Automatic driving boundary extension method and extension device

Abstract

The present application relates to the field of autonomous driving technology, and provides an autonomous driving boundary extension method and device. The method obtains a reference line of an autonomous driving vehicle, traverses all reference points in the reference line to obtain global information, and obtains the self-driving vehicle information, determines the value of each flag in the flag set based on the global information and the self-driving vehicle information, and finally traverses the value of each flag in the flag set. When it is determined that the value of the target flag indicates that a boundary extension is currently required, the reference line of the autonomous driving vehicle is extended using a preset strategy corresponding to the target flag, providing an automatic boundary extension method for autonomous driving vehicles in special scenarios, improving the decision-making accuracy of the PnC module, and enhancing the user experience.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to an autonomous driving boundary extension method and extension device. Background Art

[0002] The autonomous driving planning and control (PnC) module needs to use the lane information provided by the map when planning the path, and this lane information will be organized into reference lines according to business needs. In general, the PnC module will read the left and right boundaries of the current position from the reference line as reference information for autonomous driving, but in the case of driving on a borrowed lane, driving on a narrow road, or driving outside the lane, the left and right boundaries provided by the current reference line need to be expanded. Summary of the invention

[0003] In view of this, an embodiment of the present application provides an autonomous driving boundary extension method and an extension device to solve the problem that the prior art cannot provide an efficient autonomous driving boundary automatic extension solution in specific scenarios such as driving on borrowed lanes, driving on narrow roads, and driving outside lanes.

[0004] According to a first aspect of an embodiment of the present application, a method for extending an autonomous driving boundary is provided, comprising:

[0005] Get the reference line of the autonomous driving vehicle, traverse all reference points in the reference line, and get global information;

[0006] Obtaining self-driving vehicle information, and determining the value of each flag bit in the flag bit set based on the global information and the self-driving vehicle information;

[0007] Traversing the values ​​of each flag bit in the flag bit set in turn, and when determining that the value of the target flag bit indicates that a boundary extension is required, using a preset strategy corresponding to the target flag bit to extend the boundary of the reference line of the autonomous driving vehicle;

[0008] The target flag bit is any flag bit in the flag bit set.

[0009] According to a second aspect of an embodiment of the present application, there is provided an autonomous driving boundary expansion device, comprising:

[0010] An acquisition module is configured to acquire a reference line of the autonomous driving vehicle, traverse all reference points in the reference line, and obtain global information;

[0011] The acquisition module is further configured to acquire vehicle information of the autonomous driving vehicle, and determine the value of each flag bit in the flag bit set based on the global information and the vehicle information;

[0012] An extension module is configured to sequentially traverse the values ​​of each flag bit in the flag bit set, and when it is determined that the value of the target flag bit indicates that a boundary extension is required, use a preset strategy corresponding to the target flag bit to perform boundary extension on the reference line of the autonomous driving vehicle;

[0013] The target flag bit is any flag bit in the flag bit set.

[0014] Compared with the prior art, the embodiments of the present application have the following beneficial effects: the embodiments of the present application obtain a reference line of the autonomous driving vehicle, traverse all reference points in the reference line to obtain global information, and obtain the vehicle information of the autonomous driving vehicle, determine the value of each flag bit in the flag bit set based on the global information and the vehicle information, and finally traverse the value of each flag bit in the flag bit set. When it is determined that the value of the target flag bit indicates that boundary extension is currently required, the preset strategy corresponding to the target flag bit is used to extend the boundary of the reference line of the autonomous driving vehicle, thereby providing a method for automatically extending the boundaries of autonomous driving vehicles in special scenarios, improving the decision-making accuracy of the PnC module, and enhancing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 It is a flow chart of an autonomous driving boundary extension method provided in an embodiment of the present application.

[0017] Figure 2 It is a flowchart of a method for determining the value of a first flag bit based on global information and vehicle information provided in an embodiment of the present application.

[0018] Figure 3 It is a flowchart of a method for determining the value of a third flag bit based on global information and vehicle information provided in an embodiment of the present application.

[0019] Figure 4 It is a flowchart of a method for determining the value of the fourth flag bit based on global information and vehicle information provided in an embodiment of the present application.

[0020] Figure 5 It is a flowchart of a method provided by an embodiment of the present application for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when determining that the value of a target flag bit indicates that boundary extension is required.

[0021] Figure 6 It is a flowchart diagram of another method provided by an embodiment of the present application for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when determining that the value of a target flag bit indicates that boundary extension is required.

[0022] Figure 7 It is a flowchart of another method provided by an embodiment of the present application for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when it is determined that the value of the target flag bit indicates that boundary extension is required.

[0023] Figure 8 It is a schematic diagram of an application scenario of autonomous driving boundary extension provided in an embodiment of the present application.

[0024] Fig. 9 It is a schematic diagram of an autonomous driving boundary extension device provided in an embodiment of the present application.

[0025] Fig.10 It is a schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0027] A method and device for extending the boundary of an autonomous driving system according to an embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0028] As mentioned above, the PnC module of autonomous driving needs to use the lane line information provided by the map when planning the path. This lane line information will be organized into reference lines according to business needs. Generally, the PnC module will read the left and right boundaries of the current position from the reference line as reference information for autonomous driving. However, when encountering situations such as driving on a borrowed lane, driving on a narrow road, or driving outside the lane, the left and right boundaries provided by the current reference line need to be expanded.

[0029] In view of this, an embodiment of the present application provides a method for extending the boundary of an autonomous driving vehicle, by acquiring a reference line of an autonomous driving vehicle, traversing all reference points in the reference line to obtain global information, and acquiring the vehicle information of the autonomous driving vehicle, determining the value of each flag bit in a flag bit set based on the global information and the vehicle information, and finally traversing the value of each flag bit in the flag bit set. When it is determined that the value of a target flag bit indicates that a boundary extension is currently required, a preset strategy corresponding to the target flag bit is used to extend the boundary of the reference line of the autonomous driving vehicle, thereby providing a method for automatically extending the boundaries of autonomous driving vehicles in special scenarios, improving the decision-making accuracy of the PnC module, and enhancing the user experience.

[0030] Figure 1 is a flow chart of an autonomous driving boundary extension method provided in an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0031] In step S101, a reference line of the autonomous driving vehicle is obtained, and all reference points in the reference line are traversed to obtain global information.

[0032] The reference line may be the center line of the lane in which the autonomous driving vehicle is currently traveling.

[0033] In step S102, the vehicle information of the autonomous driving vehicle is obtained, and the value of each flag in the flag set is determined based on the global information and the vehicle information.

[0034] In step S103, the values ​​of each flag bit in the flag bit set are traversed in turn, and when it is determined that the value of the target flag bit indicates that boundary extension is required, the boundary of the reference line of the autonomous driving vehicle is extended using a preset strategy corresponding to the target flag bit.

[0035] The target flag bit is any flag bit in the flag bit set.

[0036] In some embodiments of the present application, the autonomous driving boundary extension method can be executed by an autonomous driving PnC module to provide boundary information for autonomous driving decisions. The boundary information may include left boundary information and right boundary information of the autonomous driving vehicle, the left boundary information further includes but is not limited to the coordinate value of the left boundary, whether the left boundary is expandable, etc., and the right boundary information further includes but is not limited to the coordinate value of the right boundary, whether the right boundary is expandable, etc.

[0037] In some implementations, a reference line of the autonomous driving vehicle may be first obtained. The reference line may be the center line of the lane in which the autonomous driving vehicle is currently traveling. The reference line may include multiple reference points, and each reference point may be configured with its corresponding lane information and road information. Based on the lane information and road information in the reference point, global information may be obtained.

[0038] On the other hand, the self-driving vehicle's own vehicle information can also be obtained, and the self-driving vehicle information includes the position information and boundary information of the self-driving vehicle, wherein the boundary information can be the maximum coordinate value of the self-driving vehicle in each direction. It can be understood that in the negative axis direction of the coordinate axis, the maximum coordinate value should be the coordinate value with the largest absolute value.

[0039] After obtaining the global information and the vehicle information, the value of each flag bit in the flag bit set can be determined based on the global information and the vehicle information. The flag bit set can be a pre-set set, including multiple custom flag bits, different flag bits represent different flag information, and these flag information or a combination of flag information can indicate whether boundary extension is currently required.

[0040] Furthermore, after determining the value of each flag bit in the flag bit set, the value of each flag bit in the flag bit set can be traversed in turn, and when it is determined that the value of the target flag bit indicates that boundary extension is required, the preset strategy corresponding to the target flag bit is used to expand the boundary of the reference line of the autonomous driving vehicle.

[0041] Among them, traversing the values ​​of each flag bit in sequence means that each flag bit in the flag bit set may have a priority order, and when traversing, it can be traversed in sequence according to the priority order of each flag bit. Alternatively, the preset strategies corresponding to each flag bit in the flag bit set may have an execution order relationship, and the judgment of some preset strategies may need to rely on the judgment results of other preset strategies. In this case, the flag bits corresponding to the preset strategies that are independently judged can be traversed first, and then the flag bits corresponding to the preset strategies that need to rely on the judgment results of other preset strategies can be judged.

[0042] At the same time, for flags with the same priority level, or flags whose corresponding preset strategies have no dependency relationship, their traversal order can be set according to actual needs and is not limited here.

[0043] According to the technical solution provided in the embodiment of the present application, by obtaining the reference line of the autonomous driving vehicle, traversing all reference points in the reference line to obtain global information, and obtaining the self-vehicle information of the autonomous driving vehicle, the value of each flag bit in the flag bit set is determined based on the global information and the self-vehicle information, and finally the value of each flag bit in the flag bit set is traversed. When it is determined that the value of the target flag bit indicates that boundary extension is currently required, the preset strategy corresponding to the target flag bit is used to extend the boundary of the reference line of the autonomous driving vehicle, thereby providing a method for automatic boundary extension of the autonomous driving vehicle in special scenarios, improving the decision-making accuracy of the PnC module, and enhancing the user experience.

[0044] In some embodiments of the present application, the value of each flag in the flag set is determined based on the global information and the vehicle information, and the Frenet coordinate system is constructed with the tangent direction of the reference line as the horizontal coordinate direction and the normal direction as the vertical coordinate direction; the value of each flag in the flag set is determined according to the coordinate value relationship of at least part of the parameters in the global information and the vehicle information in the Frenet coordinate system. Further, the horizontal coordinate value of each element in the Frenet coordinate system can be recorded as L, and the vertical coordinate value can be recorded as S.

[0045] In some embodiments of the present application, the global information includes at least the left boundary of the road, the left boundary of the lane, the right boundary of the road, the right boundary of the lane, and whether the left and right boundaries of the road are expandable for each reference point in the reference line. The vehicle information includes at least the left boundary of the vehicle, the right boundary of the vehicle, and the planning center of the vehicle.

[0046] In one example, the left boundary of the road of the reference point may be recorded as left_road_bound, the left boundary of the lane may be recorded as left_lane_bound, the right boundary of the road may be recorded as right_road_bound, and the right boundary of the lane may be recorded as right_lane_bound.

[0047] In another example, the left boundary of the ego vehicle information may be a point in the ego vehicle contour of the autonomous driving vehicle that has the maximum absolute value on the negative half axis of the horizontal axis of the Frenet coordinate system, for example, recorded as adc_left_1. The right boundary of the ego vehicle information may be a point in the ego vehicle contour of the autonomous driving vehicle that has the maximum absolute value on the positive half axis of the horizontal axis of the Frenet coordinate system, for example, recorded as adc_right_1. The ego vehicle planning center may be the rear axle center of the autonomous driving vehicle.

[0048] In some embodiments of the present application, the flag set may include at least the following flags: a first flag indicating whether the autonomous driving vehicle has entered a narrow road state, for example, recorded as is_narrow_extend; a second flag indicating whether the autonomous driving vehicle is in a motor vehicle lane borrowing state, for example, recorded as is_detour; a third flag indicating whether the autonomous driving vehicle is outside the boundary of the current lane, for example, recorded as is_outside_lane; a fourth flag indicating whether the distance between the autonomous driving vehicle and the lane boundary of the current lane is less than a preset distance threshold, for example, recorded as is_near_bound; a borrowing direction flag indicating the borrowing direction, for example, recorded as extend_direction; a first entry direction flag indicating entry from the left, for example, recorded as is_from_left_outside; a second entry direction flag indicating entry from the right, for example, recorded as is_from_right_outside.

[0049] Among them, the values ​​of the first flag bit, the second flag bit, the third flag bit, the fourth flag bit, the first entry direction flag bit and the second entry direction flag bit are Boolean types, and the initial default values ​​are all false. The value of the borrowing direction flag bit is an enumeration type, and the initial default value is none.

[0050] When expanding the boundaries of autonomous driving, the PnC module first needs to make a preliminary decision to determine the value of each flag bit in the flag bit set.

[0051] Figure 2 is a flow chart of a method for determining the value of the first flag bit based on global information and vehicle information provided by an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0052] In step S201, a forward detection distance, a backward detection distance and a narrow road threshold are set.

[0053] In step S202, all reference points in the Frenet coordinate system whose longitudinal distances to the planned center of the vehicle are less than the forward detection distance and the backward detection distance are determined as the target reference point set.

[0054] In step S203, the target reference point set is traversed.

[0055] In step S204, in response to determining that the target reference point set includes at least one hit reference point that satisfies the narrow road judgment condition, the value of the first flag bit is determined to be true, the value of the borrowing direction flag bit is determined to be left, and the value of the first entering direction flag bit is true.

[0056] In some embodiments of the present application, when determining the value of the first flag, the parameters forward detection distance, backward detection distance and narrow path threshold may be set first, wherein the forward detection distance may be recorded as adc_front_dis, the backward detection distance may be recorded as adc_back_dis and the narrow path threshold narrow_thresh.

[0057] Next, the target reference point set can be determined based on the forward detection distance and the backward detection distance. In one example, all reference points in the Frenet coordinate system whose longitudinal distance to the planned center of the vehicle is less than the forward detection distance and the backward detection distance can be determined as the target reference point set. In other words, the reference points outside adc_s-adc_back_dis can be ignored, and the reference points outside adc_s+adc_front_dis can be ignored, and the remaining reference points in the reference line can be used as the target reference point set.

[0058] All reference points in the determined target reference point set can be traversed, and when it is determined that at least one hit reference point meets the narrow road judgment condition, the value of the first flag bit is determined to be true, the value of the borrowing direction flag bit is determined to be left, and the value of the first entering direction flag bit is true.

[0059] Among them, the narrow road judgment condition can be that the lane width of the reference point is less than the narrow road threshold, and the right boundary of the road at the reference point coincides with the right boundary of the lane and the right road cannot be expanded, and the left boundary of the road at the reference point does not coincide with the left boundary of the lane and the left road can be expanded.

[0060] That is, for each reference point in the target reference point set, determine whether it satisfies the following conditions at the same time: the lane width of the reference point is less than narrow_thresh, the right_lane_bound of the reference point coincides with the right_road_bound, and the right lane cannot be expanded, and the left_lane_bound of the reference point does not coincide with the left_road_bound, and the left lane can be expanded. The lane width of the reference point can be obtained by subtracting right_lane_bound and left_lane_bound.

[0061] When at least one of the target reference points meets the above narrow road judgment condition, it is determined that the current narrow road state is met, so the value of is_narrow_extend is updated to true, the value of extend_direction is updated to left (LEFT), and the value of is_from_left_outside is updated to true. In this way, the narrow road state check can be implemented, and the value of the flag is_narrow_extend is updated according to the check result.

[0062] The value of the second flag is_detour can be directly determined based on the global information. If the autonomous driving vehicle is currently in the detour state, the global information will record this state. By reading this record, it can be known that the autonomous driving vehicle is currently in the detour state, and the value of is_detour can be updated to true.

[0063] Furthermore, when the value of is_detour is updated to true, the value of the detour direction flag bit extend_direction may be updated to left or right according to the vehicle's detour status, i.e., whether to detour to the left or right node.

[0064] In some scenarios of autonomous driving, for example, after the autonomous driving vehicle is manually taken over and driven outside the lane, it switches back to autonomous driving mode before returning to the current lane. At this time, the boundary is still the boundary corresponding to the lane where the autonomous driving vehicle was before the manual takeover. It needs to be expanded to the boundary that matches the current position of the vehicle. Before expansion, it is first necessary to determine whether the autonomous driving vehicle is outside the lane, and update the value of the flag is_outside_lane according to the determination result.

[0065] Figure 3 is a flow chart of a method for determining the value of the third flag bit based on global information and vehicle information provided by an embodiment of the present application. Figure 3 As shown, the method comprises the following steps:

[0066] In step S301, the current reference point is determined.

[0067] Among them, the current reference point is the reference point in the reference line that is closest to the planned center of the vehicle.

[0068] In step S302, in response to determining that the absolute value of the horizontal coordinate of the left boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the left boundary of the lane of the current reference point, the value of the third flag is determined to be true, the value of the borrowing direction flag is determined to be left, and the value of the first entering direction flag is true.

[0069] In step S303, in response to determining that the absolute value of the horizontal coordinate of the right boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the right boundary of the lane of the current reference point, the value of the third flag is determined to be true, the value of the borrowing direction flag is determined to be right, and the value of the second entering direction flag is true.

[0070] In certain embodiments of the embodiments of the present application, when determining the value of the third flag, the reference point in the reference line that is closest to the planned center of the vehicle can be first determined as the current reference point. Then compare the left boundary of the vehicle with the left boundary of the lane of the current reference point, or compare the right boundary of the vehicle with the right boundary of the lane of the current reference point. If the comparison result is that the absolute value of the horizontal coordinate of the left boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the left boundary of the lane of the current reference point, or the absolute value of the horizontal coordinate of the right boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the right boundary of the lane of the current reference point, then it can be determined that the value of the third flag is true, and the value of the borrowing direction flag is determined to be left, and the value of the first entry direction flag is true.

[0071] That is to say, if it is determined that adc_left_l is to the left of left_lane_bound of the current reference point in the horizontal direction of the Frenet coordinate system, is_outside_lane can be updated to true, extend_direction can be updated to LEFT, and is_from_left_outside can be updated to true.

[0072] Alternatively, if it is determined that adc_right_l is to the right of left_right_bound of the current reference point in the horizontal direction of the Frenet coordinate system, is_outside_lane can also be updated to true, extend_direction can be updated to right (RIGHT), and is_from_right_outside can be updated to true.

[0073] When the autonomous driving vehicle returns to the lane from outside the lane, if the boundary is immediately updated to the boundary of the lane, the sudden change of the boundary may cause the vehicle to make a sudden turn, which in turn causes the vehicle to drive unsteadily. Therefore, after the vehicle returns to the lane from outside the lane, the boundary can be extended for a period of time to allow it to return smoothly to the boundary of the lane. At this time, it is necessary to detect whether the vehicle has just returned to the lane from the outside lane. This detection can be achieved by whether the distance between the autonomous driving vehicle and the boundary of the lane is less than the preset distance threshold and whether the vehicle enters from the left or right lane. After the detection is completed, the is_near_bound flag can be updated according to the detection result.

[0074] Figure 4 is a flow chart of a method for determining the value of the fourth flag bit based on global information and vehicle information provided by an embodiment of the present application. Figure 4 As shown, the method comprises the following steps:

[0075] In step S401, the current reference point is determined.

[0076] Among them, the current reference point is the reference point in the reference line that is closest to the planned center of the vehicle.

[0077] In step S402, in response to determining that the value of the first entry direction flag is true, and the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is greater than a preset boundary detection threshold, the fourth flag is determined to be false, and the value of the borrowing direction flag is determined to be empty, and the value of the first entry direction flag is false.

[0078] In step S403, in response to determining that the value of the first entry direction flag is true, and the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is less than or equal to a preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be left, and the value of the first entry direction flag is true.

[0079] In step S404, in response to determining that the value of the second entry direction flag is true, and the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the vehicle is greater than a preset boundary detection threshold, the fourth flag is determined to be false, and the value of the borrowing direction flag is determined to be empty, and the value of the second entry direction flag is false.

[0080] In step S405, in response to determining that the value of the second entry direction flag is true, and the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the vehicle is less than or equal to the preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be right, and the value of the second entry direction flag is true.

[0081] In some embodiments of the present application, when determining the value of the fourth flag, the reference point closest to the planned center of the vehicle in the reference line can also be first determined as the current reference point. Then, it is determined whether the first entry direction flag and the second entry direction flag include a true value. If not, that is, the values ​​of the first entry direction flag and the second entry direction flag are both false, then the value of the fourth flag is kept false.

[0082] If the first entry direction flag is true, it is further determined whether the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is greater than the preset boundary detection threshold. If so, the fourth flag is determined to be false, and the value of the borrowing direction flag is determined to be empty, and the value of the first entry direction flag is false. On the contrary, if the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is less than or equal to the preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be left, and the value of the first entry direction flag is true.

[0083] If the second entry direction flag is true, it is further determined whether the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the vehicle is greater than the preset boundary detection threshold. If so, the fourth flag is determined to be false, and the value of the borrowing direction flag is determined to be empty, and the value of the second entry direction flag is false. Conversely, if the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the vehicle is less than or equal to the preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be right, and the value of the second entry direction flag is true. Among them, the preset boundary detection threshold is a custom parameter, which can be recorded as kBoundNearBuffer.

[0084] That is to say, when is_from_left_outside is true, if adc_left_l is to the right of the left_lane_bound of the current reference point shifted to the right by kBoundNearBuffer in the horizontal direction of the Frenet coordinate system, it indicates that although the autonomous driving vehicle has entered the lane from the left lane, the current distance to the left boundary of the lane is far enough and it has basically entered the stable driving stage. Therefore, the value of is_near_bound can be updated to false, the value of extend_direction can be updated to empty (NONE), and the value of is_from_left_outside can be updated to false.

[0085] On the contrary, if adc_left_l is to the left of the current reference point's left_lane_bound shifted right by kBoundNearBuffer in the horizontal direction of the Frenet coordinate system, or the two coincide, it means that the autonomous driving vehicle has just entered the lane from the left lane, and the distance to the left boundary of the lane is still small. At this time, it is necessary to maintain the boundary extension for a period of time to ensure smooth driving of the vehicle. Therefore, the value of is_near_bound can be updated to true, and the value of extend_direction can be updated to LEFT.

[0086] On the other hand, when is_from_right_outside is true, if adc_right_l is to the left of the right_lane_bound of the current reference point shifted left by kBoundNearBuffer in the horizontal direction of the Frenet coordinate system, the value of is_near_bound can be updated to false, the value of extend_direction can be updated to NONE, and the value of is_from_right_outside can be updated to false.

[0087] On the contrary, if adc_right_l is to the right of the current reference point's right_lane_bound shifted left by kBoundNearBuffer in the horizontal direction of the Frenet coordinate system, or the two coincide, the value of is_near_bound can be updated to true, and the value of extend_direction can be updated to RIGHT.

[0088] It should be noted that the above preliminary decision-making process can be executed in a certain order. For example, first determine the value of the second flag is_detour, and after the value of the flag is_detour is determined, determine whether the flag extend_direction is NONE. If so, it means that there is no need to extend the boundary at present, exit the current boundary extension execution task, and wait for the next task to start. On the contrary, if extend_direction is not NONE, the values ​​of the first flag is_narrow_extend and the third flag is_outside_lane can be determined respectively, and after the value of the third flag is updated and the flag is_from_left_outside or is_from_right_outside is assigned, determine the value of the fourth flag is_near_bound.

[0089] It is understandable that the flags in the preliminary decision-making process can be adjusted according to actual needs, including addition, deletion, and update, and the execution order of the preliminary decisions can also be adjusted according to actual needs, which is not limited here.

[0090] After the pre-decision is completed, the post-decision can be executed to determine the extended boundary according to the value of each flag. For example, the narrow road extension strategy can be used to extend the autonomous driving boundary when the is_narrow_extend flag is true.

[0091] Figure 5 1 is a flow chart of a method for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when determining that the value of the target flag bit indicates that a boundary extension is required, provided by an embodiment of the present application. Figure 5 As shown, the method comprises the following steps:

[0092] In step S501, in response to determining that the first flag is true, a first extended left boundary of each reference point in the reference line is calculated using a narrow path extension strategy.

[0093] In step S502, the first extended left boundary of each reference point is determined to be the extended left boundary, and the boundary left extension flag is set to true.

[0094] Among them, when the first extended left boundary of the reference point is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the value is the sum of the left boundary of the lane of the reference point and the first preset extension distance; when it is less than or equal to the left boundary of the lane of the reference point, the value is the left boundary of the lane of the reference point; and when it is greater than or equal to the left boundary of the road of the reference point, the value is the left boundary of the road of the reference point.

[0095] In certain embodiments of the present application, when the first flag is true, the narrow path expansion strategy is used to calculate the first extended left boundary of each reference point in the reference line, and then the first extended left boundary of each reference point is determined to be the expanded left boundary, and the boundary left extension flag is set to true.

[0096] That is, you can also define the flags extend_left and extend_right. When extend_direction is LEFT and the left_lane_bound and left_road_bound of the reference point do not overlap (indicating that the left side can be extended), the value of extend_left is updated to true. Similarly, when extend_direction is RIGHT and the right_lane_bound and right_road_bound of the reference point do not overlap (indicating that the right side can be extended), the value of extend_right is updated to true.

[0097] Furthermore, when is_narrow_extend is true, the left_lane_bound of each reference point in the reference line can be extended to the first extended left boundary, which is the sum of the left boundary of the lane of the reference point and the first preset extended distance when it is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the left boundary of the lane of the reference point when it is less than or equal to the left boundary of the lane of the reference point, and the left boundary of the road of the reference point when it is greater than or equal to the left boundary of the road of the reference point. The first preset extended distance can be recorded as left_extend_dis.

[0098] That is, the first extended left boundary of each reference point in the reference line calculated by using the narrow road extension strategy may be: the first extended left boundary of each reference point=clamp(left_lane_bound+left_extend_dis, left_lane_bound, left_road_bound), wherein clamp is a limiting function.

[0099] After completing the narrow road expansion, the value of extend_left can also be updated to true. It should be noted that the value of extend_left is updated to true because the borrowing is usually done to the left, taking into account the local traffic rules. In actual use, for example, if this solution is applied to a right-hand drive area, since the borrowing is usually done to the right, it can be adjusted according to the actual situation, for example, by changing to update the value of extend_right to true.

[0100] Furthermore, other extension strategies may be used to extend the autonomous driving boundary when the is_detour flag, the is_outside_lane flag, or the is_near_bound flag is true.

[0101] Figure 6 1 is another flowchart of a method for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when determining that the value of the target flag bit indicates that a boundary extension is required, provided by an embodiment of the present application. Figure 6 As shown, the method comprises the following steps:

[0102] In step S601, in response to determining that the boundary left extension flag is true, a second extended left boundary of each reference point in the reference line is calculated, and a third extended left boundary of each reference point in the reference line is calculated using a filtering extension strategy.

[0103] In step S602, in response to determining that the value of the second flag is true, the maximum value of the second extended left boundary and the third extended left boundary of each reference point is determined as the extended left boundary.

[0104] In step S603, in response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, the second extended left boundary of each reference point is determined to be the extended left boundary.

[0105] In some embodiments of the present application, if it is determined that the left extension flag is true, the second extended left boundary of each reference point in the reference line can be first calculated, and the third extended left boundary of each reference point in the reference line can be calculated using the filtering extension strategy. Further, if it is determined that the value of the second flag is true, the maximum value of the second extended left boundary and the third extended left boundary of each reference point is determined to be the extended left boundary. Further, if it is determined that the value of the second flag is false, and at least one of the third flag and the fourth flag is true, the second extended left boundary of each reference point is determined to be the extended left boundary.

[0106] Among them, when the second extended left boundary of the reference point is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the value is the sum of the left boundary of the lane of the reference point and the second preset extension distance; when it is less than or equal to the left boundary of the lane of the reference point, the value is the left boundary of the lane of the reference point; and when it is greater than or equal to the left boundary of the road of the reference point, the value is the left boundary of the road of the reference point.

[0107] And, the third extended left boundary of the reference point is taken as the filter left boundary when it is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, is taken as the left boundary of the lane of the reference point when it is less than or equal to the left boundary of the lane of the reference point, and is taken as the left boundary of the road of the reference point when it is greater than or equal to the left boundary of the road of the reference point. The value of the filter left boundary is the sum of the first filter item and the second filter item, the first filter item is the product of the left boundary of the lane of the reference point and the filter coefficient, and the second filter item is 1 minus the difference of the filter coefficient multiplied by the sum of the left boundary of the lane of the reference point and the preset adjacent width.

[0108] That is, when extend_left is true, the formula left_bound_l_by_adc'=clamp(left_bound_l_by_adc, left_lane_bound, left_road_bound) can be used to calculate the second extended left boundary of each reference point in the reference line, where left_bound_l_by_adc=adc_left_l+second preset extension distance, and left_bound_l_by_adc' is the second extended boundary. The value of the second preset extension distance can be customized according to actual needs. In one example, the second preset extension distance can be equal to the width of the autonomous driving vehicle.

[0109] At the same time, the third extended left boundary of each reference point in the reference line can also be calculated using a filtering extension strategy. The filtering extension strategy can be, first, using the formula left_bound_l_by_weight= left_lane_bound*extend_ratio+(1-extend_ratio)*(left_lane_bound+nearby_width) to calculate the filtered left boundary, where left_bound_l_by_weight is the filtered left boundary, and extend_ratio is the filtering coefficient. The value of the filtering coefficient can be determined according to actual needs, but needs to be greater than 0 and less than 1. Then, the formula left_bound_l_by_weight'=clamp(left_bound_l_by_weight, left_lane_bound, left_road_bound) is used to calculate the third extended left boundary, where left_bound_l_by_weight' is the third extended left boundary.

[0110] After calculating the second extended left boundary and the third extended left boundary, we can first determine whether the value of is_detour is true. If so, we use max(left_bound_l_by_adc', left_bound_l_by_weight') as the extended left boundary of each reference point, where max is the maximum value function. Otherwise, if the value of is_detour is false, we continue to determine whether the values ​​of is_outside_lane and is_near_bound are true. If one of the values ​​of is_outside_lane and is_near_bound is true, we use left_bound_l_by_adc' as the extended left boundary of each reference point.

[0111] Figure 7 1 is another flowchart of a method for extending the boundary of a reference line of an autonomous driving vehicle using a preset strategy corresponding to a target flag bit when determining that the value of the target flag bit indicates that a boundary extension is required, provided by an embodiment of the present application. Figure 7 As shown, the method comprises the following steps:

[0112] In step S701, in response to determining that the boundary right extension flag is true, a first extended right boundary of each reference point in the reference line is calculated, and a second extended right boundary of each reference point in the reference line is calculated using a filtering extension strategy.

[0113] In step S702, in response to determining that the value of the second flag is true, the maximum value of the first extended right boundary and the second extended right boundary of each reference point is determined as the extended right boundary.

[0114] In step S703, in response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, the first extended right boundary of each reference point is determined to be the extended right boundary.

[0115] In some embodiments of the present application, if it is determined that the right extension flag of the boundary is true, the first extended right boundary of each reference point in the reference line can be calculated first, and the second extended right boundary of each reference point in the reference line can be calculated using the filtering extension strategy. Further, if it is determined that the value of the second flag is true, the maximum value of the first extended right boundary and the second extended right boundary of each reference point is determined to be the extended right boundary. Further, if it is determined that the value of the second flag is false, and at least one of the third flag and the fourth flag is true, the first extended right boundary of each reference point is determined to be the extended right boundary.

[0116] Among them, when the first extended right boundary of the reference point is greater than the right boundary of the lane of the reference point and less than the right boundary of the road of the reference point, the value is the difference between the right boundary of the lane of the reference point and the second preset extension distance; when it is less than or equal to the right boundary of the lane of the reference point, the value is the right boundary of the lane of the reference point; and when it is greater than or equal to the right boundary of the road of the reference point, the value is the right boundary of the road of the reference point.

[0117] And, when the second extended right boundary of the reference point is greater than the right boundary of the lane of the reference point and less than the right boundary of the road of the reference point, the value is the filter right boundary, when it is less than or equal to the right boundary of the lane of the reference point, the value is the lane right boundary of the reference point, and when it is greater than or equal to the right boundary of the road of the reference point, the value is the road right boundary of the reference point. The value of the filter right boundary is the sum of the third filter item and the fourth filter item, the third filter item is the product of the right boundary of the lane of the reference point and the filter coefficient, and the fourth filter item is 1 minus the difference of the filter coefficient multiplied by the difference between the right boundary of the lane of the reference point and the preset adjacent width.

[0118] That is, when extend_right is true, the formula right_bound_l_by_adc'=clamp(right_bound_l_by_adc, right_lane_bound, right_road_bound) can be used to calculate the first extended right boundary of each reference point in the reference line, where right_bound_l_by_adc=adc_left_l-second preset extended distance, and right_bound_l_by_adc' is the first extended right boundary. The value of the second preset extended distance can be customized according to actual needs. In one example, the second preset extended distance can be equal to the width of the autonomous driving vehicle.

[0119] At the same time, the second extended right boundary of each reference point in the reference line can also be calculated using a filtering extension strategy. The filtering extension strategy can be, first, using the formula right_bound_l_by_weight= right_lane_bound*extend_ratio+(1-extend_ratio)*(right_lane_bound-nearby_width) to calculate the filtering right boundary, where right_bound_l_by_weight is the filtering right boundary, and extend_ratio is the filtering coefficient. The value of the filtering coefficient can be determined according to actual needs, but needs to be greater than 0 and less than 1. Then the formula right_bound_l_by_weight'=clamp(right_bound_l_by_weight, right_lane_bound, right_road_bound) is used to calculate the second extended right boundary, where right_bound_l_by_weight' is the second extended right boundary.

[0120] After calculating the first extended right boundary and the second extended right boundary, we can first determine whether the value of is_detour is true. If so, we use max(right_bound_l_by_adc', right_bound_l_by_weight') as the right boundary of each reference point after extension, where max is the maximum value function. Otherwise, if the value of is_detour is false, we continue to determine whether the values ​​of is_outside_lane and is_near_bound are true. If one of the values ​​of is_outside_lane and is_near_bound is true, we use right_bound_l_by_adc' as the right boundary of each reference point after extension.

[0121] In some embodiments of the present application, the preset adjacent width can be determined in the following manner: in response to determining that the value of the borrowing direction flag is left, obtaining the left lane width corresponding to the planned center of the vehicle; limiting the left lane width based on the preset left lane width upper limit and lower limit to obtain the left lane width after limiting; determining the left lane width after limiting as the preset adjacent width; in response to determining that the left lane width corresponding to the planned center of the vehicle cannot be obtained, determining the preset adjacent width as the preset threshold.

[0122] Alternatively, in response to determining that the value of the borrowing direction flag is right, the right lane width corresponding to the planned center of the vehicle is obtained; the right lane width is limited based on a preset upper limit value and a lower limit value of the right lane width to obtain the right lane width after limiting; the right lane width after limiting is determined to be a preset adjacent width; in response to determining that the right lane width corresponding to the planned center of the vehicle cannot be obtained, the preset adjacent width is determined to be a preset threshold.

[0123] The preset nearby width can be recorded as nearby_width, which can be created and assigned a value before executing the later decision, or can be created and assigned a value at other appropriate times, without limitation. The value of nearby_width can be written into each reference point in the reference line.

[0124] For any reference point in the reference line, when determining the value of nearby_width, you can first determine the value of the extend_direction flag. If the value of extend_direction is LEFT, read the left lane information of the current lane from the map module, and read the lane width of the left lane at the corresponding longitudinal position of adc_s. Then customize the upper and lower bounds of the left lane width, use the clamp function to limit the lane width read above using the customized upper and lower bounds of the left lane width, and update the limited result to nearby_width to obtain the determined nearby_width value.

[0125] In some embodiments, if the left lane information of the current lane cannot be read from the map module, and thus the lane width of the left lane at the longitudinal position corresponding to adc_s cannot be determined, a preset threshold can be used as the nearby_width value. The preset threshold can be customized according to actual needs and is not limited here.

[0126] On the other hand, if the value of extend_direction is RIGHT, the right lane information of the current lane is read from the map module, and the lane width of the right lane at the corresponding longitudinal position of adc_s is read. Then the upper and lower bounds of the right lane width are customized, and the clamp function is used to clamp the lane width read above using the customized upper and lower bounds of the right lane width, and the result after clamping is updated to nearby_width to obtain the determined nearby_width value.

[0127] In some embodiments, if the right lane information of the current lane cannot be read from the map module, and thus the lane width of the right lane at the longitudinal position corresponding to adc_s cannot be determined, a preset threshold can be used as the nearby_width value. The preset threshold can be customized according to actual needs and is not limited here.

[0128] Figure 8 Schematic diagram of an application scenario of autonomous driving boundary extension provided by an embodiment of the present application. Figure 8 As shown, the autonomous driving vehicle starts changing lanes at position 1 and ends the lane change at position 2 to prepare to enter the new lane. Boundary 1 is the original boundary at the end of the lane change. At this time, since part of the vehicle's parking space is still outside the current lane, the original boundary can be expanded to obtain the expanded boundary 2.

[0129] By adopting the technical solution of the embodiment of the present application, a complete set of decision-making methods for boundary expansion processing are provided for scenarios such as autonomous driving driving via other lanes, driving on narrow roads, and entering from outside the lane, which provides more reliable reference information for autonomous driving decisions and improves the stability and efficiency of autonomous driving.

[0130] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0131] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0132] Fig. 9 Schematic diagram of an automatic driving boundary expansion device provided in an embodiment of the present application. Fig. 9 As shown, the device comprises:

[0133] The acquisition module 901 is configured to acquire a reference line of the autonomous driving vehicle, traverse all reference points in the reference line, and obtain global information.

[0134] The acquisition module 901 is also configured to obtain the vehicle information of the autonomous driving vehicle, and determine the value of each flag bit in the flag bit set based on the global information and the vehicle information.

[0135] The expansion module 902 is configured to sequentially traverse the values ​​of each flag bit in the flag bit set, and when it is determined that the value of the target flag bit indicates that boundary extension is required, use a preset strategy corresponding to the target flag bit to perform boundary extension on the reference line of the autonomous driving vehicle.

[0136] The target flag bit is any flag bit in the flag bit set.

[0137] According to the technical solution provided in the embodiment of the present application, by obtaining the reference line of the autonomous driving vehicle, traversing all reference points in the reference line to obtain global information, and obtaining the self-vehicle information of the autonomous driving vehicle, the value of each flag bit in the flag bit set is determined based on the global information and the self-vehicle information, and finally the value of each flag bit in the flag bit set is traversed. When it is determined that the value of the target flag bit indicates that boundary extension is currently required, the preset strategy corresponding to the target flag bit is used to extend the boundary of the reference line of the autonomous driving vehicle, thereby providing a method for automatic boundary extension of the autonomous driving vehicle in special scenarios, improving the decision-making accuracy of the PnC module, and enhancing the user experience.

[0138] In some embodiments, the value of each flag in the flag set is determined based on the global information and the vehicle information, including: constructing a Frenet coordinate system with the tangent direction of the reference line as the horizontal coordinate direction and the normal direction as the vertical coordinate direction; determining the value of each flag in the flag set according to the coordinate value relationship of at least some parameters in the global information and the vehicle information in the Frenet coordinate system.

[0139] In some embodiments, the global information includes at least the left boundary of the road, the left boundary of the lane, the right boundary of the road, the right boundary of the lane, and whether the left and right boundaries of the road are expandable for each reference point in the reference line; the vehicle information includes at least the left boundary of the vehicle, the right boundary of the vehicle, and the planning center of the vehicle.

[0140] In some embodiments, the flag set includes at least the following flags: a first flag indicating whether the autonomous driving vehicle has entered a narrow road state; a second flag indicating whether the autonomous driving vehicle is in a motor vehicle lane borrowing state; a third flag indicating whether the autonomous driving vehicle is outside the boundary of the current lane; a fourth flag indicating whether the distance between the autonomous driving vehicle and the lane boundary of the current lane is less than a preset distance threshold; a borrowing direction flag indicating the borrowing direction; a first entry direction flag indicating entry from the left; a second entry direction flag indicating entry from the right; wherein the values ​​of the first flag, the second flag, the third flag, the fourth flag, the first entry direction flag and the second entry direction flag are Boolean types, and the value of the borrowing direction flag is an enumeration type.

[0141] In some embodiments, the value of the first flag is determined based on global information and vehicle information, including: setting a forward detection distance, a rearward detection distance, and a narrow road threshold; determining all reference points in the Frenet coordinate system whose longitudinal distances to the planned center of the vehicle are less than the forward detection distance and the rearward detection distance as a target reference point set; traversing the target reference point set; in response to determining that the target reference point set includes at least one hit reference point that satisfies a narrow road judgment condition, determining that the value of the first flag is true, determining that the value of the borrowing direction flag is left, and the value of the first entry direction flag is true.

[0142] In some embodiments, the narrow road judgment conditions include: the lane width of the reference point is less than the narrow road threshold; the right boundary of the road at the reference point coincides with the right boundary of the lane and the road on the right side cannot be expanded; and the left boundary of the road at the reference point does not coincide with the left boundary of the lane and the road on the left side can be expanded.

[0143] In some embodiments, the value of the third flag is determined based on the global information and the vehicle information, including: determining the current reference point, the current reference point being the reference point in the reference line that is closest to the planned center of the vehicle; in response to determining that the absolute value of the horizontal coordinate of the left boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the left boundary of the lane of the current reference point, determining that the value of the third flag is true, and determining that the value of the borrowing direction flag is left, and the value of the first entry direction flag is true; in response to determining that the absolute value of the horizontal coordinate of the right boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the right boundary of the lane of the current reference point, determining that the value of the third flag is true, and determining that the value of the borrowing direction flag is right, and the value of the second entry direction flag is true.

[0144] In some embodiments, determining the value of the fourth flag bit based on global information and vehicle information includes: determining a current reference point, the current reference point being a reference point in the reference line that is closest to a planned center of the vehicle; in response to determining that the value of the first entry direction flag bit is true, and the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is greater than a preset boundary detection threshold, determining that the fourth flag bit is false, and determining that the value of the borrowing direction flag bit is empty, and the value of the first entry direction flag bit is false; in response to determining that the value of the first entry direction flag bit is true, and the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is less than or equal to a preset boundary detection threshold, determining that the fourth flag bit is true, and determining that the value of the borrowing direction flag bit is left, and the value of the first entry direction flag bit is true.

[0145] In some embodiments, determining the value of the fourth flag bit based on global information and vehicle information includes: determining a current reference point, the current reference point being a reference point in the reference line that is closest to a planned center of the vehicle; in response to determining that the value of the second entry direction flag bit is true, and the difference between the horizontal coordinate of the right boundary of the lane at the current reference point and the horizontal coordinate of the right boundary of the vehicle is greater than a preset boundary detection threshold, determining that the fourth flag bit is false, and determining that the value of the borrowing direction flag bit is empty, and the value of the second entry direction flag bit is false; in response to determining that the value of the second entry direction flag bit is true, and the difference between the horizontal coordinate of the right boundary of the lane at the current reference point and the horizontal coordinate of the right boundary of the vehicle is less than or equal to a preset boundary detection threshold, determining that the fourth flag bit is true, and determining that the value of the borrowing direction flag bit is right, and the value of the second entry direction flag bit is true.

[0146] In some embodiments, when it is determined that the value of the target flag indicates that boundary extension is required, a preset strategy corresponding to the target flag is used to extend the boundary of a reference line of the autonomous driving vehicle, including: in response to determining that the first flag is true, using a narrow road expansion strategy to calculate a first extended left boundary for each reference point in the reference line; determining the first extended left boundary of each reference point as the extended left boundary, and setting the boundary left extension flag to true; wherein, when the first extended left boundary of the reference point is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the value of the first extended left boundary of the reference point is the sum of the left boundary of the lane of the reference point and the first preset extension distance, when it is less than or equal to the left boundary of the lane of the reference point, the value of the left boundary of the lane of the reference point is taken, and when it is greater than or equal to the left boundary of the road of the reference point, the value of the first extended left boundary of the reference point is taken.

[0147] In some embodiments, when it is determined that the value of the target flag indicates that boundary extension is required, a preset strategy corresponding to the target flag is used to extend the boundary of the reference line of the autonomous driving vehicle, including: in response to determining that the boundary left extension flag is true, calculating the second extended left boundary of each reference point in the reference line, and using the filtering extension strategy to calculate the third extended left boundary of each reference point in the reference line; in response to determining that the value of the second flag is true, determining the maximum value of the second extended left boundary and the third extended left boundary of each reference point as the extended left boundary; in response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, determining the second extended left boundary of each reference point as the extended left boundary.

[0148] In some embodiments, the second extended left boundary of the reference point is taken as the sum of the left boundary of the lane of the reference point and the second preset extension distance when it is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, and is taken as the left boundary of the lane of the reference point when it is less than or equal to the left boundary of the lane of the reference point, and is taken as the left boundary of the road of the reference point when it is greater than or equal to the left boundary of the road of the reference point; the third extended left boundary of the reference point is taken as the filtered left boundary when it is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, and is taken as the left boundary of the lane of the reference point when it is less than or equal to the left boundary of the lane of the reference point, and is taken as the left boundary of the road of the reference point when it is greater than or equal to the left boundary of the road of the reference point; wherein the value of the filtered left boundary is the sum of the first filter item and the second filter item, the first filter item is the product of the left boundary of the lane of the reference point and the filter coefficient, and the second filter item is 1 minus the difference of the filter coefficient multiplied by the product of the sum of the left boundary of the lane of the reference point and the preset adjacent width.

[0149] In some embodiments, when it is determined that the value of the target flag indicates that boundary extension is required, a preset strategy corresponding to the target flag is used to extend the boundary of the reference line of the autonomous driving vehicle, including: in response to determining that the boundary right extension flag is true, calculating the first extended right boundary of each reference point in the reference line, and using the filtering extension strategy to calculate the second extended right boundary of each reference point in the reference line; in response to determining that the value of the second flag is true, determining the maximum value of the first extended right boundary and the second extended right boundary of each reference point as the extended right boundary; in response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, determining the first extended right boundary of each reference point as the extended right boundary.

[0150] In some embodiments, the first extended right boundary of the reference point is taken as the difference between the right boundary of the lane of the reference point and the second preset extended distance when it is greater than the right boundary of the lane of the reference point and less than the right boundary of the road of the reference point, and is taken as the right boundary of the lane of the reference point when it is less than or equal to the right boundary of the lane of the reference point, and is taken as the right boundary of the road of the reference point when it is greater than or equal to the right boundary of the road of the reference point; the second extended right boundary of the reference point is taken as the filtered right boundary when it is greater than the right boundary of the lane of the reference point and less than the right boundary of the road of the reference point, and is taken as the right boundary of the lane of the reference point when it is less than or equal to the right boundary of the lane of the reference point, and is taken as the right boundary of the road of the reference point when it is greater than or equal to the right boundary of the road of the reference point; wherein the value of the filtered right boundary is the sum of the third filter item and the fourth filter item, the third filter item is the product of the right boundary of the lane of the reference point and the filter coefficient, and the fourth filter item is 1 minus the difference in the filter coefficient multiplied by the difference between the right boundary of the lane of the reference point and the preset adjacent width.

[0151] In some embodiments, the preset adjacent width is determined in the following manner: in response to determining that the value of the borrowing direction flag is left, the left lane width corresponding to the planned center of the self-vehicle is obtained; the left lane width is limited based on the preset upper and lower limits of the left lane width to obtain the left lane width after limiting; the left lane width after limiting is determined to be the preset adjacent width; in response to determining that the left lane width corresponding to the planned center of the self-vehicle cannot be obtained, the preset adjacent width is determined to be a preset threshold; in response to determining that the value of the borrowing direction flag is right, the right lane width corresponding to the planned center of the self-vehicle is obtained; based on the preset upper and lower limits of the right lane width, the right lane width is limited to obtain the right lane width after limiting; the right lane width after limiting is determined to be the preset adjacent width; in response to determining that the right lane width corresponding to the planned center of the self-vehicle cannot be obtained, the preset adjacent width is determined to be the preset threshold.

[0152] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0153] Fig.10 Schematic diagram of an electronic device provided in an embodiment of the present application. Fig.10 As shown, the electronic device 10 of this embodiment includes: a processor 1001, a memory 1002, and a computer program 1003 stored in the memory 1002 and executable on the processor 1001. When the processor 1001 executes the computer program 1003, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 1001 executes the computer program 1003, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0154] The electronic device 10 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 10 may include, but is not limited to, a processor 1001 and a memory 1002. Those skilled in the art will appreciate that Fig.10 The electronic device 10 is merely an example and does not limit the electronic device 10 . The electronic device 10 may include more or fewer components than shown in the figure, or different components.

[0155] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0156] The memory 1002 may be an internal storage unit of the electronic device 10, for example, a hard disk or memory of the electronic device 10. The memory 1002 may also be an external storage device of the electronic device 10, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (FlashCard), etc. equipped on the electronic device 10. The memory 1002 may also include both an internal storage unit of the electronic device 10 and an external storage device. The memory 1002 is used to store computer programs and other programs and data required by the electronic device.

[0157] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0158] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, which may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0159] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for extending the boundary of an autonomous driving system, characterized in that: include: Obtain a reference line of the autonomous driving vehicle, traverse all reference points in the reference line, and obtain global information, wherein the global information at least includes a left boundary of the road, a left boundary of the lane, a right boundary of the road, a right boundary of the lane, and whether the left and right boundaries of the road are expandable for each reference point in the reference line, and the reference line is the center line of the lane in which the autonomous driving vehicle is currently traveling; Obtaining vehicle information of the autonomous driving vehicle, and determining a value of each flag in a flag set based on the global information and the vehicle information, wherein the vehicle information at least includes a planning center of the vehicle, a left boundary of the vehicle, and a right boundary of the vehicle; Traversing the values ​​of each flag bit in the flag bit set in turn, and when it is determined that the value of the target flag bit indicates that a boundary extension is required, using a preset strategy corresponding to the target flag bit to perform boundary extension on the reference line of the autonomous driving vehicle; The target flag bit is any flag bit in the flag bit set; and the step of determining the value of each flag bit in the flag bit set based on the global information and the vehicle information includes: Constructing a Frenet coordinate system with the tangent direction of the reference line as the horizontal coordinate direction and the normal direction as the vertical coordinate direction; Determine the value of each flag bit in the flag bit set according to the coordinate value relationship of at least part of the parameters in the global information and the vehicle information in the Frenet coordinate system; The flag bit set includes at least a first flag bit indicating whether the autonomous driving vehicle has entered a narrow road state, a borrowing direction flag bit indicating a borrowing direction, and a first entering direction flag bit indicating entering from the left side; The determining the value of the first flag bit based on the global information and the vehicle information includes: Set forward detection distance, backward detection distance and narrow road threshold; Determine all reference points in the Frenet coordinate system whose longitudinal distances from the planned center of the vehicle are less than the forward detection distance and the backward detection distance as a target reference point set; Traversing the target reference point set; In response to determining that the target reference point set includes at least one hit reference point that satisfies a narrow road judgment condition, the value of the first flag is determined to be true, the value of the borrowing direction flag is determined to be left, and the value of the first entering direction flag is true.

2. The method according to claim 1, characterized in that The flag bit set also includes the following flag bits: A second flag indicating whether the autonomous driving vehicle is in a state of borrowing lanes on a motor vehicle lane; A third marker indicating whether the autonomous driving vehicle is outside the boundary of the current lane; A fourth flag indicating whether the distance between the autonomous driving vehicle and the lane boundary of the current lane is less than a preset distance threshold; The second entry direction sign indicating entry from the right; Among them, the values ​​of the first flag bit, the second flag bit, the third flag bit, the fourth flag bit, the first entry direction flag bit and the second entry direction flag bit are of Boolean type, and the value of the borrowing direction flag bit is of enumeration type.

3. The method according to claim 1, characterized in that The narrow road judgment conditions include: The lane width at the reference point is smaller than the narrow road threshold; and The right edge of the road at the reference point coincides with the right edge of the lane and the right side of the road cannot be extended; and The left boundary of the road at the reference point does not coincide with the left boundary of the lane and the left side of the road can be extended.

4. The method according to claim 2, characterized in that: Determining a value of a third flag bit based on the global information and the vehicle information includes: Determine a current reference point, where the current reference point is the reference point on the reference line that is closest to the planned center of the vehicle; In response to determining that the absolute value of the horizontal coordinate of the left boundary of the own vehicle is greater than the absolute value of the horizontal coordinate of the left boundary of the lane of the current reference point, determining that the value of the third flag is true, determining that the value of the borrowing direction flag is left, and determining that the value of the first entering direction flag is true; In response to determining that the absolute value of the horizontal coordinate of the right boundary of the vehicle is greater than the absolute value of the horizontal coordinate of the right boundary of the lane of the current reference point, the value of the third flag is determined to be true, the value of the borrowing direction flag is determined to be right, and the value of the second entry direction flag is determined to be true.

5. The method according to claim 2, characterized in that: Determining a value of a fourth flag bit based on the global information and the vehicle information includes: Determine a current reference point, where the current reference point is the reference point on the reference line that is closest to the planned center of the vehicle; In response to determining that the value of the first approaching direction flag is true, and the difference between the horizontal coordinate of the left boundary of the own vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is greater than a preset boundary detection threshold, determining that the fourth flag is false, determining that the value of the borrowing direction flag is empty, and determining that the value of the first approaching direction flag is false; In response to determining that the value of the first entry direction flag is true, and the difference between the horizontal coordinate of the left boundary of the vehicle and the horizontal coordinate of the left boundary of the lane of the current reference point is less than or equal to a preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be left, and the value of the first entry direction flag is true.

6. The method according to claim 2, characterized in that Determining a value of a fourth flag bit based on the global information and the vehicle information includes: Determine a current reference point, where the current reference point is the reference point on the reference line that is closest to the planned center of the vehicle; In response to determining that the value of the second approaching direction flag is true, and the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the vehicle is greater than a preset boundary detection threshold, determining that the fourth flag is false, determining that the value of the borrowing direction flag is empty, and determining that the value of the second approaching direction flag is false; In response to determining that the value of the second entry direction flag is true, and the difference between the horizontal coordinate of the right boundary of the lane of the current reference point and the horizontal coordinate of the right boundary of the own vehicle is less than or equal to a preset boundary detection threshold, the fourth flag is determined to be true, and the value of the borrowing direction flag is determined to be right, and the value of the second entry direction flag is true.

7. The method according to any one of claims 2 to 3, characterized in that When determining that the value of the target flag indicates that boundary extension is required, using a preset strategy corresponding to the target flag to extend the boundary of the reference line of the autonomous driving vehicle includes: In response to determining that the first flag is true, a first extended left boundary of each reference point in the reference line is calculated using a narrow path extension strategy; Determine the first extended left boundary of each reference point as the extended left boundary, and set the boundary left extension flag bit to true; Among them, when the first extended left boundary of the reference point is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the value is the sum of the left boundary of the lane of the reference point and the first preset extension distance; when it is less than or equal to the left boundary of the lane of the reference point, the value is the left boundary of the lane of the reference point; and when it is greater than or equal to the left boundary of the road of the reference point, the value is the left boundary of the road of the reference point.

8. The method according to claim 2, characterized in that: When determining that the value of the target flag indicates that boundary extension is required, using a preset strategy corresponding to the target flag to extend the boundary of the reference line of the autonomous driving vehicle includes: In response to determining that the boundary left extension flag is true, calculating a second extended left boundary of each reference point in the reference line, and calculating a third extended left boundary of each reference point in the reference line using a filtering extension strategy; In response to determining that the value of the second flag bit is true, determining the maximum value of the second extended left boundary and the third extended left boundary of each reference point as the extended left boundary; In response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, the second extended left boundary of each reference point is determined to be the extended left boundary.

9. The method according to claim 8, characterized in that When the second extended left boundary of the reference point is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point, the value is the sum of the left boundary of the lane of the reference point and the second preset extended distance; when it is less than or equal to the left boundary of the lane of the reference point, the value is the left boundary of the lane of the reference point; and when it is greater than or equal to the left boundary of the road of the reference point, the value is the left boundary of the road of the reference point; The third extended left boundary of the reference point is taken as the filter left boundary when it is greater than the left boundary of the lane of the reference point and less than the left boundary of the road of the reference point; is taken as the left boundary of the lane of the reference point when it is less than or equal to the left boundary of the lane of the reference point; and is taken as the left boundary of the road of the reference point when it is greater than or equal to the left boundary of the road of the reference point; Among them, the value of the filtered left boundary is the sum of a first filtering item and a second filtering item, the first filtering item is the product of the left boundary of the lane of the reference point and the filtering coefficient, and the second filtering item is 1 minus the difference of the filtering coefficient multiplied by the product of the sum of the left boundary of the lane of the reference point and the preset adjacent width.

10. The method according to claim 2, characterized in that When determining that the value of the target flag indicates that boundary extension is required, using a preset strategy corresponding to the target flag to extend the boundary of the reference line of the autonomous driving vehicle includes: In response to determining that the boundary right extension flag is true, calculating a first extended right boundary of each reference point in the reference line, and calculating a second extended right boundary of each reference point in the reference line using a filtering extension strategy; In response to determining that the value of the second flag bit is true, determining the maximum value of the first extended right boundary and the second extended right boundary of each reference point as the extended right boundary; In response to determining that the value of the second flag is false and at least one of the third flag and the fourth flag is true, the first extended right boundary of each reference point is determined to be the extended right boundary.

11. The method according to claim 10, characterized in that When the first extended right boundary of the reference point is greater than the right boundary of the lane of the reference point and less than the right boundary of the road of the reference point, the value is the difference between the right boundary of the lane of the reference point and the second preset extended distance; when it is less than or equal to the right boundary of the lane of the reference point, the value is the right boundary of the lane of the reference point; and when it is greater than or equal to the right boundary of the road of the reference point, the value is the right boundary of the road of the reference point; The second extended right boundary of the reference point is taken as the filtering right boundary when it is greater than the lane right boundary of the reference point and less than the road right boundary of the reference point, is taken as the lane right boundary of the reference point when it is less than or equal to the lane right boundary of the reference point, and is taken as the road right boundary of the reference point when it is greater than or equal to the road right boundary of the reference point; Among them, the value of the right boundary of the filter is the sum of the third filter item and the fourth filter item, the third filter item is the product of the right boundary of the lane of the reference point and the filter coefficient, and the fourth filter item is 1 minus the difference of the filter coefficient multiplied by the product of the difference between the right boundary of the lane of the reference point and the preset adjacent width.

12. The method according to claim 9 or 11, characterized in that: The preset proximity width is determined in the following manner: In response to determining that the value of the borrowing direction flag is left, obtaining a left lane width corresponding to the planned center of the vehicle; Based on a preset upper limit value and a lower limit value of the left lane width, the left lane width is limited to obtain a limited left lane width; Determine the left lane width after the clipping as the preset adjacent width; In response to determining that the left lane width corresponding to the planned center of the vehicle cannot be obtained, determining the preset adjacent width to be a preset threshold; In response to determining that the value of the borrowing direction flag is right, obtaining a right lane width corresponding to the planned center of the vehicle; Based on a preset upper limit value and a lower limit value of the right lane width, the right lane width is limited to obtain a limited right lane width; Determine the right lane width after the clipping as the preset adjacent width; In response to determining that the right lane width corresponding to the planned center of the vehicle cannot be obtained, the preset adjacent width is determined to be a preset threshold.

13. An automatic driving boundary expansion device, characterized in that: include: an acquisition module, configured to acquire a reference line of the autonomous driving vehicle, traverse all reference points in the reference line, and obtain global information, wherein the global information at least includes a left boundary of the road, a left boundary of the lane, a right boundary of the road, a right boundary of the lane, and whether the left and right boundaries of the road are expandable for each reference point in the reference line, and the reference line is a center line of a lane in which the autonomous driving vehicle is currently traveling; The acquisition module is further configured to acquire vehicle information of the autonomous driving vehicle, and determine the value of each flag bit in the flag bit set based on the global information and the vehicle information, wherein the vehicle information at least includes a planning center of the vehicle, a left boundary of the vehicle, and a right boundary of the vehicle; an expansion module, configured to sequentially traverse the values ​​of each flag bit in the flag bit set, and when it is determined that the value of a target flag bit indicates that a boundary expansion is required, use a preset strategy corresponding to the target flag bit to perform boundary expansion on the reference line of the autonomous driving vehicle; Wherein, the target flag bit is any flag bit in the flag bit set; The determining the value of each flag bit in the flag bit set based on the global information and the vehicle information includes: Constructing a Frenet coordinate system with the tangent direction of the reference line as the horizontal coordinate direction and the normal direction as the vertical coordinate direction; Determine the value of each flag bit in the flag bit set according to the coordinate value relationship of at least part of the parameters in the global information and the vehicle information in the Frenet coordinate system; The flag bit set includes at least a first flag bit indicating whether the autonomous driving vehicle has entered a narrow road state, a borrowing direction flag bit indicating a borrowing direction, and a first entering direction flag bit indicating entering from the left side; The determining the value of the first flag bit based on the global information and the vehicle information includes: Set forward detection distance, backward detection distance and narrow road threshold; Determine all reference points in the Frenet coordinate system whose longitudinal distances from the planned center of the vehicle are less than the forward detection distance and the backward detection distance as a target reference point set; Traversing the target reference point set; In response to determining that the target reference point set includes at least one hit reference point that satisfies a narrow road judgment condition, the value of the first flag is determined to be true, the value of the borrowing direction flag is determined to be left, and the value of the first entering direction flag is true.

Images