Intersection turning path reference line generation method, device, vehicle and storage medium

Abstract

The application relates to a method and device for generating a turning path reference line at an intersection, a vehicle and a storage medium. The method comprises the following steps: determining a first endpoint and a second endpoint of a turning path reference line at a target turning intersection; obtaining a curvature value of the first endpoint; obtaining a spiral line parameter value of a preset spiral line formula; and obtaining position data of a plurality of trajectory reference points on a spiral line corresponding to the preset spiral line formula between the first endpoint and the second endpoint according to the curvature value of the first endpoint and the spiral line parameter value. The scheme provided by the application can generate a turning path reference line at an intersection that conforms to the motion behavior of a vehicle, so that an autonomous vehicle can reasonably plan a trajectory and improve the safety of the autonomous vehicle passing through the intersection.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a method, device, vehicle, and storage medium for generating reference lines for turning paths at intersections. Background Technology

[0002] With the development of artificial intelligence, vehicles are moving towards achieving autonomous driving. Reference lines (usually the road centerline) serve as guide lines for trajectory planning of autonomous vehicles in structured roads, playing a crucial role in autonomous driving technology.

[0003] In urban autonomous driving scenarios, intersections present unique and complex challenges, and smooth traffic flow at intersections is a crucial evaluation metric. While high-precision map-based autonomous driving solutions exist, the road lines within intersections provided by these maps are manually drawn, resulting in high creation and maintenance costs and a lack of adaptability. Alternatively, high-precision maps may not even provide road lines within intersections, preventing autonomous vehicles from rationally planning their trajectories and creating safety hazards when navigating intersections. Summary of the Invention

[0004] To address or partially address the problems existing in related technologies, this application provides a method, device, vehicle, and storage medium for generating intersection turning path reference lines. This method can generate turning path reference lines at intersections that conform to vehicle motion behavior, enabling autonomous vehicles to rationally plan their trajectories and improving the safety of autonomous vehicles passing through intersections.

[0005] This application provides a method for generating a reference line for turning paths at an intersection, the method comprising:

[0006] Determine the first and second endpoints of the turning path reference line at the target turning intersection;

[0007] Obtain the curvature value of the first endpoint;

[0008] Obtain the spiral parameter values ​​of the preset spiral formula;

[0009] Based on the curvature value of the first endpoint and the spiral parameter value, the position data of multiple trajectory reference points on the spiral line corresponding to the preset spiral formula between the first endpoint and the second endpoint are obtained.

[0010] In one embodiment, obtaining the curvature value of the first endpoint includes:

[0011] Based on the position data and orientation angle of the first endpoint, and the position data of the second endpoint, the radius of an imaginary circle tangent to the first endpoint and intersecting the second endpoint is obtained; and the curvature value of the first endpoint is obtained by using the radius of the imaginary circle as the radius of curvature of the first endpoint; or,

[0012] Obtain the size data of the target turning intersection, and based on the size data and the correspondence between multiple pre-stored intersection size data and multiple curvature values, select the curvature value of the first endpoint from the multiple curvature values.

[0013] In one embodiment, obtaining the spiral parameter values ​​of the preset spiral formula includes:

[0014] Based on the curvature value of the first endpoint and the orientation angle of the first endpoint, the spiral parameter values ​​of the preset spiral formula are obtained.

[0015] In one embodiment, the preset cyclotron formula is the Fresnel integral formula;

[0016] The step of obtaining position data of multiple trajectory reference points on the spiral line corresponding to the preset spiral line formula between the first and second endpoints based on the curvature value of the first endpoint and the spiral line parameter value includes:

[0017] Based on the curvature value of the first endpoint, the integral interval of the preset spiral formula corresponding to the first endpoint is obtained;

[0018] By taking multiple values ​​between the integration intervals, multiple integration intervals are obtained for multiple trajectory reference points on the spiral line corresponding to the preset spiral line formula;

[0019] For each of the plurality of integration intervals, based on the spiral parameter value, the position data of the corresponding trajectory reference point is obtained through a preset solution algorithm corresponding to the Fresnel integral formula.

[0020] In one embodiment, the first endpoint is the end point of the turning path reference line, and the second endpoint is the start point of the turning path reference line;

[0021] The step of obtaining the integration interval of the preset spiral formula corresponding to the first endpoint based on the curvature value of the first endpoint includes: obtaining the upper limit of the integration of the Fresnel integral formula corresponding to the endpoint based on the curvature value of the endpoint;

[0022] The step of taking multiple values ​​between the integration intervals to obtain multiple integration intervals corresponding to multiple trajectory reference points on the spiral line of the preset spiral line formula includes: uniformly taking values ​​between the upper and lower limits of integration corresponding to the Fresnel integral formula and the endpoint to obtain multiple upper limits of integration corresponding to multiple trajectory reference points on the spiral line of the preset spiral line formula.

[0023] In one embodiment, the turning path reference line is a first turning path reference line; the first endpoint of the first turning path reference line is located at the first exit lane line of the target turning intersection, and the second endpoint of the first turning path reference line is located at the first entry lane line of the target turning intersection.

[0024] The method further includes:

[0025] Determine the first and second endpoints of the second turning path reference line at the target turning intersection;

[0026] Based on the position data of the first and second endpoints of the first turning path reference line, the plurality of trajectory reference points, and the first and second endpoints of the second turning path reference line, the position data of the plurality of trajectory reference points between the first and second endpoints of the second turning path reference line are obtained through translation processing.

[0027] In one embodiment, obtaining the position data of multiple trajectory reference points between the first and second endpoints of the second turning path reference line through translation processing, based on the position data of the first and second endpoints of the first turning path reference line, the plurality of trajectory reference points, and the first and second endpoints of the second turning path reference line, includes:

[0028] Obtain the first Euclidean distance between the first endpoint of the first turning path reference line and the first endpoint of the second turning path reference line;

[0029] Obtain the second Euclidean distance between the second endpoint of the first turning path reference line and the second endpoint of the second turning path reference line;

[0030] Based on the first Euclidean distance, the second Euclidean distance, and the number of translation trajectory reference points of the first turning path reference line, obtain the translation distance of each point among the m translation trajectory reference points of the first turning path reference line;

[0031] Based on the position data and translation distance of each of the m translation trajectory reference points of the first turning path reference line, the coordinates of the m trajectory reference points of the second turning path reference line are obtained.

[0032] In one embodiment, the first endpoint is the end point of the corresponding turning path reference line, and the second endpoint is the start point of the corresponding turning path reference line;

[0033] The starting and ending points of the first turning path reference line and the second turning path reference line are different. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (1). Pj :

[0034] d Pj =d S1S2 +j*(d T1T2 -d S1S2 ) / m, j=1,2,3,…,m; Formula (1),

[0035] Where, d S1S2 Let d be the Euclidean distance between the starting points of the first turning path reference line and the second turning path reference line. T1T2 The Euclidean distance between the endpoints of the first turning path reference line and the second turning path reference line;

[0036] or,

[0037] The first turning path reference line and the second turning path reference line have different starting points but the same ending point. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (2). Pj :

[0038] d Pj =d S1S2 -j*d S1S2 / m, j = 1, 2, 3, ..., m; Formula (2);

[0039] or,

[0040] The first turning path reference line and the second turning path reference line have different endpoints but the same starting point. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (3). Pj :

[0041] d Pj =j*d T1T2 / m, j=1,2,3,…,m. Formula (3).

[0042] In one embodiment, one of the first endpoint and the second endpoint is the starting point of the turning path reference line, and the other is the ending point of the turning path reference line.

[0043] The starting point of the turning path reference line is the end point of the corresponding entry lane line, or the intersection of the corresponding entry lane line and the entry stop line of the target turning intersection.

[0044] The endpoint of the turning path reference line is the starting point of the corresponding exit lane line, or the intersection of the corresponding exit lane line and the intersection frame of the target turning intersection.

[0045] The entry lane line is the driving lane line before the vehicle enters the target turning intersection, and the exit lane line is the driving lane line after the vehicle leaves the target turning intersection.

[0046] This application also provides a computing device, including:

[0047] At least one processor; and

[0048] At least one memory storing executable code, which, when executed by the at least one processor, causes the at least one processor to perform the method as described above.

[0049] This application also provides an autonomous vehicle, including the computing device described above.

[0050] This application also provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor, causes the processor to perform the method described above.

[0051] The technical solution provided in this application may include the following beneficial effects:

[0052] The technical solution of this application can generate a turning path reference line connecting the entering lane line and the exit lane line in the target turning intersection area based on the entering lane line and the exit lane line of the intersection. The generated turning path reference line can conform to the vehicle's motion behavior, and the curvature of the turning path reference line is continuous and the curvature change is smooth, making the planned autonomous driving turning path more reasonable and improving the safety of autonomous vehicles passing through intersections.

[0053] 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

[0054] 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.

[0055] Figure 1 This is a flowchart illustrating a method for generating a reference line for a turning path at an intersection according to an embodiment of this application.

[0056] Figure 2 This is a flowchart illustrating a method for generating a reference line for a turning path at an intersection, according to another embodiment of this application.

[0057] Figure 3 yes Figure 2 A schematic diagram of a scenario in an embodiment;

[0058] Figure 4 This is a schematic diagram of a turning intersection scenario according to another embodiment of this application;

[0059] Figure 5 This is a schematic diagram of a turning intersection scenario according to another embodiment of this application;

[0060] Figure 6 This is a schematic diagram of the structure of a computing device according to an embodiment of this application. Detailed Implementation

[0061] 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.

[0062] 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.

[0063] 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.

[0064] The related technologies for autonomous driving solutions based on high-precision maps have limitations. Because the road lines within intersections provided by high-precision maps are manually drawn, the cost of drawing and maintaining them is very high and they lack adaptability. Alternatively, high-precision maps may not provide road lines within intersections at all, causing autonomous vehicles to be unable to plan their trajectories reasonably, which poses safety hazards for autonomous vehicles passing through intersections.

[0065] To address the aforementioned issues, this application provides a method for generating intersection turning path reference lines. This method can generate turning path reference lines at intersections that conform to vehicle motion behavior, enabling autonomous vehicles to rationally plan their trajectories and improving the safety of autonomous vehicles passing through intersections.

[0066] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0067] Figure 1 This is a flowchart illustrating a method for generating a reference line for a turning path at an intersection according to an embodiment of this application.

[0068] See Figure 1 A method for generating reference lines for turning paths at intersections, applied to a computing device, includes:

[0069] In S110, the first and second endpoints of the turning path reference line of the target turning intersection are determined.

[0070] Understandably, the target turning intersection can be a single left-turn intersection, a single right-turn intersection, or a bidirectional turning intersection where both left and right turns are permitted; the turning path reference line can be a left-turn path reference line for a single left-turn intersection, a right-turn path reference line for a single right-turn intersection, or a left-turn path reference line or a right-turn path reference line for a bidirectional turning intersection.

[0071] In S120, the curvature value of the first endpoint is obtained, as well as the spiral parameter value of the preset spiral formula is obtained.

[0072] In some embodiments, obtaining the curvature value of the first endpoint includes: obtaining the radius of an imaginary circle tangent to the first endpoint and intersecting the second endpoint based on the position data and orientation angle of the first endpoint, and the position data of the second endpoint; and obtaining the curvature value of the first endpoint using the radius of the imaginary circle as the curvature radius of the first endpoint. It is understood that the curvature value of the first endpoint can be obtained in other ways. For example, in other embodiments, the curvature value of the first endpoint can be selected from the multiple curvature values ​​based on the size data of the target turning intersection and the pre-stored correspondence between multiple intersection size data and multiple curvature values. For example, the curvature values ​​of the first endpoints corresponding to multiple intersections with different aspect ratios can be obtained in advance, and then the matching curvature value of the first endpoint can be obtained by looking up a table based on the aspect ratio of the target turning intersection.

[0073] In some embodiments, the preset cyclotron formula is the Fresnel integral formula.

[0074] In S130, based on the curvature value of the first endpoint and the spiral parameter value, position data of multiple trajectory reference points on the spiral line corresponding to the preset spiral formula between the first endpoint and the second endpoint are obtained.

[0075] In this embodiment, the turning path reference line adopts the form of a spiral curve. The position data of multiple trajectory reference points on the spiral curve corresponding to the preset spiral curve formula between the first and second endpoints can represent the turning path reference line. Furthermore, autonomous driving trajectory planning for the target turning intersection can be performed based on the obtained turning path reference line. It is understood that the autonomous driving trajectory planning for the target turning intersection can directly use the turning path reference line as the planned turning path, or it can be obtained by performing preset post-processing on the turning path reference line. In this way, the turning path reference line has the characteristic of continuous and linear curvature change; and since the curvature of the vehicle's trajectory during operation is basically proportional to the steering wheel, autonomous driving trajectory planning for the target turning intersection based on the obtained turning path reference line has the advantages of curvature friendliness and comfort.

[0076] It is understood that in this application, the computing device includes a processor and a memory storing a computer program. When the processor executes the stored computer program, it can implement a method for generating reference lines for turning paths at intersections. The computing device can be one or more computer terminals, a server, or a combination of computer terminals and servers. It is understood that a server can be a single physical server or a logical server virtualized from multiple physical servers. A server can also be a server cluster composed of multiple interconnected servers, with each functional module distributed across different servers in the server cluster. In some embodiments, the computing device is an in-vehicle electronic device, such as, but not limited to, a vehicle's electronic control unit, an autonomous driving system controller, a smart navigation device, a smartphone, a smart tablet, or other mobile devices.

[0077] In some embodiments, the first endpoint is the starting point of the turning path reference line, and the second endpoint is the ending point of the turning path reference line.

[0078] In other embodiments, the first endpoint is the end point of the turning path reference line, and the second endpoint is the start point of the turning path reference line.

[0079] In some embodiments, the starting point of the turning path reference line is the ending point of the corresponding entering lane line, or the intersection of the corresponding entering lane line and the entry stop line of the target turning intersection. The turning path reference line can connect the entering lane line and the exit lane line. The ending point of the turning path reference line is the starting point of the corresponding exit lane line, or the intersection of the corresponding exit lane line and the intersection frame of the target turning intersection. The entering lane line is the driving lane line before the vehicle enters the target turning intersection, and the exit lane line is the driving lane line after the vehicle leaves the target turning intersection. In other words, the vehicle enters the target turning intersection along the entering lane line and leaves the target turning intersection along the exit lane line. The entering lane line and the exit lane line can be the lane center line of the lane the vehicle is traveling on, or other types of lane reference lines for the lane the vehicle is traveling on, such as lane edge lines, lane dividing lines, etc.

[0080] According to the embodiments of this application, a turning path reference line connecting the entering lane line and the exit lane line in the target turning intersection area can be generated based on the entering lane line and the exit lane line of the intersection. The generated turning path reference line can conform to the vehicle's motion behavior, and the curvature of the turning path reference line is continuous and the curvature change is smooth, making the planned autonomous driving turning path more reasonable and improving the safety of autonomous vehicles passing through intersections.

[0081] Figure 2 This is a flowchart illustrating a method for generating a reference line for a turning path at an intersection, according to another embodiment of this application. Figure 3 yes Figure 2 A schematic diagram of the scenario in the embodiment. The following description uses the example where the first endpoint is the end point of the turning path reference line and the second endpoint is the start point of the turning path reference line.

[0082] Combination Figure 3 In this embodiment, the generation of a first turning path reference line 3051 and a second turning path reference line 3052 at a left-turn intersection is used as an example for illustration. The figure shows a two-lane intersection, where the intersection frame 301 indicates the intersection range of the left-turn intersection. This left-turn intersection has two entering lane lines and two exit lane lines, namely a first entering lane line 3021, a second entering lane line 3022, a first exit lane line 3031, and a second exit lane line 3032. According to the embodiment of this application, a first turning path reference line 3051 connecting the first entering lane line 3021 and the first exit lane line 3031 can be generated, and a second turning path reference line 3052 connecting the second entering lane line 3022 and the second exit lane line 3032 can be further generated.

[0083] In one specific implementation, an offline execution method can be used to pre-generate the required turning path reference lines for each turning intersection in the high-precision map. Understandably, the generated turning path reference lines can be used as part of the high-precision map data for autonomous vehicles to directly use as turning paths during autonomous driving, or they can be post-processed and used as turning paths as needed.

[0084] In another specific implementation, the autonomous vehicle generates a turning path reference line in real time based on its current lane (which is the entry lane) and the upcoming turning intersection and exit lane, and then plans the turning path in real time based on the generated turning path reference line.

[0085] See Figure 2 A method for generating a reference line for turning paths at an intersection, comprising:

[0086] In S210, the start and end points of the first turning path reference line at the target turning intersection are determined.

[0087] In some embodiments, road sign information of the vehicle's current driving route can be obtained based on the vehicle's positioning information and high-precision map data. Based on this road sign information, the upcoming turning intersection and related road surface information can be determined. The starting and ending points of the first turning path reference line for the upcoming turning intersection can be determined based on the road surface information. The road surface information related to the upcoming turning intersection may include the entry lane lines, exit lane lines, entry stop lines, and intersection frames. The vehicle's positioning information may include latitude and longitude information. High-precision map data can be stored on the autonomous vehicle's onboard equipment or in the cloud. The autonomous vehicle can obtain high-precision map data matching its current latitude and longitude information; and obtain road surface information related to the upcoming turning intersection based on the high-precision map data.

[0088] In other embodiments, the autonomous vehicle includes radar and / or cameras, which can obtain road surface information related to the turning intersection ahead of the vehicle's current driving road through data collected by the radar and / or cameras.

[0089] exist Figure 3 In the example shown, road surface information related to the left-turn intersection ahead of the vehicle's current driving road is obtained based on the vehicle's positioning information and high-precision map data. This includes obtaining the intersection frame 301 of the left-turn intersection, the first entry lane line 3021 and the second entry lane line 3022 of the left-turn intersection, the first exit lane line 3031 and the second exit lane line 3032, and the entry stop line 304 of the left-turn intersection.

[0090] In one embodiment, the intersection of the first entering lane line 3021 and the intersection stop line 304 can be determined as the starting point S1 of the first turning path reference line, and the intersection of the intersection frame 301 and the first exit lane line 3031 can be determined as the ending point T1 of the first turning path reference line.

[0091] In S220, the curvature value of the endpoint is obtained based on the position data of the starting point of the first turning path reference line, the position data of the endpoint, and the orientation angle.

[0092] Curvature represents the degree of curvature of a curve at a given point. The larger the curvature value, the greater the curvature of the curve. Curvature can be obtained using the formula k = 1 / R, where k is the curvature and R is the radius of curvature at that point.

[0093] Figure 3 In the example shown, the radius of an imaginary circle that is tangent to the end point T1 and intersects the start point S1 can be obtained based on the position data and orientation angle of the end point T1 of the first turning path reference line and the position data of the start point S1; and the curvature value of the end point T1 can be obtained by using the radius of the imaginary circle as the curvature radius of the end point T1.

[0094] In one embodiment, point attribute information of each point on the first entering lane line and the second exit lane line can be obtained from high-precision map data. The point attribute information includes at least the coordinates (X, Y, X) of the point. P Y P ), orientation angle θ.

[0095] In S230, the spiral parameter values ​​of the preset spiral formula are obtained based on the curvature value of the endpoint and the orientation angle of the endpoint.

[0096] The turning path reference line adopts the form of a spiral, which is a curve whose curvature changes proportionally with the curve length. A spiral can be represented by a preset spiral formula, which contains a spiral parameter indicating the rate of change in the spiral's curvature. For a single spiral, the spiral parameter is a constant.

[0097] In some embodiments, the preset cyclotron formula can be the Fresnel integral formula.

[0098] In some embodiments, the coordinates (X, Y) of any point on the turning path reference line can be obtained using the following Fresnel integral formula:

[0099]

[0100]

[0101] In the formula, the interval of t represents the range of curvature variation, and a is the spiral parameter. The relationship between the arc length s and t of the spiral satisfies the formula s = a * t. Based on the time attribute of t, the relationship between curvature k and a, t conforms to the formula k = π * t / a, which can be transformed to obtain: t = a * k / π; substituting into the above integral formula, we can obtain the following integral formula:

[0102]

[0103]

[0104] Additionally, based on the slope tanθ of the endpoint T1 of the first turning path reference line... T1 The condition that the slopes of points of tangency on a circle are equal determines that the orientation angle θ and t satisfy the formula: θ = t 2 *π / 2.

[0105] According to t = a * k / π, θ = t 2 *π / 2, obtain the spiral parameters.

[0106] The orientation angle θ of the endpoint T1 T1 and curvature value k T1 Substituting the value into the above formula, we can proceed according to the formula. Calculate the value of the spiral parameter a.

[0107] In S240, based on the curvature value of the endpoint of the first turning path reference line, the integral interval corresponding to the endpoint of the preset spiral formula is obtained, and the value is uniformly taken within the integral interval to obtain multiple integral upper limits corresponding to multiple trajectory reference points.

[0108] The spiral formula is a variable upper limit integral formula, where the integral formula corresponds to the upper limit of integration t = a*k at the endpoint T1. T1 / π, the lower limit of integration is 0, that is, the integration interval corresponding to the endpoint of the pre-defined spiral formula is [0, a*k]. T1 / π]; n values ​​can be uniformly selected between the integration intervals corresponding to the lower and upper limits of integration to obtain n integration upper limits corresponding to n trajectory reference points.

[0109] In S250, for each of the multiple integration upper limits, that value is used as the integration upper limit of the Fresnel integral formula mentioned above. Based on the calculated spiral line parameter value, the position data of the corresponding trajectory reference point is obtained through the preset solution algorithm corresponding to the Fresnel integral formula.

[0110] In one embodiment, the coordinates (Xi, Yi) of each trajectory reference point Pi on the first turning path reference line between the starting point S1 and the ending point T1 can be obtained by substituting different i (i = 1, 2, ..., n) into the following formula:

[0111]

[0112]

[0113] Where n is the number of sampled trajectory reference points on the first turning path reference line.

[0114] Understandably, in other embodiments, the spiral curve formula can be a variable lower limit integral formula. The integral interval corresponding to one endpoint of the first turning path reference line can be obtained based on the curvature value of that endpoint. Then, the value is uniformly taken within the integral interval to obtain multiple integral lower limits corresponding to multiple trajectory reference points. Based on the calculated spiral curve parameter values, the position data of the corresponding trajectory reference points can be obtained through a preset solution algorithm corresponding to the variable lower limit integral formula.

[0115] In some embodiments, the coordinates of multiple trajectory reference points of a second turning path reference line connecting the second entering lane line and the second exit lane line can be obtained by translation processing based on the first turning path reference line.

[0116] In some embodiments, based on the first turning path reference line, the coordinates of multiple trajectory reference points of the second turning path reference line connecting the second entering lane line and the second exiting lane line are obtained through translation processing, including:

[0117] Determine the first and second endpoints of the second turning path reference line at the target turning intersection; and

[0118] Based on the position data of the first and second endpoints of the first turning path reference line, and multiple trajectory reference points, as well as the first and second endpoints of the second turning path reference line, the position data of multiple trajectory reference points between the first and second endpoints of the second turning path reference line are obtained through translation processing.

[0119] For example, Figure 3 In the scenario shown, multiple trajectory reference points for the second turning path reference line 3052 can be obtained through translation processing, based on the starting point S1, ending point T1, and multiple trajectory reference points of the first turning path reference line 3051, and the starting point S2 and ending point T2 of the second turning path reference line 3052. It is understandable that the determination of the two endpoints S2 and T2 of the second turning path reference line 3052 and the acquisition of their point attribute information can be achieved by referring to the two endpoints S1 and T1 of the first turning path reference line 3051, and will not be elaborated further.

[0120] In some embodiments, the translation process includes:

[0121] SA1, obtain the first Euclidean distance between the first endpoint of the first turning path reference line and the first endpoint of the second turning path reference line.

[0122] SA2, obtain the second Euclidean distance between the second endpoint of the first turning path reference line and the second endpoint of the second turning path reference line.

[0123] SA3, based on the first Euclidean distance, the second Euclidean distance, and the number of translation trajectory reference points of the first turning path reference line, obtain the translation distance of each of the m translation trajectory reference points of the first turning path reference line.

[0124] In this embodiment, after determining the translation distance of m translation trajectory reference points of the first turning path reference line, each translation trajectory reference point is translated to obtain m trajectory reference points of the second turning path reference line.

[0125] Understandably, the m translation trajectory reference points of the first turning path reference line can be the n trajectory reference points Pi on the first turning path reference line between the starting point S1 and the ending point T1 obtained in S250 above, i.e., m = n; or, they can be added, deleted, or uniformly sampled again according to m based on the n trajectory reference points Pi of the first turning path reference line.

[0126] For example, the value of m can be determined using the following formula, based on the set sampling distance delta_l and the length S of the first turning path reference line 3051:

[0127] m = S / delta_l.

[0128] SA4: Based on the position data and translation distance of each of the m translation trajectory reference points of the first turning path reference line, obtain the coordinates of the m trajectory reference points of the second turning path reference line.

[0129] See Figure 3 As shown, starting point S1 is located at the first entering lane line 3021, starting point S2 is located at the second entering lane line 3022, ending point T1 is located at the first exit lane line 3031, and ending point T2 is located at the second exit lane line 3032. Starting point S1 and starting point S2 are both on the stop line 304, and ending point T1 and ending point T2 are both on the left-turn exit side of intersection frame 301. The coordinates of starting point S1, starting point S2, ending point T1, and ending point T2 are obtained through high-precision map data; based on the coordinates of starting point S1 and starting point S2, the first Euclidean distance d between starting point S1 and starting point S2 is calculated. S1S2 ; Calculate the second Euclidean distance d between endpoint T1 and endpoint T2 based on the coordinates of endpoint T1 and endpoint T2. T1T2 Based on the first Euclidean distance d between the two starting points. S1S2The second Euclidean distance d between the two endpoints T1T2 The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated using the following formula. Pj ,

[0130] d Pj =d S1S2 +j*(d T1T2 -d S1S2 ) / m; j=1, 2, 3,..., m;

[0131] Where m is the total number of translation trajectory reference points of the first turning path reference line.

[0132] Understandably, the translation distance d of the translation trajectory reference point Pj is... Pj It is the length of the translation trajectory reference point Pj extended in the direction of the normal vector at that point.

[0133] The coordinates of the trajectory reference point Pj' obtained after translating the translation trajectory reference point Pj are (X... i ', Y i '),in:

[0134] X j =X Pj +d Pj *cos(θ Pj -π / 2);

[0135] Y j '=Y Pj +d Pj *sin(θ Pj -π / 2).

[0136] Among them, X Pj Y Pj θ represents the coordinates of the reference point Pj on the translation trajectory. Pj This indicates the orientation angle of point Pj.

[0137] It is understood that the second turning path reference line is not limited to the outside of the first turning path reference line. In other embodiments, the first turning path reference line can be translated inward to obtain the second turning path reference line inside the first turning path reference line.

[0138] According to the embodiments of this application, a turning path reference line connecting the entering lane line and the exit lane line in the target turning intersection area can be generated based on the entering lane line and the exit lane line of the intersection. The generated turning path reference line can conform to the vehicle's motion behavior, and the curvature of the turning path reference line is continuous and the curvature change is smooth, making the planned autonomous driving turning path more reasonable and improving the safety of autonomous vehicles passing through intersections.

[0139] Furthermore, according to the embodiments of this application, after generating the first turning path reference line according to the Fresnel integral formula, a simple translation process is used to generate corresponding turning path reference lines for other lanes within the target turning intersection range; in this way, turning path reference lines connecting all corresponding entering and exiting lane lines of the intersection can be generated quickly, and other turning path reference lines obtained by translation processing also have the characteristics of continuous curvature and smooth curvature change.

[0140] Figure 4 This diagram illustrates another type of intersection. (Example:) Figure 4 As shown in the figure, the left-turn intersection is a type of merged intersection, which is an intersection with multiple entering lanes and one or more exit lanes, where the number of exit lanes is less than the number of entering lanes. In this embodiment, the generation of a turning path reference line for a left-turn merged intersection is used as an example. The figure uses two entering lanes and one exit lane as an example. The intersection frame 501 shows the intersection range of the left-turn intersection, which has two entering lane lines and one exit lane line, namely, the first entering lane line 5021, the second entering lane line 5022, and the exit lane line 5031. According to the embodiments of this application, a first turning path reference line 5051 connecting the first entering lane line 5021 and the exit lane line 5031, and a second turning path reference line 5052 connecting the second entering lane line 5022 and the exit lane line 5031 can be generated. The starting point S1 of the first turning path reference line 5051 is the intersection of the first entering lane line 5021 and the intersection entry stop line 504. The ending point T1 of the first turning path reference line 5051 is the intersection of the exit lane line 5031 and the intersection frame 501. The starting point S2 of the second turning path reference line 5052 is the intersection of the second entering lane line 5022 and the intersection entry stop line 504. The second turning path reference line 5052 has the same ending point as the first turning path reference line 5051.

[0141] In some embodiments, the position data of multiple trajectory reference points of the first turning path reference line 5051 can be obtained according to steps S210 to S250. Then, according to steps SA1 to SA4, the m horizontal trajectory reference points of the first turning path reference line 5051 are translated to obtain the coordinates of the m trajectory reference points of the second turning path reference line 5052. The difference is that, since the second turning path reference line 5052 and the first turning path reference line 5051 have the same endpoint, the second Euclidean distance d between the two endpoints is... T1T2 The value is 0, therefore the translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is 0. Pj for:

[0142] d Pj =d S1S2-j*d S1S2 / m; j=1, 2, 3,..., m.

[0143] Figure 5 This diagram illustrates another type of intersection. (Example:) Figure 5 As shown in the figure, the left-turn intersection is a type of split intersection, which is an intersection with one or more entering lanes and multiple exit lanes, where the number of entering lanes is less than the number of exit lanes. In this embodiment, the generation of a turning path reference line for a left-turn split intersection is used as an example. The figure uses one entering lane and two exit lanes as an example. The intersection frame 701 shows the intersection range of the left-turn intersection, which has one entering lane line and two exit lane lines, namely, entering lane line 7021, first exit lane line 7031, and second exit lane line 7032. According to the embodiments of this application, a first turning path reference line 7051 connecting the entering lane line 7021 and the first exit lane line 7031, and a second turning path reference line 7052 connecting the entering lane line 7021 and the second exit lane line 7032 can be generated. The starting point S1 of the first turning path reference line 7051 is the intersection of the entry lane line 7021 and the intersection entry stop line 704. The ending point T1 of the first turning path reference line 7051 is the intersection of the first exit lane line 7031 and the intersection frame 701. The second turning path reference line 7052 has the same starting point S1 as the first turning path reference line 7051. The ending point T2 of the second turning path reference line 7052 is the intersection of the second exit lane line 7032 and the intersection frame 701.

[0144] In some embodiments, the position data of multiple trajectory reference points of the first turning path reference line 7051 can be obtained according to steps S210 to S250. Then, according to steps SA1 to SA4, the m horizontal trajectory reference points of the first turning path reference line 7051 are translated to obtain the coordinates of the m trajectory reference points of the second turning path reference line 7052. The difference is that since the second turning path reference line 7052 and the first turning path reference line 7051 have the same starting point, the first Euclidean distance d between the two starting points is... S1S2 The value is 0, therefore the translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is 0. Pj for:

[0145] d Pj =j*d T1T2 / m; j=1, 2, 3,..., m.

[0146] Corresponding to the aforementioned application function implementation method embodiments, this application also provides a computing device, a vehicle, and corresponding embodiments.

[0147] Figure 6This is a schematic diagram of the structure of a computing device shown in an embodiment of this application.

[0148] See Figure 6 The computing device 900 includes a memory 901 and a processor 902.

[0149] The processor 902 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. A general-purpose processor can be a microprocessor or any conventional processor.

[0150] Memory 901 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 902 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 901 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, memory 901 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.

[0151] The memory 901 stores executable code, which, when processed by the processor 902, can cause the processor 902 to execute part or all of the methods described above.

[0152] In addition, this application also provides an autonomous vehicle, including the computing device described above.

[0153] 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.

[0154] Alternatively, this application may also 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 a vehicle (or server, etc.), causes the processor to perform part or all of the steps of the above-described method according to this application.

[0155] 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 generating reference lines for turning paths at intersections, applied to a computing device, characterized in that, include: Determine the first and second endpoints of the turning path reference line at the target turning intersection; Obtain the curvature value of the first endpoint; Obtain the spiral parameter values ​​of the preset spiral formula; The preset cyclotron formula is the Fresnel integral formula; Based on the curvature value of the first endpoint and the spiral parameter value, position data of multiple trajectory reference points on the spiral line corresponding to the preset spiral formula between the first endpoint and the second endpoint are obtained, including: Based on the curvature value of the first endpoint, the integral interval of the preset spiral formula corresponding to the first endpoint is obtained; By taking multiple values ​​between the integration intervals, multiple integration intervals are obtained for multiple trajectory reference points on the spiral line corresponding to the preset spiral line formula; For each of the plurality of integration intervals, based on the spiral parameter value, the position data of the corresponding trajectory reference point is obtained through a preset solution algorithm corresponding to the Fresnel integral formula.

2. The method according to claim 1, characterized in that, Obtaining the curvature value of the first endpoint includes: Based on the position data and orientation angle of the first endpoint, and the position data of the second endpoint, the radius of an imaginary circle tangent to the first endpoint and intersecting the second endpoint is obtained; and the curvature value of the first endpoint is obtained by using the radius of the imaginary circle as the radius of curvature of the first endpoint; or, Obtain the size data of the target turning intersection, and based on the size data and the correspondence between multiple pre-stored intersection size data and multiple curvature values, select the curvature value of the first endpoint from the multiple curvature values.

3. The method according to claim 1, characterized in that, The process of obtaining the spiral parameter values ​​of the preset spiral formula includes: Based on the curvature value of the first endpoint and the orientation angle of the first endpoint, the spiral parameter values ​​of the preset spiral formula are obtained.

4. The method according to claim 1, characterized in that: The first endpoint is the end point of the turning path reference line, and the second endpoint is the start point of the turning path reference line; The step of obtaining the integration interval of the preset spiral formula corresponding to the first endpoint based on the curvature value of the first endpoint includes: obtaining the upper limit of the integration of the Fresnel integral formula corresponding to the endpoint based on the curvature value of the endpoint; The step of taking multiple values ​​between the integration intervals to obtain multiple integration intervals corresponding to multiple trajectory reference points on the spiral line of the preset spiral line formula includes: uniformly taking values ​​between the upper and lower limits of integration corresponding to the Fresnel integral formula and the endpoint to obtain multiple upper limits of integration corresponding to multiple trajectory reference points on the spiral line of the preset spiral line formula.

5. The method according to any one of claims 1 to 4, characterized in that, The turning path reference line is a first turning path reference line; the first endpoint of the first turning path reference line is located at the first exit lane line of the target turning intersection, and the second endpoint of the first turning path reference line is located at the first entry lane line of the target turning intersection. The method further includes: Determine the first and second endpoints of the second turning path reference line at the target turning intersection; Based on the position data of the first and second endpoints of the first turning path reference line, the plurality of trajectory reference points, and the first and second endpoints of the second turning path reference line, the position data of the plurality of trajectory reference points between the first and second endpoints of the second turning path reference line are obtained through translation processing.

6. The method according to claim 5, characterized in that, The step of obtaining the position data of multiple trajectory reference points between the first and second endpoints of the first turning path reference line, the position data of the plurality of trajectory reference points, and the first and second endpoints of the second turning path reference line through translation processing includes: Obtain the first Euclidean distance between the first endpoint of the first turning path reference line and the first endpoint of the second turning path reference line; Obtain the second Euclidean distance between the second endpoint of the first turning path reference line and the second endpoint of the second turning path reference line; Based on the first Euclidean distance, the second Euclidean distance, and the number of translation trajectory reference points of the first turning path reference line, obtain the translation distance of each point among the m translation trajectory reference points of the first turning path reference line; Based on the position data and translation distance of each of the m translation trajectory reference points of the first turning path reference line, the coordinates of the m trajectory reference points of the second turning path reference line are obtained.

7. The method according to claim 6, characterized in that, The first endpoint is the end point of the corresponding turning path reference line, and the second endpoint is the start point of the corresponding turning path reference line; The starting and ending points of the first turning path reference line and the second turning path reference line are different. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (1). Pj : d Pj = d S1S2 + j* (d T1T2 - d S1S2 ) / m, j=1,2,3,…,m; Formula (1). Where, d S1S2 Let d be the Euclidean distance between the starting points of the first turning path reference line and the second turning path reference line. T1T2 The Euclidean distance between the endpoints of the first turning path reference line and the second turning path reference line; or, The first turning path reference line and the second turning path reference line have different starting points but the same ending point. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (2). Pj : d Pj = d S1S2 -j* d S1S2 / m, j=1,2,3,…,m; Formula (2); or, The first turning path reference line and the second turning path reference line have different endpoints but the same starting point. The translation distance d of the j-th translation trajectory reference point Pj on the first turning path reference line is calculated by the following formula (3). Pj : d Pj = j* d T1T2 / m, j=1,2,3,…,m; Formula (3).

8. The method according to any one of claims 1 to 4, characterized in that, One of the first endpoint and the second endpoint is the starting point of the turning path reference line, and the other is the ending point of the turning path reference line. The starting point of the turning path reference line is the end point of the corresponding entry lane line, or the intersection of the corresponding entry lane line and the entry stop line of the target turning intersection. The endpoint of the turning path reference line is the starting point of the corresponding exit lane line, or the intersection of the corresponding exit lane line and the intersection frame of the target turning intersection. The entry lane line is the driving lane line before the vehicle enters the target turning intersection, and the exit lane line is the driving lane line after the vehicle leaves the target turning intersection.

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

10. An autonomous vehicle, characterized in that, Includes the computing device as described in claim 9.

11. A computer-readable storage medium, characterized in that, It stores executable code that, when executed by a processor, causes the processor to perform the method as described in any one of claims 1-8.

Images