively; the at least two segmentation locations include at least two intersection locations on the intersection line between the drivable area of ​​the second vehicle and the area where the second vehicle is located; The third positional relationship is determined based on the magnitude relationship between the at least two fifth distances.

9. The method according to any one of claims 1 to 8, wherein, The step of determining multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane based on the image frame includes: Determine the inclusion relationship between the image frame and the first lane; In response to the inclusion relationship that the image frame contains the first lane, multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane are determined based on the image frame.

10. A device for determining the positional relationship between a vehicle and a lane line during driving, comprising: The first determining module is used to determine the image frames captured when the first vehicle is traveling in the first lane; The second determining module is used to determine, based on the image frame, multiple spatial location regions where the second vehicle has shifted from the second lane to the first lane; wherein, the multiple spatial location regions refer to at least two location regions in the image coordinate system that represent the positional relationship between the second vehicle and the first lane; The third determining module determines the positional relationship of the second vehicle relative to the first lane in a preset direction based on the multiple spatial location regions; Specifically, the second determining module is used to extract multiple spatial location regions from the image frame where the second vehicle deviates from the second lane to the first lane using multiple neural network models; the multiple spatial location regions include the wheel frame location region of the second vehicle, the wheel contact location region of the second vehicle, and the segmented location region of the second vehicle.

11. A computer-readable storage medium storing a computer program for performing the method for determining the positional relationship between a vehicle and a lane line during driving, as described in any one of claims 1-9.

12. An electronic device, the electronic device comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method for determining the positional relationship between a vehicle and lane lines during driving as described in any one of claims 1-9.

Images