Method, device, vehicle and readable storage medium for constructing vehicle access road

Abstract

The application relates to a vehicle road passing construction method, device, vehicle and readable storage medium. The method comprises the following steps: acquiring a vehicle moving track and road boundaries distributed on both sides of the vehicle moving track according to a driving route of the vehicle; acquiring a plurality of left side road boundary points and a plurality of right side road boundary points corresponding to a plurality of track points sampled on the vehicle moving track; forming a left boundary of a passable road according to the plurality of left side road boundary points, and forming a right boundary of the passable road according to the plurality of right side road boundary points, so that the vehicle drives in a target passable road formed by the left boundary of the passable road and the right boundary of the passable road. The scheme provided by the application can provide prior information when a subsequent vehicle enters an automatic driving application scene of a fixed route, and the automatic driving is performed by using a self-built target passable road, so that the reliability and safety of the automatic driving of the vehicle are improved when the vehicle automatically drives along a fixed commuting route from a departure place to a destination.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a method, apparatus, vehicle, and readable storage medium for constructing a road for vehicles. Background Technology

[0002] In autonomous driving, vehicles typically plan their routes using lane-level navigation based on high-definition maps. However, the widespread adoption of high-definition map-based autonomous driving technology is hampered. On one hand, some regions lack high-definition map support, making autonomous driving impossible. On the other hand, when communication signals are weak, vehicle positioning information transmission is interrupted, preventing accurate positioning on high-definition maps and posing safety risks. Summary of the Invention

[0003] To address or partially address the problems existing in related technologies, this application provides a method, apparatus, vehicle, and readable storage medium for constructing vehicle access roads, which enables autonomous driving using self-built target access roads, and is reliable and safe.

[0004] The first aspect of this application provides a method for constructing a vehicle traffic road, comprising:

[0005] Based on the vehicle's driving route, obtain the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory.

[0006] Based on multiple trajectory points sampled from the vehicle's movement trajectory, obtain corresponding multiple left road boundary points and multiple right road boundary points;

[0007] A left boundary of a passable road is formed based on a plurality of left road boundary points, and a right boundary of a passable road is formed based on a plurality of right road boundary points, so that the vehicle can travel within the target passable road formed by the left boundary of the passable road and the right boundary of the passable road.

[0008] In some implementations, obtaining multiple left-side road boundary points and multiple right-side road boundary points based on multiple trajectory points sampled from the vehicle's movement trajectory includes:

[0009] Samples are taken on the vehicle's movement trajectory at preset intervals to obtain multiple trajectory points;

[0010] The road widths of each trajectory point and the road boundary are obtained respectively; wherein, the road widths include the left road width and the right road width respectively;

[0011] The trajectory point is moved along a first preset direction according to the corresponding left road width to obtain the corresponding left road boundary point; and the trajectory point is moved along a second preset direction according to the corresponding right road width to obtain the corresponding right road boundary point.

[0012] In some implementations, obtaining the road width between each trajectory point and the road boundary includes:

[0013] Using the trajectory point as the endpoint, draw a first ray along the first preset direction and a second ray along the second preset direction respectively;

[0014] Obtain the intersection point of the first ray and the road boundary to obtain the left road spacing; and obtain the intersection point of the second ray and the road boundary to obtain the right road spacing;

[0015] The left and right road spacings are compared with preset distance thresholds to obtain the corresponding left and right road widths.

[0016] In some implementations, after obtaining the road width of each trajectory point and the road boundary, the method further includes:

[0017] Each of the road widths is smoothed to obtain an optimized target width; the target width includes the left target width and the right target width.

[0018] In some implementations, the step of smoothing each of the road widths to obtain an optimized target width includes:

[0019] Calculate the difference in width between each pair of adjacent roads on the same side to cluster the widths of multiple roads on the same side that meet the preset clustering rules, and obtain at least one set of width sequences.

[0020] If the width of all roads in the current width sequence is greater than the width of the last road in the adjacent preceding width sequence and greater than the width of the first road in the adjacent following width sequence, then all road widths in the current width sequence are updated to the width of the last road or the width of the first road to obtain the corresponding target width.

[0021] In some embodiments, forming a passable left boundary of a road based on a plurality of left road boundary points, and forming a passable right boundary of a road based on a plurality of right road boundary points, includes:

[0022] The trajectory point is moved along a first preset direction according to the corresponding left target width to obtain the corresponding left road boundary point; and the trajectory point is moved along a second preset direction according to the corresponding right target width to obtain the corresponding right road boundary point.

[0023] Connect the left road boundary points in sequence to generate the left boundary of the passable road; connect the right road boundary points in sequence to generate the right boundary of the passable road.

[0024] In some implementations, before calculating the difference in width between any two adjacent roads on the same side, the method further includes:

[0025] The preset distances between each trajectory point and the starting point of the vehicle's movement trajectory are obtained respectively;

[0026] Based on the corresponding preset distance values, the widths of the roads on the same side are sorted in ascending order to obtain the sequentially arranged road widths on the same side.

[0027] The second aspect of this application provides an application for constructing a target passable road according to the method for constructing a vehicle passageway as described in the first aspect of this application.

[0028] A third aspect of this application provides a device for constructing a vehicle traffic road, comprising:

[0029] The data acquisition module is used to acquire the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory based on the vehicle's driving route.

[0030] The boundary point generation module is used to obtain multiple left road boundary points and multiple right road boundary points respectively based on multiple trajectory points sampled on the vehicle's movement trajectory.

[0031] The road generation module is used to form a left boundary of a passable road based on multiple left road boundary points, and to form a right boundary of a passable road based on multiple right road boundary points, so that the vehicle can travel within the target passable road formed by the left boundary of the passable road and the right boundary of the passable road.

[0032] A fourth aspect of this application provides a vehicle, comprising:

[0033] Processor; and

[0034] A memory that stores executable code, which, when executed by the processor, causes the processor to perform the method described above.

[0035] A fifth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method described above.

[0036] The technical solution provided in this application may include the following beneficial effects:

[0037] The technical solution of this application obtains the corresponding vehicle movement trajectory and road boundary based on the vehicle's driving route, and constructs the target passable road corresponding to the driving route using the vehicle movement trajectory and road boundary. This provides prior information for subsequent autonomous driving application scenarios where the vehicle enters a fixed route. For example, when the vehicle performs autonomous driving along a fixed commuting route with a starting point and destination, it can perform autonomous driving without the support of high-precision maps, thereby improving the reliability of autonomous driving, reducing the safety risks caused by the inability to use high-precision maps during autonomous driving, and preventing the vehicle from taking the wrong route during autonomous driving because the target passable road is unique.

[0038] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0039] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0040] Figure 1 This is a schematic flowchart illustrating the method for constructing a vehicle passageway according to an embodiment of this application;

[0041] Figure 2 This is a schematic diagram of the area map shown in the embodiments of this application;

[0042] Figure 3 This is another schematic flowchart illustrating the method for constructing a vehicle passageway as shown in the embodiments of this application;

[0043] Figure 4 This is another schematic flowchart illustrating the method for constructing a vehicle passageway as shown in the embodiments of this application;

[0044] Figure 5 This is a schematic diagram of the structure of a vehicle passageway construction device shown in an embodiment of this application;

[0045] Figure 6 This is another structural schematic diagram of the vehicle passageway construction device shown in the embodiments of this application;

[0046] Figure 7 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application. Detailed Implementation

[0047] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0048] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0049] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. 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, "multiple" means two or more, unless otherwise explicitly specified.

[0050] Among related technologies, autonomous driving technology based on high-precision maps is difficult to popularize. On the one hand, some regions lack the support of high-precision maps, making autonomous driving impossible. On the other hand, when communication signals are poor, the transmission of vehicle positioning information is interrupted, making it impossible to obtain accurate positioning on high-precision maps, which poses safety risks.

[0051] To address the aforementioned issues, this application provides a method for constructing vehicle traffic roads, which can effectively improve the reliability of autonomous driving without the support of high-precision map data and reduce the safety risks caused by the inability to use high-precision maps during autonomous driving.

[0052] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart illustrating a method for constructing a vehicle traffic road according to an embodiment of this application.

[0054] See Figure 1 and Figure 2 The method for constructing a vehicle passageway according to this application includes:

[0055] S110: Based on the vehicle's driving route, obtain the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory.

[0056] During a vehicle's journey, the path it takes from its starting point to its destination forms a physical route. For example, a vehicle's route might be from point A to point C via a certain path. In other words, a route is a fixed path with direction.

[0057] During vehicle operation, the vehicle can obtain trajectory coordinates along its route using positioning technology. Multiple trajectory coordinates distributed sequentially over time form a corresponding vehicle movement trajectory. Correspondingly, within the same coordinate system, the vehicle can perceive and store road boundary information distributed along both sides of the vehicle's movement trajectory using various sensors installed on the vehicle body. That is, based on the vehicle's direction of travel, the road boundaries include the left road boundary located to the left of the vehicle's movement trajectory and the right road boundary located to the right of the vehicle's movement trajectory. In some implementations, a corresponding regional map is constructed based on the vehicle's movement trajectory and the corresponding road boundary information. It can be understood that in the regional map, the relative positional relationship between the vehicle's movement trajectory and the road boundaries can be determined based on the position coordinates of the vehicle's movement trajectory and the road boundaries.

[0058] In some implementations, the vehicle trajectory can consist of a continuous curve or multiple discontinuous curve segments. For example, smoothly connecting the trajectory coordinates in a time sequence can form a corresponding curve or multiple curve segments. Optionally, the vehicle trajectory can consist of multiple discrete points. Optionally, the vehicle trajectory can consist of multiple continuous or discontinuous line segments. That is, the vehicle trajectory is a set of position coordinates traversed by the vehicle in a time sequence during its journey, and its representation is not limited thereto. In some implementations, the road boundary can be represented using point cloud data, or it can be represented using continuous or discontinuous curves.

[0059] Furthermore, the vehicle's trajectory can correspond to the entire or partial route of the vehicle's journey. In other words, the start and end points of the vehicle's trajectory are equivalent to the start and end points of the corresponding route. Alternatively, the start point of the vehicle's trajectory can be a location along the route, and the end point can be another location along the route.

[0060] Furthermore, the road boundary can be acquired using relevant environmental perception technologies during vehicle operation to represent the physical division boundary of the road in a real-world scenario. Specifically, the road boundary can be acquired by onboard sensors, which may include one or more of the following: onboard cameras, onboard LiDAR, onboard ultrasonic radar, onboard millimeter-wave radar, onboard infrared sensors, etc., without limitation. In some embodiments, the road boundary may include at least one of ground markings, physical dividing facilities, or the boundary contours of buildings. Ground markings may be, for example, traffic markings indicating road boundaries such as guide lines or double solid yellow lines. Physical dividing facilities may be, for example, road barriers such as fences or water-filled barriers. The boundary contours of buildings may be the boundary contours of buildings located on both sides of the road.

[0061] S120: Based on multiple trajectory points sampled from the vehicle's movement trajectory, obtain the corresponding multiple left-side road boundary points and multiple right-side road boundary points respectively.

[0062] In this step, multiple trajectory points are obtained by sampling the vehicle's movement trajectory. Based on each trajectory point, the corresponding left and right road boundary points can be obtained separately.

[0063] It should be noted that, due to objective factors of the scenario and the sensing capabilities of the vehicle's sensors, road boundaries may be interrupted. Therefore, not every left-side road boundary point is entirely located on the left-side road boundary, and not every right-side road boundary point is entirely located on the right-side road boundary.

[0064] S130, a drivable road left boundary is formed based on multiple left road boundary points, and a drivable road right boundary is formed based on multiple right road boundary points, so that vehicles can travel within the target drivable road formed by the drivable road left boundary and the drivable road right boundary.

[0065] In this step, multiple left-side road boundary points are connected sequentially to form the left boundary of the passable road, and multiple right-side road boundary points are connected sequentially to form the right boundary of the passable road. Based on the left and right boundaries of the passable road, the target passable road can be formed, allowing the vehicle to drive safely within the target passable road during autonomous driving.

[0066] It is understandable that the target traversable road can be used to define the roads that vehicles are allowed to travel on during autonomous driving. Based on the permitted traversable roads, the vehicle's traversable area is limited, preventing the vehicle from entering incorrect paths. At the same time, it can reduce the amount of environmental information collected by the vehicle outside the traversable area during autonomous driving, reducing errors in the environmental information collected by the vehicle's sensors and improving the stability of the autonomous driving process.

[0067] Furthermore, the target traversable road obtained in this application can be used for autonomous driving when a self-driving vehicle or other vehicles use the same driving route, such as a vehicle driving autonomously along a fixed commuting route with a fixed origin and destination. That is, when a self-driving vehicle or other vehicle drives autonomously again along the same driving route, the target traversable road constructed in this application, due to its uniqueness, effectively reduces the impact of other factors outside the traversable area on autonomous driving, such as excluding road structures outside the field of vision, reducing some perception errors beyond line of sight and uncertainties in blind spots, enabling the vehicle to drive autonomously along the unique traversable area, avoiding the vehicle taking the wrong route during autonomous driving, and thus improving the stability of the vehicle's autonomous driving process.

[0068] In this embodiment, the technical solution of this application obtains the corresponding vehicle movement trajectory and the corresponding road boundary according to the vehicle's driving route, and constructs the corresponding target passable road using the vehicle movement trajectory and road boundary. This provides prior information for subsequent autonomous driving application scenarios where the vehicle enters a fixed route, such as when the vehicle is driving autonomously along a fixed commuting route from the origin to the destination (e.g., when the vehicle enters commuting mode). The corresponding target passable road is applied to the vehicle's autonomous driving process, which effectively improves the reliability of autonomous driving without the support of high-precision map data, reduces the safety risks caused by the inability to use high-precision maps during autonomous driving, and avoids the vehicle from taking the wrong route during autonomous driving because the target passable road is unique.

[0069] Figure 3 This is a flowchart illustrating a method for constructing a vehicle traffic road according to an embodiment of this application.

[0070] See Figure 3 The method for constructing a vehicle passageway according to this application includes:

[0071] S210 generates a regional map based on the vehicle's travel route. The regional map includes the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory.

[0072] In this step, based on the vehicle's travel route, the vehicle's trajectory and the road boundaries distributed on both sides of the trajectory are obtained. In the same coordinate system, the vehicle trajectory and the road boundaries on both sides of the trajectory form a regional map.

[0073] S220 samples the vehicle's movement trajectory at preset intervals to obtain multiple trajectory points.

[0074] In this step, a preset interval is used to sample trajectory points at intervals along the vehicle's trajectory, starting from the starting point and proceeding sequentially to obtain multiple trajectory points. It should be understood that the trajectory composed of these multiple trajectory points is used to obtain road boundary points.

[0075] The preset spacing can be set according to actual application requirements. It should be understood that a smaller preset spacing indicates higher calculation accuracy, allows for the collection of more trajectory points, and results in more accurate and reliable road boundaries. However, it also generates a larger amount of data computation. In some implementations, the preset interval threshold can be selected from 0.5 meters to 2 meters. This ensures calculation accuracy while maintaining appropriate computational load and data storage space for the system.

[0076] The starting point of the vehicle's movement trajectory is the first trajectory point, and the ending point is the last trajectory point.

[0077] S230, obtain the road width between each trajectory point and the road boundary; the road width includes the left road width and the right road width.

[0078] In this step, starting from the first trajectory point, the road width from each trajectory point to the road boundaries located on both sides of the vehicle's movement trajectory is obtained. This road width includes the left road width from the trajectory point to the left road boundary and the right road width from the trajectory point to the right road boundary. It can be understood that the left and right road widths corresponding to the same trajectory point are independent and may be the same or different.

[0079] See Figure 2 In some specific implementations, obtaining the road width between each trajectory point and the road boundary may include the following steps:

[0080] S231, using the trajectory point as the endpoint, draw a first ray along a first preset direction and a second ray along a second preset direction.

[0081] It can be understood that each trajectory point has corresponding road boundary points on both sides, and further, each trajectory point and its two corresponding road boundary points can be located on the same straight line. Specifically, the first preset direction faces the road boundary located to the left of the driving direction, and the second preset direction faces the road boundary located to the right of the driving direction.

[0082] Specifically, in some implementations, a first preset direction and a second preset direction can be determined based on the orientation angle of each trajectory point. The orientation angle of each trajectory point can be obtained using a unified preset rule. In some specific implementations, within the same angular coordinate system, a preset reference direction is set as the starting edge of the orientation angle corresponding to each trajectory point, and the tangent direction of the current trajectory point on the vehicle's movement trajectory, pointing towards the next trajectory point, is set as the ending edge of the orientation angle. The angle between the starting and ending edges is the orientation angle corresponding to the current trajectory point. Furthermore, the preset reference direction can be arbitrarily defined within the same angular coordinate system, and all trajectory points use the same preset reference direction as the starting edge of their orientation angles.

[0083] As an example, such as Figure 3 As shown, let θ be the orientation angle of the trajectory point. Then, the first preset direction of the trajectory point can be the terminal side of the angle (θ + π / 2). According to the rule that the two road boundary points are located on the same straight line, the second preset direction of the trajectory point can be the terminal side corresponding to the angle (θ - π / 2).

[0084] After determining the first preset direction and the second preset direction, a first ray can be drawn along the first preset direction, and a second ray can be drawn along the second preset direction. For example... Figure 3 As shown, assuming the trajectory point is P, a first ray can be drawn along the first preset direction where the terminal side of (θ+π / 2) lies. A second ray can be drawn along the second preset direction where the terminal side of (θ-π / 2) lies.

[0085] S232, obtain the intersection point of the first ray and the road boundary to obtain the left road spacing; and obtain the intersection point of the second ray and the road boundary to obtain the right road spacing.

[0086] like Figure 3 As shown, in this step, after drawing the corresponding first ray along the first preset direction, the left intersection point L corresponding to the left road boundary can be obtained. The straight-line distance between the left intersection point and the trajectory point is the left road spacing D. L Similarly, after drawing the corresponding second ray along the second preset direction, the right intersection point R between the second ray and the right road boundary can be obtained. The straight-line distance between the right intersection point and the trajectory point is the right road spacing D. R .

[0087] S233, compare the left road spacing and right road spacing with preset distance thresholds respectively to obtain the corresponding left road width and right road width.

[0088] The preset distance threshold is the maximum extension length of the first and second rays, calculated using the trajectory point as the endpoint. This preset distance threshold limits the maximum width of the left and right roads, ensuring compliance with real-world road boundary widths. The preset distance threshold can be set according to actual application requirements. For example, it can be between 20 and 25 meters. When the left road spacing is less than or equal to the preset distance threshold, the current left road spacing is the left road width. Conversely, when the left road spacing is greater than the preset distance threshold, the preset distance threshold is the corresponding left road width. Similarly, when the right road spacing is less than or equal to the preset distance threshold, the current right road spacing is the right road width. Conversely, when the right road spacing is greater than the preset distance threshold, the preset distance threshold is the corresponding right road width.

[0089] By comparing the distance between the left and right roads with preset distance thresholds, the values ​​of the left and right road widths can be determined. This effectively ensures the consistency and stability of the road boundary point selection process and avoids the loss or failure of road boundary points.

[0090] It should be understood that during actual vehicle sensing, it is impossible to detect every road boundary; there may be instances where road boundaries are missing or interrupted. Therefore, not every first or second ray emanating from a trajectory point can find an intersecting road boundary, thus failing to form a corresponding left or right intersection point. When an intersection point is missing, the preset distance threshold is the width of the left or right road corresponding to the current trajectory point.

[0091] After obtaining the left and right road widths corresponding to each trajectory point, the target passable road can be generated using the following different methods. For example... Figure 3 As shown, S240, S2501, and S260 can be executed sequentially. For example... Figure 4 As shown, step S2502 and S260 can be executed directly without executing S240.

[0092] S240, each road width is smoothed to obtain the optimized target width; the target width includes the left target width and the right target width.

[0093] To smooth the boundaries of the subsequently obtained target passable roads and avoid situations where local road sections are too wide or too narrow, this application can also smooth the road widths of some trajectory points. In this step, based on the actual values ​​of each obtained road width, road widths with large value fluctuations are smoothed to obtain optimized target widths. It can be understood that the left and right road widths are smoothed independently. For example, for the same trajectory point, its left road width is optimized to the left target width, while its right road width may remain unchanged; or the left road width remains unchanged, while its right road width is optimized to the right target width.

[0094] It should be understood that factors such as vehicle swaying or lateral deviation during operation can lead to poor smoothness in the sensed vehicle trajectory and road boundaries, resulting in unstable road widths on both sides of the trajectory points. This step smooths out some road widths, and the optimized target width is used to calculate road boundary points in subsequent steps. This effectively avoids the adverse chain reaction of fluctuations in the front-end data on subsequent calculation data, thereby improving the smoothness of the target passable road obtained later.

[0095] It is understandable that, for road widths in different directions, the left-side road width and the right-side road width can be smoothed separately to achieve a smooth effect on the road boundaries on the same side.

[0096] In some specific implementations, smoothing each road width to obtain an optimized target width may include the following steps:

[0097] S241, calculate the difference in width between each pair of adjacent roads on the same side, so as to cluster the widths of multiple roads on the same side that meet the preset clustering rules and obtain at least one set of width sequences.

[0098] In this step, the sorting results of all road widths on the same side are first obtained, forming an ordered sequence. Taking the left-side road width as an example, the left-side road width corresponding to each trajectory point can be sorted sequentially from beginning to end according to the index of the corresponding trajectory point. For example, the sequence of left-side road widths can be represented as follows: `w` represents the road width, the superscript `l` indicates the left side, and the subscript `N` is the index corresponding to each road width. The value of `N` corresponds to the number of trajectory points. It can be understood that when representing the road width on the right side, the superscript `r` can be used to distinguish it from the left side. Other sorting methods can also be used; only examples are given here.

[0099] Next, the difference in width between each pair of adjacent roads on the same side is calculated. Then, based on the calculated differences, the widths of multiple roads on the same side that conform to the preset clustering rules are clustered to obtain a corresponding set of width sequences. A set of width sequences can be represented as follows: Where M is the index corresponding to each road width in the current width sequence. Clustering can be understood as grouping the widths of multiple adjacent roads on the same side that conform to a preset clustering rule into the same width sequence. All road widths on the same side can be clustered into multiple width sequences. All width sequences on the same side can be represented as X = {C1, C2, ..., C...} k}, where k represents the number of width sequences.

[0100] In some specific implementations, the preset clustering rule can be: when the difference between the widths of two adjacent roads on the same side is less than or equal to a clustering width threshold, the widths of the two adjacent roads on the same side are grouped together. In some implementations, the clustering width threshold can be 1 meter to 5 meters.

[0101] To facilitate understanding, let's take the clustering of left-side road widths as an example. We select the left-side road widths corresponding to nine trajectory points that are consecutively distributed over time. w1 to w9 are arranged sequentially. The values ​​of w1 to w9 are: w1=1, w2=3, w3=4, w4=8, w5=9, w6=11, w7=2, w8=3, w9=4, and the width of each left-hand road is in meters. Assume that the cluster width threshold is set to 3 meters.

[0102] Calculate the absolute value of the difference between the widths of the left-hand roads of every two adjacent rows. For example, the difference between w1 and w2 is 2 meters; the difference between w2 and w3 is 1 meter; the difference between w3 and w4 is 4 meters; the difference between w4 and w5 is 1 meter; the difference between w5 and w6 is 2 meters; the difference between w6 and w7 is 9 meters; the difference between w7 and w8 is 1 meter; and the difference between w8 and w9 is 1 meter.

[0103] By comparing the above difference with the cluster width threshold, it can be seen that three width sequences can be obtained by clustering: C1 = {w1, w2, w3}, C2 = {w4, w5, w6}, and C3 = {w7, w8, w9}. Then, the complete width sequence of the left road can be denoted as X = {C1, C2, C3}.

[0104] Of course, other preset clustering rules can also be used to cluster multiple road widths on the same side in this application, and there are no restrictions here.

[0105] In some implementations, when obtaining the sorting results of all road widths located on the same side, the results can also be obtained in the following manner:

[0106] S2411, obtain the preset distance between each trajectory point and the starting point of the vehicle's movement trajectory.

[0107] The preset distance can be the sum of the distances between the current trajectory point and the starting point of the vehicle's movement trajectory, as well as the distances between all adjacent trajectory points.

[0108] For example, a preset distance S can be defined as:

[0109]

[0110] Where n is the index of the corresponding trajectory point; d i This represents the Euclidean distance from the i-th point to the (i-1)-th point. It can be understood that the first trajectory point is the starting point of the vehicle's movement trajectory, and there is no need to calculate the corresponding preset distance.

[0111] S2412, based on the corresponding preset distance values, sort the widths of the roads on the same side in ascending order to obtain the sequentially arranged road widths on the same side.

[0112] Sort the roads according to the preset distance corresponding to each road width, and obtain the road widths on the same side in the order.

[0113] By sorting the widths of roads on the same side using the above method, the correspondence of vehicle routes on subsequently acquired target passable roads is further ensured, effectively reducing data errors. Of course, other related technologies can also be used to sort the widths of roads on the same side.

[0114] After obtaining multiple width sequences on the same side, the following adjustment will be performed using the road width in the entire width sequence as the unit to be adjusted.

[0115] S242, if all road widths in the current width sequence are greater than the width of the last road in the adjacent preceding width sequence and greater than the width of the first road in the adjacent following width sequence, then update all road widths in the current width sequence to the width of the last road or the width of the first road to obtain the corresponding target width.

[0116] In this step, adjacent width sequences after clustering are compared. If all road widths in the current width sequence are greater than the width of the last road in the preceding width sequence and greater than the width of the first road in the following width sequence, then the smoothness of the current width sequence is considered unsatisfactory. The current width sequence is wider than the road widths before and after it, forming a non-smooth "narrow-wide-narrow" boundary. By updating all road widths in the current width sequence to the width of the last or first road, the target width is obtained for all road widths in the current width sequence. For example, all road widths in the current width sequence can be updated to the width of the last road; or all road widths in the current width sequence can be updated to the width of the first road; or all road widths in the current width sequence can be updated to the larger of the width of the last or the first road.

[0117] To facilitate understanding of the technical solution of this application, let's take the width sequence X = {C1, C2, C3} summarized in the above example as an example. Assume C2 is the current width sequence, with w4 = 8, w5 = 9, and w6 = 11, all greater than w3 = 4 and w7 = 2. Clearly, the width values ​​of each road in C2 are greater than the last road width w3 in the preceding C1, and also greater than the first road width w7 in the following C3. This indicates that all road widths in C2 need to be optimized. Accordingly, w4, w5, and w6 can be updated to the values ​​of w3 or w7. To avoid excessive changes, we can choose to update to the larger of w3 and w7, for example, updating all w4, w5, and w6 to the value of w3.

[0118] By employing the aforementioned smoothing process, the smoothness between different road widths is improved, thereby enhancing the smoothness of the target passable road in subsequent steps. Of course, other smoothing methods from related technologies can also be used to improve the smoothness between different road widths; no limitation is imposed here.

[0119] It's understandable that if all road widths in the current width sequence are greater than the width of the last road in the preceding width sequence, but less than the width of the first road in the following width sequence, then there's no need to optimize the current width sequence. Similarly, if all road widths in the current width sequence are less than the width of the last road in the preceding width sequence, but greater than the width of the first road in the following width sequence, then there's also no need to optimize the current width sequence.

[0120] Based on steps S241 and S242 above, the width of the left road is smoothed to obtain the target width on the left. Similarly, the width of the right road is smoothed to obtain the target width on the right. It can be understood that for road widths that do not require optimization and remain unchanged, the original value is used as the corresponding target width.

[0121] S2501, Move the trajectory point along the first preset direction according to the corresponding left target width to obtain the corresponding left road boundary point; and Move the trajectory point along the second preset direction according to the corresponding right target width to obtain the corresponding right road boundary point.

[0122] In this step, based on the left target width and the first preset direction obtained in the previous steps, the corresponding trajectory point is translated along the first preset direction according to the left road width to obtain the corresponding left road boundary point. Similarly, based on the right target width and the second preset direction obtained in the previous steps, the corresponding trajectory point is moved along the second preset direction according to the right road width to obtain the corresponding right road boundary point.

[0123] Similarly, using the same method, all trajectory points are translated to both sides to obtain multiple left road boundary points and multiple right road boundary points.

[0124] As an example, assuming the first preset direction of the trajectory point is an angle of (θ+π / 2) and the second preset direction is an angle of (θ-π / 2), then by translating the trajectory point along the angle of (θ+π / 2) to the corresponding width of the left road, the left road boundary point is obtained. Similarly, by translating the trajectory point along the angle of (θ-π / 2) to the corresponding width of the right road, the right road boundary point is obtained.

[0125] S2502, the trajectory point is moved along the first preset direction according to the corresponding left road width to obtain the corresponding left road boundary point; and the trajectory point is moved along the second preset direction according to the corresponding right road width to obtain the corresponding right road boundary point.

[0126] See Figure 4 The difference between step S2502 and step S2501 is that the road widths are not smoothed in step S2502. The trajectory points can be directly obtained by translating the left and right road widths generated in step S230 along their corresponding directions to obtain the left and right road boundary points.

[0127] S260, connect the left road boundary points in sequence to generate the left boundary of the passable road; connect the right road boundary points in sequence to generate the right boundary of the passable road, so that vehicles can travel within the target passable road formed by the left and right boundaries of the passable road.

[0128] In this step, relevant techniques can be used for fitting to smooth the lines connecting the left-side road boundary points, forming the left boundary of the passable road. Similarly, the lines connecting the right-side road boundary points are smoothed to form the right boundary of the passable road. The target passable road is formed between the left and right boundaries of the passable road. When the vehicle is in autonomous driving mode, the passable area is the target passable road.

[0129] In this embodiment, the technical solution of this application samples multiple trajectory points on the vehicle's movement trajectory, obtains the corresponding left and right road widths using these sampled trajectory points, and optimizes the road widths that need smoothing through clustering. After obtaining the corresponding target width, each trajectory point is translated based on the left and right target widths to obtain multiple left and right road boundary points. These points are then quickly connected to form the left and right boundaries of the target passable road, creating a smooth and reliable target passable road for safer and smoother autonomous driving. Simultaneously, when performing autonomous driving based on a regional map, road structures outside the vehicle's field of vision can be determined in advance based on prior information such as road boundaries and road widths. This reduces some beyond-line-of-sight perception errors and uncertainties in blind spots, providing a stable and reliable map foundation for autonomous driving.

[0130] This application also provides an application of a vehicle driving road constructed according to the vehicle driving road construction method in any of the above embodiments. The vehicle driving road constructed in this application can be applied to autonomous driving of vehicles. The application scenario corresponding to autonomous driving in this application can include: commuting scenarios. Commuting scenarios can be considered as scenarios where users drive frequently.

[0131] It's understandable that when a vehicle operates autonomously in a commuting scenario, its origin (e.g., home) and destination (e.g., office) are fixed. The vehicle can pre-construct target traversable routes from home to office, and vice versa. This means that by constructing a corresponding target traversable route based on the origin, destination, and driving route, and maintaining the uniqueness and reliability of each route, the vehicle won't take the wrong path when traveling along any target traversable route, providing a stable and reliable map foundation for autonomous driving.

[0132] The following example illustrates an autonomous driving method based on the road in which the vehicle travels.

[0133] In some implementations, the autonomous driving method of this application may include the following steps:

[0134] S310: Obtain the vehicle's current location information, preset destination location information, and the target passable road.

[0135] The target passable roads are pre-constructed according to the above-mentioned vehicle driving road construction method.

[0136] The destination location information can be the user-selected location coordinates. Generally, the current location information is not equal to the destination location information. The current location information may be the same as or different from the starting point of the target accessible road. The preset destination location information may be the same as or different from the destination location of the target accessible road. S320, the current location information and the destination location information are matched with the target accessible road respectively to obtain the corresponding location matching results.

[0137] If both the vehicle's current location and destination location are within the target accessible road area, it indicates a successful location match with the target accessible road, and the vehicle can autonomously drive to the preset destination location based on the target accessible road. Otherwise, the opposite applies.

[0138] S330: If the current location information and the destination location information are located on the target passable road, the vehicle will perform autonomous driving based on the target passable road.

[0139] The vehicle driving road constructed by the aforementioned method in this application is applied to the autonomous driving process, enabling the vehicle to complete autonomous driving without relying on high-precision maps, and ensuring the reliability of the autonomous driving process. It reduces the impact of other factors besides the target passable road on autonomous driving, such as excluding road structures outside the field of vision, reducing some perception errors beyond line of sight and the uncertainty of blind spots, thereby improving the stability of the vehicle's autonomous driving process. Furthermore, since the target passable road is unique, it can prevent the vehicle from taking the wrong road during the autonomous driving process.

[0140] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a vehicle driving road construction device, electronic device, and corresponding embodiments.

[0141] Figure 5 This is a schematic diagram of the structure of the vehicle driving road construction device shown in the embodiments of this application.

[0142] See Figure 5 The vehicle driving road construction device 400 of this application includes a data acquisition module 410, a boundary point generation module 420 and a road generation module 430.

[0143] The data acquisition module 410 is used to acquire the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory based on the vehicle's driving route.

[0144] The boundary point generation module 420 is used to obtain multiple left road boundary points and multiple right road boundary points respectively based on multiple trajectory points sampled on the vehicle's movement trajectory.

[0145] The road generation module 430 is used to form a left boundary of a passable road based on multiple left road boundary points and a right boundary of a passable road based on multiple right road boundary points, so that vehicles can travel within the target passable road composed of the left and right boundaries of the passable road.

[0146] See Figure 6 In one specific implementation, the boundary point generation module 420 includes a trajectory point sampling module 421, a width calculation module 422, and a boundary point determination module 423.

[0147] In some implementations, the trajectory point sampling module 421 is used to sample on the vehicle's movement trajectory at preset intervals to obtain multiple trajectory points.

[0148] In some implementations, the width calculation module 422 is used to obtain the road width between each trajectory point and the road boundary; wherein the road width includes the left road width and the right road width.

[0149] In some specific implementations, the width calculation module 422 may be used to obtain the road width between each trajectory point and the road boundary, including: drawing a first ray along a first preset direction and a second ray along a second preset direction, starting from the trajectory point; obtaining the intersection of the first ray and the road boundary to obtain the left road spacing; obtaining the intersection of the second ray and the road boundary to obtain the right road spacing; and comparing the left road spacing and the right road spacing with preset distance thresholds to obtain the corresponding left road width and right road width.

[0150] In some implementations, the boundary point determination module 423 is used to move the trajectory point along a first preset direction according to the corresponding left road width to obtain the corresponding left road boundary point; and to move the trajectory point along a second preset direction according to the corresponding right road width to obtain the corresponding right road boundary point.

[0151] In some implementations, the road generation module 430 is used to sequentially connect the left road boundary points to generate the left boundary of the passable road; and sequentially connect the right road boundary points to generate the right boundary of the passable road, so that the left boundary and the right boundary of the passable road constitute the target passable road.

[0152] The apparatus of this application also includes a smoothing module 440, which is used to smooth each road width to obtain an optimized target width; the target width includes the left target width and the right target width.

[0153] In some specific implementations, the smoothing module 440 includes a width clustering module 441 and a width optimization module 442. The width clustering module 441 calculates the difference between the widths of two adjacent roads on the same side to cluster multiple road widths on the same side that conform to a preset clustering rule, obtaining a corresponding set of width sequences. The width optimization module 442 updates all road widths in the current width sequence to the width of the last road in the preceding width sequence or the width of the first road in the following width sequence if all road widths in the current width sequence are greater than the width of the last road in the preceding width sequence and greater than the width of the first road in the following width sequence, thus obtaining the corresponding target width. The target widths include a left target width and a right target width.

[0154] In some implementations, the boundary point determination module 423 is used to move the trajectory point along a first preset direction according to the corresponding left target width to obtain the corresponding left road boundary point; and to move the trajectory point along a second preset direction according to the corresponding right target width to obtain the corresponding right road boundary point.

[0155] In this embodiment, the technical solution of this application obtains the corresponding vehicle movement trajectory and the corresponding road boundary according to the vehicle's driving route, and uses the vehicle movement trajectory and road boundary to obtain the target passable road corresponding to the driving route, providing prior information for the subsequent autonomous driving process of the vehicle on the same driving route, effectively improving the reliability of autonomous driving without the support of high-precision map data, and reducing the safety risks caused by the inability to use high-precision maps during autonomous driving.

[0156] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated further here.

[0157] Figure 7 This is a schematic diagram of the vehicle structure shown in the embodiments of this application.

[0158] See Figure 7 The vehicle 1000 includes a memory 1010 and a processor 1020.

[0159] The processor 1020 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0160] Memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by processor 1020 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, a high-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.

[0161] The memory 1010 stores executable code, which, when processed by the processor 1020, can cause the processor 1020 to execute part or all of the methods described above.

[0162] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0163] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0164] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for constructing a vehicle traffic road, characterized in that, include: Based on the vehicle's driving route, obtain the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory. Based on multiple trajectory points sampled from the vehicle's movement trajectory and the road width corresponding to each trajectory point and the road boundary, multiple left road boundary points and multiple right road boundary points are obtained respectively; wherein, the road width includes the left road width and the right road width, the road width is determined according to the road distance between the intersection of the trajectory point and the road boundary, and both the left road width and the right road width are less than or equal to a preset distance threshold; wherein, the trajectory point is moved along a first preset direction according to the corresponding left road width to obtain the corresponding left road boundary point; and the trajectory point is moved along a second preset direction according to the corresponding right road width to obtain the corresponding right road boundary point; A left boundary of a passable road is formed based on a plurality of left road boundary points, and a right boundary of a passable road is formed based on a plurality of right road boundary points, so that the vehicle can travel within the target passable road formed by the left boundary of the passable road and the right boundary of the passable road.

2. The method according to claim 1, characterized in that, The method further includes: The first preset direction and the second preset direction are determined according to the orientation angle of the trajectory point; wherein, in the angular coordinate system, the preset reference direction is set as the starting side of the orientation angle corresponding to each trajectory point, and the tangent direction of the current trajectory point on the vehicle movement trajectory and the tangent pointing to the next trajectory point is set as the ending side of the orientation angle, and the angle between the starting side and the ending side is the orientation angle corresponding to the current trajectory point; if the orientation angle of the current trajectory point is θ, then the first preset direction of the current trajectory point is the ending side of the angle (θ+π / 2), and the second preset direction of the current trajectory point is the ending side corresponding to the angle (θ-π / 2).

3. The method according to claim 1, characterized in that, The step of obtaining the road width between each trajectory point and the road boundary includes: Using the trajectory point as the endpoint, draw a first ray along the first preset direction and a second ray along the second preset direction respectively; Obtain the intersection point of the first ray and the road boundary to obtain the left road spacing; and obtain the intersection point of the second ray and the road boundary to obtain the right road spacing; The left and right road spacings are compared with preset distance thresholds to obtain the corresponding left and right road widths.

4. The method according to claim 1, characterized in that, After obtaining the road width of each trajectory point and the road boundary, the method further includes: Each of the road widths is smoothed to obtain an optimized target width; the target width includes the left target width and the right target width.

5. The method according to claim 4, characterized in that, The step of smoothing each of the road widths to obtain the optimized target width includes: Calculate the difference in width between each pair of adjacent roads on the same side to cluster the widths of multiple roads on the same side that meet the preset clustering rules, and obtain at least one set of width sequences. If the width of all roads in the current width sequence is greater than the width of the last road in the adjacent preceding width sequence and greater than the width of the first road in the adjacent following width sequence, then all road widths in the current width sequence are updated to the width of the last road or the width of the first road to obtain the corresponding target width.

6. The method according to claim 5, characterized in that, The process of forming a passable left boundary of a road based on multiple left road boundary points and forming a passable right boundary of a road based on multiple right road boundary points includes: The trajectory point is moved along a first preset direction according to the corresponding left target width to obtain the corresponding left road boundary point; and the trajectory point is moved along a second preset direction according to the corresponding right target width to obtain the corresponding right road boundary point. Connect the left road boundary points in sequence to generate the left boundary of the passable road; connect the right road boundary points in sequence to generate the right boundary of the passable road.

7. The method according to claim 5, characterized in that, Before calculating the difference in width between any two adjacent roads on the same side, the method further includes: The preset distances between each trajectory point and the starting point of the vehicle's movement trajectory are obtained respectively; Based on the corresponding preset distance values, the widths of the roads on the same side are sorted in ascending order to obtain the sequentially arranged road widths on the same side.

8. The method according to any one of claims 1 to 7, characterized in that: The road boundary includes at least one of ground markings, physical dividing facilities, or the boundary outline of a building.

9. An application of a target navigable road constructed by a method for constructing a vehicle access road according to any one of claims 1 to 8.

10. A device for constructing a vehicle passageway, characterized in that, include: The data acquisition module is used to acquire the vehicle's movement trajectory and the road boundaries distributed on both sides of the vehicle's movement trajectory based on the vehicle's driving route. A boundary point generation module is used to obtain multiple left-side road boundary points and multiple right-side road boundary points based on multiple trajectory points sampled on the vehicle's movement trajectory and the road width corresponding to each trajectory point and the road boundary. The road widths include both left-side and right-side road widths, determined by the road distance between the intersections of the trajectory points and the road boundaries, and both the left-side and right-side road widths are less than or equal to a preset distance threshold. The module obtains the corresponding left-side road boundary point by moving the trajectory point along a first preset direction according to the corresponding left-side road width, and obtains the corresponding right-side road boundary point by moving the trajectory point along a second preset direction according to the corresponding right-side road width, where both the left-side and right-side road widths are less than or equal to the preset distance threshold. The road generation module is used to form a left boundary of a passable road based on multiple left road boundary points, and to form a right boundary of a passable road based on multiple right road boundary points, so that the vehicle can travel within the target passable road formed by the left boundary of the passable road and the right boundary of the passable road.

11. A vehicle, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-8.

12. A computer-readable storage medium having executable code stored thereon, characterized in that: When the executable code is executed by the processor of the electronic device, the processor performs the method as described in any one of claims 1-8.

Images