WO2024114620A1 - 车辆变道成功率的预测方法、装置、计算机设备和存储介质 - Google Patents

车辆变道成功率的预测方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2024114620A1
WO2024114620A1 PCT/CN2023/134641 CN2023134641W WO2024114620A1 WO 2024114620 A1 WO2024114620 A1 WO 2024114620A1 CN 2023134641 W CN2023134641 W CN 2023134641W WO 2024114620 A1 WO2024114620 A1 WO 2024114620A1
Authority
WO
WIPO (PCT)
Prior art keywords
lane
vehicle
lanes
original
line
Prior art date
Application number
PCT/CN2023/134641
Other languages
English (en)
French (fr)
Inventor
肖宁
Original Assignee
腾讯科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Publication of WO2024114620A1 publication Critical patent/WO2024114620A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Definitions

  • the present application relates to the field of traffic technology, and in particular to a method, device, computer equipment, storage medium and computer program product for predicting a vehicle lane change success rate.
  • a method, apparatus, computer device, computer-readable storage medium, and computer program product for predicting a vehicle lane change success rate are provided.
  • a success rate of the vehicle changing lanes from the first lane to the second lane is predicted based on the current position offset, the target position offset, the remaining feature set, and the driving information of the vehicle.
  • the present application further provides a device for predicting a lane change success rate of a vehicle, wherein the vehicle is on a driving road.
  • the driving road comprises a plurality of lane groups, each lane group comprises a lane section of a plurality of lanes, and the device comprises:
  • An extraction module is used to obtain lane group data of the multiple lane groups; determine feature sets corresponding to the multiple lane lines of the driving road from the lane group data of the multiple lane groups, wherein the feature sets respectively include multiple feature subsets corresponding to the multiple lane groups, and each feature subset includes a corresponding lane line type and line segment offset;
  • a processing module configured to remove the feature subsets satisfying the first preset condition from the corresponding feature set among the multiple feature subsets, to obtain a remaining feature set
  • a determination module configured to determine, during the driving of the vehicle, a current position offset of a driving position of the vehicle in the plurality of lane groups, the driving position being in a first lane among the plurality of lanes; and determine a target position offset of a preset target position of the vehicle in the plurality of lane groups, the preset target position being in a second lane among the plurality of lanes different from the first lane;
  • a determination module is used to predict a success rate of the vehicle changing lanes from the first lane to the second lane based on the current position offset, the target position offset, the remaining feature set and the driving information of the vehicle.
  • the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method for predicting the vehicle lane change success rate are implemented.
  • the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method for predicting the vehicle lane change success rate when the computer program is executed by a processor.
  • the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the method for predicting the vehicle lane change success rate are implemented.
  • FIG1 is a diagram showing an application environment of a method for predicting a vehicle lane change success rate in one embodiment
  • FIG2 is a schematic flow chart of a method for predicting a vehicle lane change success rate in one embodiment
  • FIG3a is a schematic diagram of high-precision data or lane-level data in one embodiment
  • FIG3 b is a schematic diagram of high-precision data or lane-level data in another embodiment
  • FIG4 is a schematic diagram of segmenting a lane group and marking an offset in a lane group in one embodiment
  • FIG5 is a schematic diagram of marking the offset of the driving position and the target position in a lane group in one embodiment
  • FIG6 is a schematic diagram of decomposing the vehicle speed according to an embodiment
  • FIG7 is a schematic diagram of prompting a vehicle to change lanes and re-plan a route according to an embodiment
  • FIG8 is a schematic diagram of prompting a vehicle to change lanes in one embodiment
  • FIG9 is a schematic diagram of a process of performing lane reorganization on original lane group data in one embodiment
  • FIG10 is a schematic diagram of performing lane reorganization on an original lane group in one embodiment
  • FIG11 is a schematic diagram of performing lane reorganization on an original lane group in another embodiment
  • FIG12 is a schematic diagram of performing lane reorganization on an original lane group in another embodiment
  • FIG13 is a schematic diagram of performing lane reorganization on an original lane group in another embodiment
  • FIG14 is a schematic diagram of a process for predicting a vehicle lane change success rate in one embodiment
  • FIG15 is a schematic diagram of the positional relationship between the driving position and the lane where the target position is located in one embodiment
  • FIG16 is a schematic diagram of prompting a vehicle to change lanes in another embodiment
  • FIG17 is a structural block diagram of a device for predicting a vehicle lane change success rate in one embodiment
  • FIG18 is a structural block diagram of a device for predicting a vehicle lane change success rate in one embodiment
  • FIG. 19 is a diagram showing the internal structure of a computer device in one embodiment.
  • first and second involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that “first and second” can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.
  • the method for predicting the success rate of lane change of a vehicle provided in the embodiment of the present application can be applied in the application environment shown in FIG1 .
  • the terminal 102 communicates with the server 104 via a network.
  • the data storage system can store data that the server 104 needs to process.
  • the data storage system can be integrated on the server 104, or placed on a cloud or other network server.
  • the terminal 102 may be an electronic device installed with an electronic map, such as a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, an IoT device, and a portable wearable device; an IoT device may be a smart speaker, a smart vehicle-mounted device, and a vehicle, etc.
  • a portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc.
  • Server 104 can be an independent physical server or a service node in a blockchain system.
  • a peer-to-peer (P2P) network is formed between the service nodes in the blockchain system.
  • the P2P protocol is an application layer protocol running on top of the Transmission Control Protocol (TCP).
  • server 104 can also be a server cluster composed of multiple physical servers, and can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.
  • cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.
  • the terminal 102 and the server 104 can be connected via Bluetooth, USB (Universal Serial Bus) or network and other communication connection methods, which are not limited in this application.
  • USB Universal Serial Bus
  • a method for predicting a vehicle lane change success rate is provided.
  • the vehicle is on a driving road, and the driving road includes multiple lane groups, each lane group includes lane sections of multiple lanes.
  • the method is applied to the terminal 102 in FIG. 1 as an example for description, and includes the following steps:
  • the vehicle may be a vehicle driven by a user on a driving road, for example, various types of motor vehicles.
  • a driving road may refer to a road on which a vehicle travels, and may include urban roads and other types of roads, such as expressways; the driving road includes multiple lanes and lane lines between lanes; in addition, the driving road also includes separation lines between lanes and the driving road environment, such as edge lines on both sides of the driving road.
  • the driving road environment may refer to the road environment on both sides of the driving road, such as green belts on both sides of the driving road. It should be noted that the lane lines between lanes and the separation lines between lanes and the driving road environment can be regarded as lane lines of the driving road.
  • a vehicle icon can be displayed in the form of a vehicle icon. The vehicle is represented by .
  • Lane group data may be data for lane groups (Lane Group) in a driving road, and may include the number of lanes, lane width, lane line types on the left and right of each lane and corresponding lane line segments, lane line markings (such as the name or number of the lane line), lane-level connectivity between the upstream and downstream of each lane, and lane center line points, etc.
  • multiple lane groups may correspond to multiple lane group data, that is, each lane group may correspond to one lane group data.
  • the markings of the multiple lane line segments contained therein may be the same.
  • the lane line segments in the same lane group data may include lane line segments of one or more lane lines, for example, lane group data a may include lane line segments of lane line 1, lane line segments of lane line 2, lane line segments of lane line 3, and lane line segments of lane line 4; therefore, the same lane group data may include one or more lane line types corresponding to the lane line segments.
  • a driving road may include multiple lane groups, which may form a lane group set; a lane group is a set of lanes corresponding to lane group data in a driving road, each lane group may include lane segments of multiple lanes, and may also include lane line segments between lane segments, and lane segments and lane line segments between adjacent lane groups are aligned, such as the first lane segment in lane group 1 (i.e., lane segment numbered 1) and the first lane segment in lane group 2 belong to the same lane, as shown in FIG3a. It should be noted that the order of lane segments in the same lane group may be from left to right.
  • a lane group may be a set of lanes in the driving road consisting of lanes in a section of road of corresponding length; correspondingly, in a map application, a lane group may be a set of lanes consisting of lanes in a section of road image displayed on a map page; for example, if FIG3a is a section of road image displayed in a map application, then a lane group may be a set consisting of lanes 1, lane 2, and lane 3 in the section of road 0 to 50.
  • the lane segment may be a segment with a certain length in the lane line, such as a lane segment with a length of 10 meters or 50 meters in the lane line; in addition, the lane segment may have a corresponding segment offset, and the segment offset may be the offset of the lane segment in the lane group set.
  • the segment offset may be the offset between the lane segment and a target point (such as a starting point or a driving position of a vehicle) in the lane group set, and the direction is from the target point to the starting point of the lane segment.
  • the positive and negative signs may be used to indicate the direction, and the distance between the target point and the starting point of the lane segment may be used to indicate the offset.
  • the segment offset may include a starting point offset and an end point offset
  • the starting point offset may refer to the offset of the starting point of the lane segment in the lane group set
  • the end point offset may refer to the offset of the end point (i.e., the end point) of the lane segment in the lane group set
  • the difference between the end point offset and the starting point offset is the length of the lane segment.
  • the above lane line type may refer to the type of lane line, including solid line type, dashed line type and virtual-real type; wherein, solid line type refers to the lane line being a solid line, so the solid line type may also be called a solid line; dashed line type refers to the lane line being a dashed line, so the dashed line type may also be called a dashed line; virtual-real type may refer to the presence of dashed lines and solid lines between two adjacent lanes at the same time, and the lane line formed by the dashed lines and solid lines existing at the same time is the virtual-real type.
  • the lane-level connectivity relationship may be determined based on the lane topology information.
  • the lane group set may be a set consisting of multiple lane groups, such as the set consisting of lane group 1, lane group 2 and lane group 3 in FIG3b.
  • the lane group data can be the data obtained by reorganizing the original lane group data of the road; wherein the original lane group data is the data corresponding to each original lane group in the road; in addition, if the section of the road near the current position of the vehicle does not have lane merging, bifurcation or other changes in the number of lanes, then the lane group data can be the original lane group data.
  • the original lane group data can be extracted from high-precision (HD) data or lane-level data; the high-precision data can be high-precision road data, also known as high-precision map data; lane-level data can be lane-level road data, which is between standard (SD) road data and high-precision road data, that is, richer than standard road data, but not as rich as high-precision road data.
  • high-precision data includes road lane line equations, coordinates of centerline points, Lane type, lane marking type, lane width, lane speed limit, lane topology information, pole coordinates, signpost location, camera and traffic light location, etc.
  • Lane-level data includes road lane equations, centerline coordinates, lane type, lane marking type, lane width, lane speed limit, and lane topology information.
  • centerline you can refer to the small black dot in Figure 3b; in addition, Pa and Pb in Figure 3b are the vehicle's driving position and target position, respectively.
  • Lane groups 1 to 3 are lane sets composed of lane segments in high-precision data or lane-level data.
  • the lane line feature may be a feature of a lane line in a driving road, including the lane line type of each lane line segment in a lane group and the line segment offset of each lane line segment in the lane group.
  • the line segment offset may be the offset of a lane line segment, including the starting point offset and the end point offset of the lane line segment. As shown in FIG3b, in lane group 1 of the lane group set, the lane line segment on the left side of lane 1 The starting offset is 0 and the ending offset is 50.
  • Each feature set includes multiple feature subsets corresponding to multiple lane groups.
  • a feature set may include feature subsets corresponding to lane segments belonging to the same lane line in multiple lane groups.
  • Each feature subset includes a corresponding lane line type and segment offset.
  • a feature subset may be composed of a segment offset and a lane line type corresponding to a lane line segment in a certain lane line. Since the segment offset may include the starting offset and the end offset of the lane line segment, the feature subset may also be called a ternary feature subset or a ternary group. It should be noted that each lane line has a corresponding feature set.
  • the terminal may first obtain the vehicle's driving position and preset target position, then determine the data value range based on the driving position and the preset target position, and then obtain the original lane group data or lane group data within the data value range from the map database, such as taking the driving position as a reference point, extending the target length (such as 100 or 200 meters) in the opposite direction of the driving direction to obtain the first position point; then, taking the preset target position as a reference point, extending the target length in the driving direction to obtain the second position point, and finally obtaining the original lane group data or lane group data between the first position point and the second position point.
  • the terminal may also obtain the original lane group data or lane group data at the driving position and the preset target position.
  • the terminal can reorganize the original lane group data to obtain corresponding lane group data. If the number of lanes between two adjacent original lane groups is inconsistent (i.e., unequal), at least one virtual lane is added to the original lane group with fewer lanes, and the virtual lane width and virtual lane line type are configured for the virtual lane, such as the virtual lane width is zero or half of the original lane width.
  • the terminal can also directly obtain the lane group data of multiple lane groups; then, read the lane line type corresponding to the lane line segment of each lane line in the driving road from the lane group data, and determine the segment offset of each lane line segment in the lane group; combine the segment offset and lane line type belonging to the same lane line to obtain the feature set corresponding to each lane line.
  • the lane line types of each lane line corresponding to the multiple lane line segments in the multiple lane groups are read, and the line segment offsets of the multiple lane line segments in the lane groups they are in are determined; for each lane line segment, the corresponding lane line type and line segment offset are combined to obtain the feature subsets corresponding to each lane line segment; for each lane line, the feature subsets corresponding to the lane line segments contained in it are combined to obtain the feature sets corresponding to each lane line. It should be pointed out that when the lane line types in two adjacent feature subsets are the same, the two feature subsets can be combined into one feature subset.
  • the lane group data includes the lane line type, so the lane line type of the corresponding lane line segment can be read from the lane group data, and then the lane line type and the line segment offset of the same lane line segment are combined to obtain the feature subset corresponding to the lane line segment.
  • lane line 3 has four lane line segments, and the feature subsets corresponding to each lane line segment are (0,50, dashed line), (50,70, dashed line), (70,100, solid line) and (100,180, dashed line).
  • two adjacent feature subsets with the same lane line type can be combined into one feature subset to obtain the synthesized (0,70, dashed line) and (70,180, solid line) of lane line 3, and then these two feature subsets are combined to obtain the feature set of lane line 3.
  • the feature set of lane 4 is
  • the terminal determines a plurality of lane line identifiers from the lane group data of the plurality of lane groups, and the plurality of lane line identifiers are respectively used to uniquely identify one of the plurality of lane lines; and reads the lane line type of each lane line segment corresponding to the same lane line identifier from the lane group data of the plurality of lane groups.
  • the lane line marking may be the name or number of the lane line.
  • the terminal obtains the number of lane segments in the lane group data of multiple lane groups, and then obtains the number of each lane line in the driving road according to the number of the lane segment. For example, among 10 lane group data, each lane group data has a lane segment numbered 1. At this time, the number of the corresponding lane line can be obtained according to the number of these 10 lane segments, that is, it can be determined that the lane line belongs to lane line 1 (that is, lane line No. 1); then, for each lane line, the terminal can read the lane line type of each lane segment of lane line 1 from the lane group data of multiple lane groups.
  • the terminal determines lane lines on the driving road based on the lane group data; divides the lane lines on the driving road into segments to obtain lane line segments in each lane group; and reads the lane line type corresponding to each lane line segment from the lane group data.
  • a lane line segment may refer to a lane line of a certain length.
  • segment the following method may be used: segment the lane line according to the change of lane line type. As shown in FIG4 , at a position of 50 meters (m) from the starting point of lane line 4, the lane line type changes from a dotted line to a solid line. At this time, segment division may be performed at the 50m position; at a position of 70m from the starting point of lane line 3, the lane line type changes from a dotted line to a solid line.
  • segment division may be performed at the 70m position; in addition, at a position of 100m from the starting point of lane line 4, the lane line type changes again from a solid line to a dotted line. At this time, segment division may be performed at the 100m position, thereby dividing these lane lines into 4 segments.
  • the first preset condition may be that the lane line type is a solid line type that cannot be changed. Therefore, the feature subset that meets the first preset condition may refer to: the lane line type and the line segment offset in the feature subset are respectively a solid line type and a line segment offset of a solid line type lane line segment. As shown in FIG4 , the lane line type and line segment offset that meet the first preset condition may be the lane line feature of the lane line segment between 50m and 100m in lane line 4, that is, the solid line type and line segment offset of the lane line segment between 50m and 100m.
  • Removal may refer to filtering, screening or deleting, for example, deleting a feature subset that meets the first preset condition from each feature set, or deleting the lane line type and line segment offset in the feature subset that meets the first preset condition from each feature set.
  • the terminal may remove the feature subset that meets the first preset condition from the feature set to obtain the remaining feature set corresponding to the feature set and other feature sets that have not been removed. It should be noted that the remaining feature set corresponding to the feature set and other feature sets that have not been removed can be collectively referred to as the remaining feature set. Specifically, the terminal may remove the feature subset that meets the first preset condition from the feature set of all lane lines to obtain the feature subset that meets the first preset condition. The remaining feature sets correspond one-to-one.
  • the terminal may combine the remaining feature sets obtained according to the lane line numbers in the driving road to obtain a feature set sequence.
  • the order of the lane line numbers may be from left to right, such as the leftmost lane line of the driving road is numbered 1, the second lane line on the left side of the driving road is numbered 2, and so on.
  • solid lane lines are retained in the relevant drawings in consideration of traffic habits.
  • the feature set sequence eliminates the features corresponding to the solid lane lines, so the lane group corresponding to the feature set sequence does not contain solid lane lines.
  • the terminal when performing a removal operation on the lane line feature, can first search for a target feature subset that meets the first preset condition in each feature set, and then remove the target feature subset in each feature set, or remove the line segment offset and lane line type in the target feature subset to obtain a remaining feature set, such as a remaining feature set that does not contain a solid line lane line type and a corresponding line segment offset. Finally, the terminal combines the remaining feature sets obtained after removal to obtain a feature set sequence.
  • the remaining feature sets of each lane line are sorted and combined according to the order of the lane lines to obtain the feature set sequence
  • the terminal can filter the feature set corresponding to the lane line between the driving position and the preset target position in the feature set sequence, and the filtered feature sets can form a new feature set sequence, that is, a feature set subsequence.
  • the driving position is in the first lane among multiple lanes, and the driving position can be the current positioning position of the vehicle, and the driving position can also be called the driving position; in addition, in some application scenarios, it can also be any position in front of the driving direction where the vehicle lane change success rate prediction is required.
  • the preset target position is in the second lane among multiple lanes, and the preset target position can be the position that the vehicle needs to go through to reach the terminal position or the terminal position to be reached. If it is the position that needs to be experienced to reach the terminal position, the preset target position can be determined on the obtained planned path after path planning.
  • the lane group is a set of lanes in the driving road corresponding to the lane group data.
  • the current position offset of the driving position may be an offset in multiple lane groups, for example, multiple lane groups of the driving position
  • the offset between the starting point of the lane segment in the lane group (i.e., the lane segment where the vehicle is located) and the driving position; and the target position offset can be the offset between the preset target position and the starting point of the lane group where the driving position is located, for example, the offset between the starting point of the lane segment in the lane group where the driving position is located and the preset target position.
  • Da is the current position offset of the driving position Pa in the lane group
  • Db is the target position offset of the preset target position Pb in the lane group.
  • the current position offset and the target position offset are vectors, and the positive and negative signs can be used to indicate the offset direction, and the numerical value can be used to indicate the offset distance.
  • the vehicle's driving information may include at least one of the vehicle's speed, acceleration, driving time, and braking information.
  • the vehicle's speed may be the vehicle's current driving speed; when the vehicle's driving position is any position in front of the driving direction where a vehicle lane change success rate prediction is required, the vehicle's speed may be a speed predicted based on the vehicle's current driving speed.
  • the vehicle's speed may be decomposed into a parallel speed and a vertical speed relative to the lane, as shown in FIG6 .
  • the terminal when predicting the success rate of a vehicle changing lanes from a first lane to a second lane, may combine two different vertical speeds for prediction, the two different vertical speeds including a first vertical speed and a second vertical speed, the first vertical speed may refer to the maximum vertical speed under the conditions of complying with traffic regulations and traffic safety, and the second vertical speed may be the minimum vertical speed without affecting the driving of the following vehicle; therefore, the first time (i.e., the shortest time) required for lane change is calculated based on the first vertical speed, and the second time (i.e., the longest time) required for lane change is calculated based on the second vertical speed, and the first time and the second time are used as two different time thresholds (i.e., the first time threshold and the second time threshold).
  • the first time i.e., the shortest time
  • the second time i.e., the longest time
  • the terminal can determine the driving distance of the vehicle based on the driving information of the vehicle; then, according to the driving distance and the current position offset of the driving position, determine the forward position offset of the vehicle, and predict the success rate of the vehicle changing lanes from the first lane to the second lane based on the forward position offset, the line segment offset of each lane line segment in the remaining feature set, and the target position offset.
  • the terminal can also predict the success rate of the vehicle changing lanes from the first lane to the second lane based on the forward position offset, the line segment offset of each lane line segment in the feature set sequence, and the target position offset.
  • the feature set sequence includes a feature set subsequence between the first lane and the second lane. Therefore, when predicting the vehicle lane change success rate, the success rate of the vehicle changing lanes from the first lane to the second lane is accurately and quickly calculated based on the forward position offset, the segment offset of each lane segment in the feature set subsequence and the target position offset.
  • the terminal determines the vehicle speed based on the driving information, and determines the vehicle speed based on the vertical speed and lane width corresponding to the vehicle speed.
  • the time required to cross the lane, and the vehicle's travel distance determined based on the time and the parallel speed corresponding to the vehicle speed predict the vehicle's success rate of changing lanes from the first lane to the second lane based on the forward position offset, the segment offset of each lane segment in the feature set subset and the target position offset.
  • the vehicle's lane change success rate can also be predicted in combination with the vehicle condition data of the driving road. For example, based on the vehicle condition data, the forward position offset, the segment offset of each lane segment in the feature set subset and the target position offset, the success rate of the vehicle changing lanes from the first lane to the second lane can be predicted, thereby effectively improving traffic safety during driving.
  • the prediction results can be applied to the following two scenarios, as follows:
  • Scenario 1 Path planning for the vehicle.
  • the terminal when it is determined based on the success rate that the vehicle cannot cross from the first lane to the second lane, the terminal can re-plan the path to the preset target location and then issue a new path prompt, thereby prompting the user to drive along the re-planned path, which is conducive to improving traffic safety.
  • the preset target position Pb is the destination of the vehicle
  • the vehicle when the vehicle is at the driving position Pa , according to the technical solution of the present application, it is determined that the vehicle can gradually cross from lane 2 to lane 4. Since the vehicle did not change lanes in time at the driving position Pa , when the vehicle travels to P′a , according to the technical solution of the present application, it is determined that the vehicle can no longer gradually cross from lane 2 to lane 4. Therefore, the path to reach the preset target position Pb is replanned and a new path prompt is issued to avoid traffic safety hazards caused by forced lane changes.
  • Scenario 2 prompting the vehicle to change lanes.
  • the terminal when it is determined based on the success rate that the vehicle can gradually cross from the first lane to the second lane, the terminal can display lane change instructions on the electronic map, prompting the user to change lanes according to the lane change instructions so that the vehicle can gradually cross from the first lane to the second lane.
  • the terminal displays lane-level lane change instructions on the navigation page of the electronic map, prompting the user to change lanes according to the lane-level lane change instructions, which is conducive to improving traffic safety.
  • lane line features are extracted from the lane group data of the driving road where the vehicle is located to obtain feature sets corresponding to each lane line; feature subsets that meet the first preset condition in each feature set are removed to obtain remaining feature sets that do not include feature subsets that meet the first preset condition, so that according to whether the remaining feature set is an empty set and in combination with the remaining feature set and the vehicle's driving information, current position offset and target position offset, the success rate of the vehicle changing lanes between different lanes while driving can be quickly and accurately predicted; in addition, since the remaining feature set that does not include the feature subset that meets the first preset condition is obtained, even if the first lane where the driving position is located is different from the second lane where the preset target position is located, If the lanes do not belong to the same lane, the feature set corresponding to the lane line between the driving position and the preset target position in the remaining feature set can be simply used to be empty, so as to quickly determine that the vehicle cannot change from the first lane where the driving position
  • the lane group data is obtained by performing lane reorganization on the original lane group data of the driving road.
  • the original lane group data is data corresponding to each original lane group in the driving road, so the terminal can perform lane reorganization on the lane group data.
  • performing lane reorganization on the original lane group data may refer to performing lane reorganization on the map corresponding to the original lane group data.
  • the original lane groups in the application are reorganized (non-physical reorganization). Considering the different lane topology information between the original lane groups, the lane group data can be reorganized in the following two ways, as follows:
  • Method 1 Expand the original lane group, as shown in Figure 9.
  • the specific expansion method is as follows:
  • the first original lane group and the second original lane group are lane groups adjacent to each other on the driving road.
  • the original lane group data may include first original lane group data and second original lane group data, wherein the first original lane group data is about the first original lane group data that has not been reorganized, and the second original lane group data is about the second original lane group data that has not been reorganized.
  • the number of lanes in the original lane group Gi is 3
  • the number of lanes in the original lane group Gi +1 is 4, and the edge lane on the right side of the original lane group Gi +1 is not on the same lane as the edge lane on the right side of the original lane group Gi .
  • FIGS. 10 to 12 illustrate the case where the number of lanes of Gi is less than the number of lanes of Gi +1 . In addition, there may also be a case where the number of lanes of Gi is greater than the number of lanes of Gi+1 .
  • the added virtual lane may refer to a lane that does not exist in the actual driving road.
  • a virtual lane is added to the corresponding original lane group, so that a lane group used for lane change success prediction can be obtained.
  • Lane alignment may refer to the alignment of a lane in the first original lane group with a corresponding lane in the second original lane group in the extension direction of the driving road (i.e., the driving direction), thereby ensuring that the number of lanes between adjacent lane groups is consistent (i.e., equal).
  • each lane in the lane group Gi has a corresponding lane in the lane group Gi +1 , and the number of lanes in the lane group is also the same, that is, the lane group Gi has 4 lane sections, and its adjacent lane group Gi +1 also has 4 lane sections.
  • the original lane group data is also updated accordingly, that is, a virtual lane and the identifier of the virtual lane are added to the original lane group data; in addition, the generation time and storage location of the virtual lane can also be marked in the original lane group data. Therefore, the original lane group data after adding the above data can be used as the required lane group data.
  • Scenario a adding virtual lanes at the edge of the original lane group.
  • the terminal when the number of first lanes is less than the number of second lanes, and the edge lane segment in the second original lane group is not in the same lane as the edge lane segment in the first original lane group, the terminal adds a virtual lane in the first original lane group that is aligned with the edge lane segment of the second original lane group.
  • the edge lane segment alignment may mean that the added virtual lane is in the same lane as the corresponding edge lane segment in the second original lane group, that is, the lane topology information between the two lanes is connected.
  • Scenario b adding a virtual lane in the middle of the original lane group.
  • the terminal converts the middle lane segment in the first original lane group into at least two virtual lanes that are respectively aligned with the at least two lane segments in the second original lane group.
  • the lane segment 2 in the original lane group Gi is converted into two virtual lanes that are respectively aligned with the lane segments 2 to 3 in the original lane group Gi +1 , thereby keeping the number of lanes consistent with the original lane group Gi +1 , thereby obtaining the reorganized lane group on the right side of FIG12 .
  • S906 Configure lane features for the virtual lane in the original lane group data to obtain the lane group data.
  • the lane feature includes the lane width and the corresponding lane line type.
  • the lane feature can be configured for the virtual lane in the above updated original lane group data.
  • the terminal can also configure the first virtual lane width and the first virtual lane line for the virtual lane in the original lane group data.
  • the first virtual lane line is a solid lane line that indicates that lane change is prohibited, and the first virtual lane width is equal to the target width value.
  • the lane width of the virtual lane can be set to 0; in addition, considering that the added virtual lane does not have a corresponding lane in the actual road, it is impossible to drive on the virtual lane and it is impossible to change lanes to the virtual lane, so the lane line corresponding to the virtual lane is set to a solid line, that is, the lane line type is a solid line type.
  • the terminal may also configure a dotted second virtual lane line between at least two virtual lanes; configure a second virtual lane width for at least two virtual lanes in the original lane group data; wherein the sum of the second virtual lane widths of at least two virtual lanes is consistent with the lane width of the middle lane section in the first original lane group.
  • the lane widths of the two virtual lanes in Figure 12 can be set to half of the original lanes respectively; in addition, considering that the two virtual lanes are actually one lane, a dotted lane line is set between the two virtual lanes.
  • the lanes of the misaligned lane sections are reorganized so that the lanes between the reorganized lane groups are aligned, which is conducive to predicting the success rate of vehicle lane changes between different lanes.
  • automatic driving or assisted driving
  • It can effectively improve the computing speed and efficiency; it can also effectively reduce the code complexity and improve the code reusability during the development phase.
  • Method 2 straighten the curved original lane group.
  • the lane group data is obtained by reorganizing the original lane group data of the driving road, and the original lane group data is the data corresponding to each original lane group in the driving road; therefore, the terminal can obtain the original lane group data of the driving road; when the original lane group in the original lane group data is a lane group with curvature, the original lane group in the original lane group data is converted into a straight lane group.
  • the curvature may refer to the degree of curvature of the original lane group. For example, as shown in FIG13 , when the original lane group is a curved lane group, the originally curved original lane group is converted into a straight lane group.
  • the time-consuming calculation required for lane change is facilitated, that is, there is no need to consider the curvature during the calculation process, which can facilitate the prediction of the vehicle lane change success rate between different lanes, and can also effectively reduce the code complexity and improve the reusability of the code.
  • the vehicle speed includes a parallel speed and a vertical speed relative to the lane;
  • the feature set sequence includes a feature set subsequence between the first lane and the second lane, so for predicting the vehicle lane change success rate between the first lane and the second lane, as shown in FIG14 , a specific method may include:
  • S1402 Determine the time required to cross each lane in the lane group based on the vertical speed and lane width.
  • crossing each lane in the lane group may refer to the time taken for a vehicle to change lanes from one lane to another lane in the lane group.
  • the lane width of each lane is usually 3.5 to 3.75 m. In the subsequent embodiments, the lane width may be 3.5 m. Since the lane group obtained after the reorganization contains virtual lanes, the lane width may be 0, 3.5/L, and 3.5.
  • n is the number of virtual lanes added in the middle of the original lane group; when the lane width is 0, it means that a virtual lane is added at the edge of the original lane group; when the lane width is 3.5/L, it means that L virtual lanes are added in the middle of the original lane group; when the lane width is 3.5/L, it means that L virtual lanes are added in the middle of the original lane group; when the lane width is 3.5, it means that no virtual lanes are added in the original lane group.
  • the terminal can use the vertical speed V ⁇ of the vehicle and the lane width to calculate the time t required to cross each lane in the lane group. It should be noted that when calculating the time t, it is considered that the vehicle is changing from one lane to another. When the vehicle changes from lane 2 to lane 3, the time required is Among them, w2 is the lane width of lane 2, and w3 is the lane width of lane 3.
  • the vehicle will move forward at the same time.
  • the distance the vehicle moves forward can be calculated based on the time t and the parallel vehicle speed V ⁇ .
  • S1406 Determine the forward position offset of the vehicle according to the travel distance and the current position offset.
  • the terminal calculates the sum of the travel distance and the current position offset, and then uses the sum as the forward position offset of the vehicle.
  • S1408 predicting the success rate of the vehicle changing lanes from the first lane to the second lane based on the preceding position offset, the line segment offset of each lane line segment in the feature set subset, and the target position offset.
  • the current position offset of the vehicle's driving position in the lane group is determined, and the target position offset of the preset target position in the lane group is determined.
  • the lane group is a set of lanes in the driving road corresponding to the lane group data. Therefore, for the prediction of the vehicle's lane change success rate, the current position offset, the target position offset, the feature set subset, the vehicle's speed and lane width can be used for comprehensive calculation.
  • the success rate of the vehicle crossing from the first lane to the second lane can be accurately calculated, thereby improving the accuracy of the lane change success rate when the vehicle changes lanes, and covering a wider range of application scenarios, effectively improving traffic safety.
  • first lane where the vehicle is located may be on the left side of the second lane, as shown in Figure 15 (a); in addition, the first lane where the vehicle is located may also be on the right side of the second lane, as shown in Figure 15 (b); the first lane where the vehicle is located may also be in the same lane as the second lane (that is, the vehicle changes lanes to other lanes and then changes back to the first lane), as shown in Figure 15 (c). Therefore, when predicting the success rate of vehicle lane change, the following scenarios can be divided for explanation, as follows:
  • Scenario 1 The first lane where the vehicle is located is on the left side of the second lane.
  • the above-mentioned A and B are the lanes where P a and P b are located respectively.
  • the vehicle in the first lane is on the right side of the second lane.
  • the terminal traverses the feature subsets of each lane segment along the driving direction of the vehicle in the i-th lane line corresponding to the feature set subset sequence to obtain a traversal result; and determines whether there is a target feature that meets the second preset condition in the traversal result.
  • the success rate of the vehicle changing lanes between the first lane and the second lane can also be quickly and accurately calculated.
  • scenario 3 the first lane and the second lane where the vehicle is located belong to the same lane.
  • the terminal can directly determine that the vehicle can cross from the first lane to the second lane based on the forward position offset, the segment offset of each lane segment in the feature set sequence (or feature set subsequence) and the target position offset, thereby quickly calculating the success rate of the vehicle's lane change.
  • the vehicle's starting position Pa (lon, lat) and preset target position Pb (lon, lat) are obtained, where lon and lat represent longitude and latitude respectively.
  • Pa here can be the vehicle's current positioning position or any starting position that needs to be determined.
  • the vehicle speed V is also obtained, and the vehicle speed V may be the actual driving speed of the vehicle, or an arbitrarily preset speed, or a speed predicted based on the actual driving speed of the vehicle.
  • the lane group data includes: multiple original lane groups, the number of lanes recorded in each lane group, the lane line types on the left and right of each lane and their corresponding intervals, the lane-level connectivity relationship between the upstream and downstream of each lane, the lane-level center line point and lane width, etc.
  • each adjacent lane group Gi and Gi +1 is reorganized until all lane groups G are reorganized to obtain a complete lane group G'.
  • a lane is expanded on the left side of Gi +1 .
  • a virtual lane can be added on the left side of Gi .
  • the left lane line type of the virtual lane is a solid edge line, and the right side is a solid line.
  • the virtual lane cannot be reached. It is just a placeholder lane.
  • Its lane width can be assigned a value of 0 (just to generate a lane logically to ensure that the road width remains unchanged), and the lane group G' on the right side of Figure 10 is obtained.
  • n virtual lanes are added on the left side.
  • a lane is expanded on the right side of Gi +1 .
  • a virtual lane can be added on the right side of Gi .
  • the right lane line type of the virtual lane is a solid edge line and the left side is a solid line. According to the traffic regulations, this virtual lane cannot be reached and is just a placeholder lane. Its lane width can be assigned a value of 0, and the lane group G' on the right side of Figure 11 is obtained.
  • n virtual lanes are added on the left side.
  • a lane is expanded in the middle of Gi +1 .
  • the lane to be expanded in the middle of Gi can be replaced with two virtual lanes (for example, lane 2 on the left side of Figure 12 is replaced with virtual lanes 2-3 on the right side).
  • the left and right lane lines of the virtual lanes are both dotted lines, indicating that the two lanes can be connected.
  • the lane widths of the two virtual lanes are half of the original lane width, and the lane group G' on the right side of Figure 12 is obtained.
  • the expanded lane is replaced with m virtual lanes, the left and right lane lines of each virtual lane are both dotted lines, and the lane width of each virtual lane is set to 1/m of the original lane width.
  • the lane group G' contains the lane feature set and width information w of each lane i, where each lane feature set can contain a triplet (start, end, type) corresponding to each lane segment, where start and end respectively represent the offset of the starting point and end point of each lane segment compared to the starting point of the lane group G' (i.e., the starting point offset and the end point offset), and type records the line type.
  • start, end, type corresponding to each lane segment
  • start and end respectively represent the offset of the starting point and end point of each lane segment compared to the starting point of the lane group G' (i.e., the starting point offset and the end point offset)
  • Figure 15 shows three possible relative relationships between Pa and Pb .
  • the lane number where Pa is located be A
  • the lane number where Pb is located be B
  • there are three situations: A>B, A ⁇ B, and A B.
  • a ⁇ B means that Pb is on the right side of Pa , corresponding to Figure 15 (a);
  • A>B means that Pb is on the left side of Pa , corresponding to Figure 15 (b);
  • 1 to L are the lane line crossing order for the vehicle to travel from lane A to lane B. If there is a Line i set that is empty, it means that there is a lane line that cannot be crossed, and it directly returns to A ⁇ B as impassable (that is, the vehicle cannot change lanes from lane A to lane B).
  • the vehicle speed V can be decomposed into a component V ⁇ parallel to the lane line (referred to as parallel speed) and a component V ⁇ perpendicular to the lane line (referred to as vertical speed), as shown in Figure 6.
  • parallel speed a component V ⁇ parallel to the lane line
  • vertical speed a component V ⁇ perpendicular to the lane line
  • a maximum vertical speed can be set. For example, if the lane width is W, and it is assumed that it takes at least 2 seconds to change lanes, then you can set At this time there are:
  • the solution of the present application can be used to quickly determine whether the current position or a certain position ahead can change lanes to the lane leading to the XX highway direction, so as to prompt the user to change lanes in advance, or when the last opportunity to change lanes is missed, a new route can be planned for the user in time.
  • the solution of this application comprehensively considers multiple factors such as vehicle speed, lane width, number of lanes crossed, traffic rules and traffic safety to determine whether different lanes can be crossed.
  • it is more general and universal, and can be applied to the fields of intelligent vehicle control technology, autonomous driving and advanced driver assistance, and is helpful for specific applications such as lane-level planning and lane-level deviation.
  • road resources can be allocated more efficiently, which can be applied to smart city transportation projects to alleviate traffic congestion and reasonably allocate lane resource usage, which can effectively alleviate traffic congestion, reduce traffic accident rates, improve traffic safety, reduce energy consumption and environmental pollution, etc.
  • the embodiment of the present application also provides a prediction device for vehicle lane change success rate for implementing the above-mentioned prediction method for vehicle lane change success rate.
  • the implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the prediction device for vehicle lane change success rate provided below can refer to the limitations of the above-mentioned method for predicting vehicle lane change success rate, and will not be repeated here.
  • a device for predicting a vehicle lane change success rate is provided.
  • the vehicle is on a driving road, and the driving road includes a plurality of lane groups, each of which includes a plurality of lane sections.
  • the device includes: an extraction module 1702, a processing module 1704, a determination module 1706, and a judgment module 1708, wherein:
  • Extraction module 1702 used to obtain lane group data of multiple lane groups; determine feature sets corresponding to multiple lane lines of the driving road from the lane group data of the multiple lane groups, the feature sets respectively include multiple feature subsets corresponding to the multiple lane groups, and each feature subset includes a corresponding lane line type and line segment offset;
  • a processing module 1704 is used to remove the feature subsets satisfying the first preset condition from the corresponding feature set among the multiple feature subsets to obtain a remaining feature set;
  • the determination module 1706 is used to determine, during the driving process of the vehicle, a current position offset of a driving position of the vehicle in a plurality of lane groups, wherein the driving position is in a first lane among the plurality of lanes; and determine a target position offset of a preset target position of the vehicle in a plurality of lane groups, wherein the preset target position is in a second lane among the plurality of lanes different from the first lane.
  • the determination module 1708 is used to predict the success rate of the vehicle changing lanes from the first lane to the second lane based on the current position offset, the target position offset, the remaining feature set and the driving information of the vehicle.
  • lane line features are extracted from the lane group data of the driving road where the vehicle is located to obtain feature sets corresponding to each lane line; feature subsets that meet the first preset condition in each feature set are removed to obtain remaining feature sets that do not include feature subsets that meet the first preset condition, so that according to whether the remaining feature set is an empty set and in combination with the remaining feature set and the vehicle's driving information, current position offset and target position offset, the success rate of the vehicle changing lanes between different lanes when driving can be quickly and accurately predicted; in addition, since the remaining feature set that does not include the feature subset that meets the first preset condition is obtained, even if the first lane where the driving position is located is different from the second lane where the preset target position is located, If the lanes do not belong to the same lane, the feature set corresponding to the lane line between the driving position and the preset target position in the remaining feature set can be simply used to be empty, so as to quickly determine that the vehicle cannot change from the first lane where the driving position
  • the extraction module 1702 is further used to read the lane line types of each lane line corresponding to the multiple lane line segments in the multiple lane groups from the lane group data of the multiple lane groups, and determine the segment offsets of the multiple lane line segments in the lane groups in which they are located; for each lane line segment, the corresponding lane line type and segment offset are combined to obtain the feature subsets corresponding to each lane line segment; for each lane line, the feature subsets corresponding to the lane line segments it contains are combined to obtain the feature sets corresponding to each lane line.
  • the extraction module 1702 is further used to determine multiple lane line identifiers from lane group data of multiple lane groups, and the multiple lane line identifiers are respectively used to uniquely identify one of the multiple lane lines; and read the lane line type of each lane line segment corresponding to the same lane line identifier from the lane group data of the multiple lane groups.
  • the lane group data is obtained by performing lane reorganization on original lane group data of the driving road, and the original lane group data is data corresponding to each original lane group in the driving road;
  • the device further includes: a first reorganization module 1710, which is used to determine the number of first lanes of a first original lane group in the original lane group data, and to determine the number of second lanes of a second original lane group in the original lane group data, wherein the first original lane group and the second original lane group are adjacent lane groups in the driving road; when the number of first lanes is less than the number of second lanes, a virtual lane is added to the first original lane group to update the original lane group data, wherein the virtual lane enables the first original lane group to achieve lane alignment with the second original lane group; optionally, the lane reorganization further includes: configuring lane features for the virtual lanes in the original lane group data to obtain lane group data.
  • the first reorganization module 1710 is further used to add a virtual lane aligned with the edge lane segment of the second original lane group in the first original lane group when the number of the first lanes is less than the number of the second lanes and the edge lane segment in the second original lane group is not in the same lane as the edge lane segment in the first original lane group.
  • the first reorganization module 1710 is also used to configure a first virtual lane width and a first virtual lane line for the virtual lane in the original lane group data; wherein the first virtual lane line is a solid line type lane line representing prohibited lane changes, and the first virtual lane width is equal to the target width value.
  • lanes of unaligned road sections are reorganized so that the lanes between the reorganized lane groups are aligned, which can help predict the success rate of vehicle lane changes between different lanes.
  • logical judgments on the number of lanes in different lane groups are avoided. Therefore, the calculation speed and efficiency can be effectively improved during automatic driving (or assisted driving); and the code complexity is effectively reduced and the code reusability is improved during the development stage.
  • the first reorganization module 1710 is further used to convert the middle lane segment in the first original lane group into at least two virtual lanes that are respectively aligned with the at least two lane segments in the second original lane group when the number of the first lanes is less than the number of the second lanes and the middle lane segment in the first original lane group is split into at least two lane segments in the second original lane group.
  • the first reorganization module 1710 is further used to configure a dotted second virtual lane line between at least two virtual lanes; and configure a second virtual lane width for at least two virtual lanes in the original lane group data; wherein the sum of the second virtual lane widths of at least two virtual lanes is consistent with the lane width of the middle lane section.
  • the time-consuming calculation required for lane change is facilitated, that is, there is no need to consider the curvature during the calculation process, which can facilitate the prediction of the success rate of vehicle lane change between different lanes, and can also effectively reduce the code complexity and improve the reusability of the code.
  • the lane group data is obtained by performing lane reorganization on original lane group data of the driving road, and the original lane group data is data corresponding to each original lane group in the driving road;
  • the device further includes: a second reorganization module 1712, which is used to obtain original lane group data of the driving road; when the original lane group in the original lane group data is a lane group with curvature, the original lane group in the original lane group data is converted into a straight lane group.
  • a second reorganization module 1712 which is used to obtain original lane group data of the driving road; when the original lane group in the original lane group data is a lane group with curvature, the original lane group in the original lane group data is converted into a straight lane group.
  • the feature set includes a feature subset of each lane line segment in the lane line, and the feature subset includes a line segment offset and a lane line type of the lane line segment;
  • the processing module 1704 is further configured to search for a target feature subset that satisfies the first preset condition in each feature set;
  • the target feature subset is removed from the collection, or the line segment offset and lane line type within the target feature subset are removed.
  • the determination module 1708 is also used to determine the driving distance of the vehicle based on the driving information of the vehicle; determine the forward position offset of the vehicle based on the driving distance and the current position offset; and predict the success rate of the vehicle changing lanes from the first lane to the second lane based on the forward position offset, the segment offset of each lane segment in the remaining feature set, and the target position offset.
  • the driving information of the vehicle includes a parallel speed and a vertical speed relative to the lane;
  • the remaining feature set includes a feature set subsequence between the first lane and the second lane;
  • the determination module 1708 is further used to determine the time taken to cross each lane in the lane group based on the vertical speed and the lane width; determine the travel distance of the vehicle based on the time taken and the parallel speed; determine the forward position offset of the vehicle based on the travel distance and the position offset of the driving position; and predict the success rate of the vehicle changing lanes from the first lane to the second lane based on the forward position offset, the line segment offset of each lane line segment in the feature set subset sequence and the target position offset.
  • the line segment offset includes an end offset, and the number of lane lines corresponding to the feature set subsequence is n, where n is a positive integer not less than 1;
  • the determination module 1708 is further configured to determine that the vehicle cannot change lanes from the first lane to the second lane when it is determined that there is no target feature subset satisfying the second preset condition in the traversal results.
  • the current position offset of the vehicle's driving position in the lane group is determined, and the target position offset of the preset target position in the lane group is determined.
  • the lane group is a set of lanes in the driving road corresponding to the lane group data. Therefore, for the prediction of the vehicle's lane change success rate, the current position offset, the target position offset, the feature set subset, the vehicle's speed and lane width can be used for comprehensive calculation.
  • the success rate of the vehicle crossing from the first lane to the second lane can be accurately calculated, thereby improving the accuracy of the lane change success rate when the vehicle changes lanes, and covering a wider range of application scenarios, effectively improving traffic safety.
  • Each module in the above-mentioned vehicle lane change success rate prediction device can be implemented in whole or in part by software, hardware and a combination thereof.
  • Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.
  • a computer device which may be a terminal, and its internal structure diagram may be shown in FIG19.
  • the computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input device.
  • the processor, the memory, and the input/output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input/output interface.
  • the processor of the computer device is used to provide computing and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system and a computer program.
  • the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium.
  • the input/output interface of the computer device is used to exchange information between the processor and an external device.
  • the communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies.
  • the display unit of the computer device is used to form a visually visible picture, which may be A display screen, a projection device or a virtual reality imaging device, the display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse, etc.
  • FIG. 19 is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied.
  • the specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.
  • a computer device including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the above-mentioned method for predicting the success rate of vehicle lane change when executing the computer program.
  • a computer-readable storage medium on which a computer program is stored.
  • the steps of the above-mentioned method for predicting the success rate of vehicle lane change are implemented.
  • a computer program product including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for predicting the success rate of vehicle lane change.
  • user information including but not limited to user device information, user personal information, etc.
  • data including but not limited to data used for analysis, stored data, displayed data, etc.
  • any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory.
  • Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc.
  • Volatile memory can include random access memory (RAM) or external cache memory, etc.
  • RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • the database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database.
  • Non-relational databases may include distributed databases based on blockchain, etc., but are not limited to this.
  • the processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, etc., but are not limited to this.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Traffic Control Systems (AREA)

Abstract

本申请涉及一种车辆变道成功率的预测方法、装置、计算机设备、存储介质和计算机程序产品。所述方法应用于交通、地图和自动驾驶等应用领域，包括：获取多个车道组的车道组数据；从多个车道组的车道组数据确定行车道路的多条车道线分别对应的特征集(202)；将多个特征子集中满足第一预设条件的特征子集从其所对应的特征集去除掉，得到剩余特征集(204)；在所述车辆行驶的过程中，确定车辆的行车位置在多个车道组中的当前位置偏移量，确定车辆的预设目标位置在多个车道组中的目标位置偏移量(206)；基于当前位置偏移量、目标位置偏移量、剩余特征集和车辆的行驶信息，预测车辆从第一车道变道至第二车道的成功率(208)。

Description

车辆变道成功率的预测方法、装置、计算机设备和存储介质
相关申请
本申请要求2022年11月30日申请的，申请号为2022115145046，名称为“车道的通行性判定方法、装置、计算机设备和存储介质”的中国专利申请的优先权，在此将其全文引入作为参考。
技术领域
本申请涉及交通技术领域，特别是涉及一种车辆变道成功率的预测方法、装置、计算机设备、存储介质和计算机程序产品。
背景技术
随着电子地图技术的发展，出行对象在驾驶车辆出行时，可以通过电子地图导航的方式进行驾驶，或者将电子地图与辅助驾驶或无人驾驶技术结合的方式进行驾驶。
在采用上述驾驶方式驾驶车辆的过程中，通常需要对行车道路上的车道进行变道成功率的预测，以便可以安全且快速地行驶到目的地。然而，在实际的行驶过程中进行变道成功率的预测时，通常受多种因素影响，从而会导致预测结果准确性不高，从而给交通带来安全隐患。
发明内容
根据本申请的各种实施例，提供了一种车辆变道成功率的预测方法、装置、计算机设备、计算机可读存储介质和计算机程序产品。
第一方面，本申请提供了一种车辆变道成功率的预测方法，所述车辆处于行车道路中，所述行车道路包含多个车道组，每个车道组包括多条车道的车道路段，所述方法由计算机设备执行，所述方法包括：
获取所述多个车道组的车道组数据；
从所述多个车道组的车道组数据确定所述行车道路的多条车道线分别对应的特征集，所述特征集分别包括对应于所述多个车道组的多个特征子集，每个所述特征子集包括对应的车道线类型和线段偏移量；
将所述多个特征子集中满足第一预设条件的特征子集从其所对应的所述特征集去除掉，得到剩余特征集；
在所述车辆行驶的过程中，确定所述车辆的行车位置在所述多个车道组中的当前位置偏移量，所述行车位置处于所述多条车道中的第一车道；
确定所述车辆的预设目标位置在所述多个车道组中的目标位置偏移量，所述预设目标位置处于所述多条车道中不同于所述第一车道的第二车道；
基于所述当前位置偏移量、所述目标位置偏移量、所述剩余特征集和所述车辆的行驶信息，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
第二方面，本申请还提供了一种车辆变道成功率的预测装置，所述车辆处于行车道路中， 所述行车道路包含多个车道组，每个车道组包括多条车道的车道路段，所述装置包括：
提取模块，用于获取所述多个车道组的车道组数据；从所述多个车道组的车道组数据确定所述行车道路的多条车道线分别对应的特征集，所述特征集分别包括对应于所述多个车道组的多个特征子集，每个所述特征子集包括对应的车道线类型和线段偏移量；
处理模块，用于将所述多个特征子集中满足第一预设条件的特征子集从其所对应的所述特征集去除掉，得到剩余特征集；
确定模块，用于在所述车辆行驶的过程中，确定所述车辆的行车位置在所述多个车道组中的当前位置偏移量，所述行车位置处于所述多条车道中的第一车道；确定所述车辆的预设目标位置在所述多个车道组中的目标位置偏移量，所述预设目标位置处于所述多条车道中不同于所述第一车道的第二车道；
判定模块，用于基于所述当前位置偏移量、所述目标位置偏移量、所述剩余特征集和所述车辆的行驶信息，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
第三方面，本申请还提供了一种计算机设备。所述计算机设备包括存储器和处理器，所述存储器存储有计算机程序，所述处理器执行所述计算机程序时实现所述车辆变道成功率的预测方法的步骤。
第四方面，本申请还提供了一种计算机可读存储介质。所述计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现所述车辆变道成功率的预测方法的步骤。
第五方面，本申请还提供了一种计算机程序产品。所述计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现所述车辆变道成功率的预测方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
图1为一个实施例中车辆变道成功率的预测方法的应用环境图；
图2为一个实施例中车辆变道成功率的预测方法的流程示意图；
图3a为一个实施例中高精数据或车道级数据的示意图；
图3b为另一个实施例中高精数据或车道级数据的示意图；
图4为一个实施例中在车道组中对车道组分段并标记偏移量的示意图；
图5为一个实施例中在车道组中标记行车位置和目标位置的偏移量的示意图；
图6为一个实施例对车辆的车速进行分解的示意图；
图7为一个实施例提示车辆变道以及重新规划路径的示意图；
图8为一个实施例中提示车辆变道的示意图；
图9为一个实施例中对原始车道组数据进行车道重组的流程示意图；
图10为一个实施例中对原始车道组进行车道重组的示意图；
图11为另一个实施例中对原始车道组进行车道重组的示意图；
图12为另一个实施例中对原始车道组进行车道重组的示意图；
图13为另一个实施例中对原始车道组进行车道重组的示意图；
图14为一个实施例中对车辆变道成功率的预测的流程示意图；
图15为一个实施例中对行车位置与目标位置所处车道的位置关系的示意图；
图16为另一个实施例中提示车辆变道的示意图；
图17为一个实施例中车辆变道成功率的预测装置的结构框图；
图18为一个实施例中车辆变道成功率的预测装置的结构框图；
图19为一个实施例中计算机设备的内部结构图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
需要说明的是，在以下的描述中，所涉及的术语“第一和第二”仅仅是区别类似的对象，不代表针对对象的特定排序，可以理解地，“第一和第二”在允许的情况下可以互换特定的顺序或先后次序，以使这里描述的本申请实施例能够以除了在这里图示或描述的以外的顺序实施。
本申请实施例提供的车辆变道成功率的预测方法，可以应用于如图1所示的应用环境中。其中，终端102通过网络与服务器104进行通信。数据存储系统可以存储服务器104需要处理的数据。数据存储系统可以集成在服务器104上，也可以放在云上或其他网络服务器上。
其中，终端102可以是安装了电子地图的电子设备，如智能手机、平板电脑、笔记本电脑、台式计算机、智能音箱、智能手表、物联网设备和便携式可穿戴设备；物联网设备可为智能音箱、智能车载设备和车辆等。便携式可穿戴设备可为智能手表、智能手环、头戴设备等。
服务器104可以是独立的物理服务器，也可以是区块链系统中的服务节点，该区块链系统中的各服务节点之间形成点对点(P2P，Peer To Peer)网络，P2P协议是一个运行在传输控制协议(TCP，Transmission Control Protocol)协议之上的应用层协议。
此外，服务器104还可以是多个物理服务器构成的服务器集群，可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(Content Delivery Network，CDN)、以及大数据和人工智能平台等基础云计算服务的云服务器。
终端102与服务器104之间可以通过蓝牙、USB(Universal Serial Bus，通用串行总线)或者网络等通讯连接方式进行连接，本申请在此不做限制。
在一个实施例中，如图2所示，提供了一种车辆变道成功率的预测方法，该车辆处于行车道路中，行车道路包含多个车道组，每个车道组包括多条车道的车道路段，以该方法应用于图1中的终端102为例进行说明，包括以下步骤：
S202，获取多个车道组的车道组数据；从多个车道组的车道组数据确定行车道路的多条车道线分别对应的特征集。
其中，该车辆可以是用户在行车道路上驾驶的车辆，例如可以是各种类型的机动车。行车道路可以指车辆所行使的道路，可以包括城市道路和其它类型的道路，如高速路；该行车道路中包含了多条车道和各车道之间的车道线；此外，该行车道路中还包含车道与行车道路环境之间的间隔线，如位于行车道路两边的边缘线。该行车道路环境可以指行车道路两边的道路环境，如行车道路两边的绿化带。需要指出的是，各车道之间的车道线以及车道与行车道路环境之间的间隔线都可以视为行车道路的车道线。在地图应用中，可以以车辆图标的形 式表示该车辆。
车道组数据可以是针对行车道路中的车道组(Lane Group)的数据，可以包括车道数量、车道宽度、各车道左右的车道线类型以及对应的车道线段、车道线标识(如车道线的名称或编号)、各车道上下游的车道级连通关系和车道中心线形点等。其中，多个车道组可以对应多个车道组数据，即每个车道组均可以对应一个车道组数据。其中，对于同一个车道线标识对应的车道线，其所包含的多个车道线段的标识可以是相同的。此外，同一个车道组数据中的车道线段可以包含一个或多个车道线的车道线段，例如车道组数据a可以包含车道线1的车道线段、车道线2的车道线段、车道线3的车道线段和车道线4的车道线段；因此，同一个车道组数据中可以包含一个或多个与车道线段对应的车道线类型。
行车道路可以包括多个车道组，这些车道组可以组成一个车道组集合；车道组是行车道路中的与车道组数据对应的车道集合，每个车道组可以包括多条车道的车道路段，此外还可以包括各车道路段之间的车道线段，此外每相邻车道组之间的车道路段和车道线段是对齐的，如车道组1中的第一个车道路段(即编号为1的车道路段)与车道组2的第一个车道路段属于同一个车道，如图3a所示。需要指出的是，同一个车道组中的车道路段的顺序可以是从左至右。需要指出的是，在实际的行车道路中，车道组可以是该行车道路中的由相应长度的一段道路中的各个车道所组成的车道集合；对应地，在地图应用中，车道组可以是显示在地图页面上的一段道路图像中各车道所组成的车道集合；例如，若图3a为地图应用中显示的一段道路图像，那么车道组可以是0～50这一段道路中的车道1、车道2和车道3所组成的集合。
上述车道线段可以是车道线中的具有一定长度的线段，如车道线中长度为10米或50米的车道线段；此外，该车道线段可以具有相应的线段偏移量，该线段偏移量可以是车道线段在车道组集合中的偏移量，具体地，该线段偏移量可以是车道线段与车道组集合中的目标点(如起点或车辆的行车位置)之间的偏移量，方向是从该目标点指向该车道线段的起点，可以采用正负符号表示方向，采用目标点与该车道线段的起点之间的距离表示偏移量大小。在实际应用中，该线段偏移量可以包括起点偏移量和末端偏移量，而该起点偏移量可以指车道线段的起点在车道组集合内的偏移量，末端偏移量可以指车道线段的末端点(即终点)在车道组集合内的偏移量，而末端偏移量与起点偏移量之间的差值即为车道线段的长度。上述的车道线类型可以指车道线的类型，包括实线型、虚线型和虚实型；其中，实线型指的是车道线为实线，因此实线型也可称为实线；虚线型指的是车道线为虚线，因此虚线型也可称为虚线；虚实型可以指相邻两个车道之间同时存在虚线和实线，该同时存在的虚线和实线所构成的车道线即为虚实型。该车道级连通关系可根据车道拓扑信息来确定。车道组集合可以是多个车道组所组成的集合，如图3b中的车道组1、车道组2和车道组3所组成的集合。
对于上述车道组数据，需要指出的是：若车辆当前位置附近的这一段行车道路存在车道合并、分叉(即一个车道分流出两个车道)或其它车道数量发生变化的情况，则该车道组数据可以是对行车道路的原始车道组数据进行车道重组所得的数据；其中，原始车道组数据是行车道路中各原始车道组对应的数据；此外，若车辆当前位置附近的这一段行车道路不存在车道合并、分叉或其它车道数量发生变化的情况，则该车道组数据可以是原始车道组数据。该原始车道组数据可以从高精(High Definition，HD)数据或车道级数据中提取出来；该高精数据可以是高精道路数据，也可称为高精度地图数据；车道级数据可以是车道级道路数据，介于标准(Standard Definition，SD)道路数据与高精度道路数据之间，即比标准道路数据丰富，但没有高精度道路数据丰富。其中，高精数据包括道路车道线方程、中心线形点的坐标、 车道类型、车道标线类型、车道宽度、车道限速、车道拓扑信息、电线杆坐标、指路牌位置、摄像头和红绿灯位置等。车道级数据包括道路车道线方程、中心线形点的坐标、车道类型、车道标线类型、车道宽度、车道限速和车道拓扑信息等。对于中心线形点，可以参考图3b中的小黑点；此外，图3b中的Pa和Pb分别为车辆的行车位置和目标位置，车道组1～3是高精数据或车道级数据中各车道路段组成的车道集合。
车道线特征可以是行车道路中车道线的特征，包括车道组内各车道线段的车道线类型和各车道线段在该车道组内的线段偏移量。该线段偏移量可以车道线段的偏移量，包括车道线段的起点偏移量和末端偏移量，如图3b所示，在车道组集合的车道组1中，车道1左侧的车道线段的起始偏移量为0，末端偏移量为50。
每个特征集分别包括对应于多个车道组的多个特征子集，如一个特征集可以包括多个车道组中的属于相同车道线的各车道线段分别对应的特征子集。每个特征子集包括对应的车道线类型和线段偏移量，如一个特征子集可以由某个车道线中的一个车道线段对应的线段偏移量和车道线类型组成，由于线段偏移量可以包括车道线段的起点偏移量和末端偏移量，因此该特征子集也可称为三元特征子集或三元组。需要指出的是，每个车道线具有对应的特征集。
在一个实施例中，终端可以先获取车辆的行车位置和预设目标位置，然后基于该行车位置和预设目标位置确定数据取值范围，然后从地图数据库中获取该数据取值范围内的原始车道组数据或车道组数据，如以行车位置为参考点，沿行驶方向的反方向延伸目标长度(如100或200米)，得到第一位置点；然后，以预设目标位置为参考点，沿行驶方向延伸目标长度，得到第二位置点，最后获取位于第一位置点和第二位置点之间的原始车道组数据或车道组数据。此外，终端也可以获取位于行车位置和预设目标位置的原始车道组数据或车道组数据。
在获取到原始车道组数据之后，终端可以对该原始车道组数据进行车道重组，得到对应的车道组数据，如两个相邻的原始车道组之间车道数量不一致(即不相等)，则在车道数量少的原始车道组中添加至少一条虚拟车道，并为该虚拟车道配置虚拟车道宽度和虚拟车道线类型，如虚拟车道宽度为零或为原始车道宽度的一半。
除了获取原始车道组数据，终端也可以直接获取多个车道组的车道组数据；然后，从该车道组数据中，读取行车道路中各车道线的车道线段对应的车道线类型，并确定各车道线段在车道组中的线段偏移量；将属于相同车道线的线段偏移量和车道线类型进行组合，得到各车道线分别对应的特征集。
具体地，从多个车道组的车道组数据中，读取各车道线的分别对应于多个车道组中的多个车道线段的车道线类型，并确定多条车道线段在各自处于的车道组中的线段偏移量；对每个车道线段，将其对应的车道线类型和线段偏移量进行组合，得到各车道线段分别对应的特征子集；对每条车道线，将其包含的车道线段对应的特征子集进行组合，得到各车道线分别对应的特征集。其中，需要指出的是，当相邻两个特征子集内的车道线类型相同时，可以将这两个特征子集合成一个特征子集。
例如，车道组数据中包含有车道线类型，因此可以从车道组数据中读取相应车道线段的车道线类型，然后将同一车道线段的车道线类型和线段偏移量进行组合，得到车道线段对应的特征子集，如图4所示，车道线3有四个车道线段，每个车道线段对应的特征子集分别为(0,50,虚线)、(50,70,虚线)、(70,100,实线)和(100,180,虚线)，此时还可以将具有相同车道线类型的相邻两个特征子集合成一个特征子集，从而得到车道线3的合成后的(0,70,虚线)和(70,180,实线)，然后将这两个特征子集进行组合，从而得到车道线3的特征集 对应地，车道线4的特征集
其中，和分别指车道2右侧的车道线和车道3左侧的车道线，从图4可以看出，和均指的是Line3，即车道线3；此外，和分别指车道3右侧的车道线和车道4左侧的车道线，从图4可以看出，和均指的是Line4，即车道线4。
在一个实施例中，终端从所述多个车道组的车道组数据中确定多个车道线标识，所述多个车道线标识分别用于唯一标识所述多条车道线中的一条；从所述多个车道组的车道组数据中读取对应于相同所述车道线标识的各所述车道线段的所述车道线类型。
其中，车道线标识可以是车道线的名称或编号。
例如，终端在多个车道组的车道组数据中获取车道线段的编号，然后根据车道线段的编号获得行车道路中各车道线的编号，如10个车道组数据中，每个车道组数据中均有编号为1的车道线段，此时根据这10个车道线段的编号可以得到对应车道线的编号，即可以确定该车道线属于车道线1(即1号车道线)；然后，对于每个车道线，终端可以从多个车道组的车道组数据读取车道线1的各车道线段的车道线类型。
在一个实施例中，终端根据车道组数据确定行车道路上的车道线；对行车道路上的车道线进行线段划分，得到各车道组中的车道线段；从车道组数据中读取各车道线段对应的车道线类型。
其中，在对行车道路上的车道线进行线段划分时，可以得到多个车道组，因此可以得到各车道组中的车道线段。
车道线段可以指一定长度的车道线。对于车道线的划分，可以采用如下方式：按照车道线类型的变化对车道线进行线段划分。如图4所示，在车道线4在距离起点50米(m)的位置处，车道线类型发生了变化，即由虚线转为实线，此时可以在50m位置处进行线段划分；在车道线3在距离起点70m的位置处，车道线类型发生了变化，即由虚线转为实线，此时可以在70m位置处进行线段划分；此外，在车道线4在距离起点100m的位置处，车道线类型再次发生了变化，即由实线转为虚线，此时可以在100m位置处进行线段划分，从而将这些车道线划分成4段。
S204，将多个特征子集中满足第一预设条件的特征子集从其所对应的特征集去除掉，得到剩余特征集。
其中，第一预设条件可以是车道线类型属于不可变道的实线型。因此满足第一预设条件的特征子集可以指：特征子集中的车道线类型和线段偏移量分别为实线型和实线型车道线段的线段偏移量。如图4所示，满足第一预设条件的车道线类型和线段偏移量可以是车道线4中的50m至100m之间的这个车道线段的车道线特征，即50m至100m之间的这个车道线段的实线型和线段偏移量。
去除可以指滤除、筛除或删除，例如将满足第一预设条件的特征子集从各特征集内中删除，或将第一预设条件的特征子集内的车道线类型和线段偏移量从各特征集内删除。
在一个实施例中，终端可以在特征集内，将满足第一预设条件的特征子集去除掉，得到与该特征集对应的剩余特征集以及其它未进行去除处理的特征集，需要指出的是，与该特征集对应的剩余特征集以及其它未进行去除处理的特征集均可统称为剩余特征集。具体地，终端可以在所有车道线的特征集内，将满足第一预设条件的特征子集去除掉，得到与所有特征 集一一对应的剩余特征集。
在一个实施例中，终端在将多个特征子集中满足第一预设条件的特征子集从其所对应的特征集去除掉之后，可以根据行车道路中各车道线编号对所得的剩余特征集进行组合，得到特征集序列。其中，车道线编号的顺序可以是从左到右的顺序，如行车道路最左侧的车道线编号为1，该行车道路的左侧第二个车道线编号为2，以此类推。需要指出的是，对于特征集序列对应的车道组附图，考虑到交通习惯，在相关附图中保留了实线型的车道线，实际应用中，特征集序列是剔除了实线型的车道线对应的特征，因此该特征集序列对应的车道组中不包含实线型的车道线。
在一个实施例中，由于特征集包括车道线中各车道线段的特征子集，特征子集包括车道线段的线段偏移量和车道线类型，因此在对车道线特征执行去除操作时，终端可以先在各特征集内，查找满足第一预设条件的目标特征子集，然后在各特征集中将目标特征子集去除掉，或目标特征子集内的线段偏移量和车道线类型去除掉，得到剩余特征集，如剩余特征集内不包含实线型的车道线类型和对应的线段偏移量。最后，终端将去除后所得的剩余特征集进行组合，得到特征集序列。
例如，对于图4中各车道线的特征集，车道线1的特征集为Line1＝{(0,180,实线)}，车道线2的特征集为Line2＝{(0,180,虚线)}，车道线3的特征集为Line3＝{(0,70,虚线),(70,180,实线)}，车道线4的特征集为Line4＝{(0,50,虚线),(50,100,实线),(100,180,虚线)}，车道线5的特征集为Line5＝{(0,180,实线)；将这些特征集内的车道线为实线的车道线特征进行去除，从而得到各车道线的剩余特征集，即：Line1＝空集，Line2＝{(0,180,虚线)}，Line3＝{(0,70,虚线)}，Line4＝{(0,50,虚线),(100,180,虚线)}，Line5＝空集。最后，按照车道线的顺序将各车道线的剩余特征集进行排序组合，得到特征集序列Line＝{Line1,Line2,Line3,Line4,Line5}＝{空集,{(0,180,虚线)},{(0,70,虚线)},{(0,50,虚线),(100,180,虚线)},空集}。
在获得特征集序列之后，终端可以在特征集序列中筛选位于行车位置和预设目标位置之间的车道线对应的特征集，筛选出来的特征集可以组成一个新的特征集序列，即特征集子序列。当特征集子序列中存在空集(即特征集子序列中有一个车道线的特征集为空集)、且第一车道与第二车道不是相同车道时，此时表示行车位置和预设目标位置之间存在至少一条不可变道的实线，因此可以判定车辆无法从行车位置所处的第一车道跨越(即变道)至预设目标位置所处的第二车道；当特征集子序列中不存在空集时，执行S206。
S206，在所述车辆行驶的过程中，确定车辆的行车位置在多个车道组中的当前位置偏移量，确定车辆的预设目标位置在多个车道组中的目标位置偏移量。
其中，行车位置处于多条车道中的第一车道，该行车位置可以是车辆当前的定位位置，该行车位置也可以称为行驶位置；此外，在一些应用场景中也可以是在行驶方向的前方任意一个需要进行车辆变道成功率预测的位置。预设目标位置处于多条车道中的第二车道，该预设目标位置可以是车辆通向终点位置所需经历的位置或所要到达的终点位置，若为通向终点位置所需经历的位置，则该预设目标位置可以是在进行路径规划之后，在所得的规划路径上确定的。车道组是行车道路中的与车道组数据对应的车道集合。需要指出的是，第一车道与第二车道可以处于不同车道，当第一车道与第二车道处于不同车道时，第一车道与第二车道可以是相邻车道，也可以是中间间隔至少一个车道的情况。
该行车位置的当前位置偏移量可以是在多个车道组的偏移量，例如行车位置多个车道组 中的车道路段(即车辆所在的车道路段)的起始点和该行车位置之间的偏移量；而目标位置偏移量可以是预设目标位置相对于行车位置所在的车道组的起点之间的偏移量，例如行车位置所处车道组中的车道路段的起始点和该预设目标位置之间的偏移量。例如，图5所示，Da为行车位置Pa在车道组中的当前位置偏移量，Db为预设目标位置Pb在车道组中的目标位置偏移量。需要指出的是，当前位置偏移量和目标位置偏移量是矢量，可以采用正负符号来表示偏移方向，采用数值大小来表示偏移距离。
S208，基于当前位置偏移量、目标位置偏移量、剩余特征集和车辆的行驶信息，预测车辆从第一车道变道至第二车道的成功率。
其中，车辆的行驶信息可以包括车辆的车速、加速度、行驶时间以及制动信息等信息中的至少一种。当行车位置为车辆当前的定位位置时，则该车辆的车速可以是车辆当前行驶的速度；当行车位置为在行驶方向的前方任意一个需要进行车辆变道成功率预测的位置时，则该车辆的车速可以是依据车辆当前行驶的速度进行预测所得的速度。该车辆的车速可以分解为相对于车道的平行速度和垂直速度，如图6所示。
成功率可以指车道从第一车道变道至第二车道的成功概率，当该成功率大于或等于概率阈值时，表示车道能够从第一车道变道至第二车道；当该成功率小于该概率阈值时，表示车道不能从第一车道变道至第二车道。该概率阈值可以根据实际的交通安全情况进行设定，如可以设定为95％。
在一个实施例中，终端在对车辆从第一车道变道至第二车道的成功率进行预测时，可以结合两种不同的垂直速度进行预测，两种不同的垂直速度包括第一垂直速度和第二垂直速度，该第一垂直速度可以指在遵守交通规则和交通安全的条件下的最大垂直速度，第二垂直速度可以是在不影响后车行驶的情况下的最小垂直速度；因此，根据第一垂直速度计算变道所需的第一时间(即最短时间)，根据第二垂直速度计算变道所需的第二时间(即最长时间)，将该第一时间和第二时间作为两个不同的时间阈值(即第一时间阈值和第二时间阈值)。因此，在剩余特征集或特征集序列中不存在空集的特征集的情况下，当车辆需要进行变道时，终端可以根据车辆实际的垂直车速计算变道时间，若变道时间落入第一时间阈值和第二时间阈值之间，则预测车辆从第一车道变道至第二车道的成功率大(如该成功率大于概率阈值)；若变道时间大于第二时间阈值，则预测车辆从第一车道变道至第二车道的成功率小；若变道时间落小于第一时间阈值，则预测车辆从第一车道变道至第二车道的成功率小，且可能存在安全隐患。
在一个实施例中，终端可以依据车辆的行驶信息确定车辆的行驶距离；然后，根据行驶距离和行车位置的当前位置偏移量，确定车辆的前行位置偏移量，依据前行位置偏移量、剩余特征集中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。此外，考虑到去除满足第一预设条件的特征子集之后所得的剩余特征集可以进行组合得到特征集序列，因此终端还可以依据前行位置偏移量、特征集序列中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。
其中，该前行位置偏移量可以是车辆向前行驶后的位置偏移量。
该特征集序列包括第一车道与所述第二车道之间的特征集子序列，因此在进行车辆变道成功率预测时，依据前行位置偏移量、特征集子序列中各车道线段的线段偏移量和目标位置偏移量，准确快速地算出车辆从第一车道变道至第二车道的成功率。
例如，终端根据行驶信息确定车辆的车速，根据该车速对应的垂直速度和车道宽度确定 跨越车道所需的耗时，以及基于该耗时和该车速对应的平行车速确定车辆的行驶距离，依据前行位置偏移量、特征集子序列中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。
需要说明的是，在进行车道变道成功率预测时，还可以结合行车道路的车况数据进行车辆变道成功率的预测，如依据车况数据、前行位置偏移量、特征集子序列中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率，从而可以在行车过程中有效地提高交通安全。
在预测出车辆第一车道与第二车道之间变道的成功率之后，可以将预测结果应用到以下两种场景，具体如下：
场景1，对车辆进行路径规划。
在一个实施例中，当根据该成功率确定车辆无法从第一车道跨越到第二车道时，终端可以重新规划通向预设目标位置的路径，然后发出新路径提示，从而可以提示用户按照重新规划的路径进行行驶，有利于提高交通安全。
例如，如图7所示，假设预设目标位置Pb是车辆的目的地，当车辆处于行车位置Pa时，根据本申请的技术方案判定车辆能从车道2逐步跨越到车道4，由于车辆在行车位置Pa处未及时进行变道，当车辆行驶到P′a时，根据本申请的技术方案判定车辆已经无法从车道2逐步跨越到车道4，因此重新规划到达预设目标位置Pb的路径，并发出新路径提示，从而避免因强行变道而造成交通安全隐患。
场景2，提示车辆进行变道。
在一个实施例中，当根据该成功率确定车辆能从第一车道逐步跨越到第二车道时，终端可以在电子地图上显示变道指引，提示用户按照变道指引进行变道，以使车辆从第一车道逐步跨越到第二车道。
例如，如图8所示，当根据本申请的技术方案判定车辆在行车位置Pa处能从车道2逐步跨越到车道4时，终端在电子地图的导航页面显示车道级变道指引，提示用户可以按照车道级变道指引进行变道，有利于提升交通安全。
上述实施例中，对车辆所处的行车道路的车道组数据进行车道线特征提取，得到各车道线分别对应的特征集；将各特征集内满足第一预设条件的特征子集去除掉，得到不包含满足第一预设条件的特征子集的剩余特征集，从而根据剩余特征集是否为空集，并结合剩余特征集与车辆的行驶信息、当前位置偏移量和目标位置偏移量，可以快速准确地预测出车辆行驶时从不同车道之间进行变道的成功率；此外，由于获取到不包含满足第一预设条件的特征子集的剩余特征集，因此即使行车位置所处的第一车道与预设目标位置所处的第二车道不属于相同车道，也可以简单地利用剩余特征集中行车位置与预设目标位置之间车道线对应的特征集为空，即可快速确定出车辆无法从行车位置所处的第一车道至预设目标位置所处的第二车道；当剩余特征集中的特征集不为空时，在进行车辆变道成功率预测的过程中，考虑到了车辆的行驶信息、车道线类型以及车辆行车位置的当前位置偏移量和预设目标位置的目标位置偏移量，可以准确地预测出车辆从第一车道变道至第二车道的成功率，提高了车辆变道成功率预测的准确率，而且覆盖的应用场景更广，有效提高交通安全。
在一个实施例中，车道组数据是对行车道路的原始车道组数据进行车道重组所得的，原始车道组数据是行车道路中各原始车道组对应的数据，因此终端可以对车道组数据进行车道重组。需要指出的是，对原始车道组数据进行车道重组可以指对原始车道组数据对应的地图 应用中的原始车道组进行重组(非物理重组)。考虑到各原始车道组之间的车道拓扑信息不同，可以分以下两种方式对车道组数据进行车道重组，具体如下：
方式1，扩展原始车道组，如图9所示，具体扩展方式如下：
S902，确定原始车道组数据中第一原始车道组的第一车道数量，以及确定原始车道组数据中第二原始车道组的第二车道数量。
其中，第一原始车道组和第二原始车道组是行车道路中互为相邻的车道组。原始车道组数据可以包括第一原始车道组数据和第二原始车道组数据，第一原始车道组数据是关于第一原始的未进行重组的车道组数据，第二原始车道组数据是关于第二原始的未进行重组的车道组数据。
如图10左侧的道路图所示，原始车道组Gi的车道数量为3，原始车道组Gi+1的车道数量为4，且原始车道组Gi+1左侧的边缘车道与原始车道组Gi左侧的边缘车道并不在相同车道上。
又如图11左侧的道路图所示，原始车道组Gi的车道数量为3，原始车道组Gi+1的车道数量为4，且原始车道组Gi+1右侧的边缘车道与原始车道组Gi右侧的边缘车道并不在相同车道上。
此外，还可以参考如图12左侧的道路图，原始车道组Gi的车道数量为3，原始车道组Gi+1的车道数量为4，且原始车道组Gi+1中间两个车道是原始车道组Gi的车道2划分出的两个车道。需要指出的是，图10～12示例了Gi的车道数量小于Gi+1的车道数量的情况，此外，还可以存在Gi的车道数量大于Gi+1的车道数量的情况。
S904，当第一车道数量小于第二车道数量时，在第一原始车道组中添加虚拟车道以更新原始车道组数据，该虚拟车道使得第一原始车道组与第二原始车道组保持车道对齐。
其中，添加的虚拟车道可以指实际行车道路中不存在的车道，为了方便且准确地预测出不同车道之间的车辆变道成功率，在相应的原始车道组中添加虚拟车道，从而可以得到用来进行车道变道成功预测的车道组。车道对齐可以指第一原始车道组中的车道与第二原始车道组中相应的一个车道在行车道路的延伸方向(即行车方向)上对齐，从而确保相邻的车道组之间的各车道数量一致(即相等)。例如，如图10的右图所示，车道组Gi中的每个车道，在车道组Gi+1中均存在相应的车道与之对应，而且车道组中的车道数量也是相同的，即车道组Gi具有4个车道路段，其相邻的车道组Gi+1也具有4个车道路段。
需要指出的是，在第一原始车道组中添加虚拟车道之后，该原始车道组数据也相应的发生更新，即在原始车道组数据中新增虚拟车道以及该虚拟车道的标识；此外，还可以在原始车道组数据标记该虚拟车道的生成时间和存储位置等数据。因此，可以将新增上述数据后的原始车道组数据作为所需的车道组数据。
对于虚拟车道的添加，可以按照虚拟车道的添加位置划分为以下两种场景，具体如下：
场景a，在原始车道组的边缘添加虚拟车道。
在一个实施例中，当第一车道数量小于第二车道数量、且第二原始车道组内的边缘车道路段与第一原始车道组内的边缘车道路段不在相同车道时，终端则在第一原始车道组中，添加与第二原始车道组的边缘车道路段对齐的虚拟车道。
其中，边缘车道路段对齐可以指添加的虚拟车道与第二原始车道组中对应的边缘车道路段处于相同车道，即两车道之间的车道拓扑信息是连通的。
例如，如图10所示，由于原始车道组Gi的车道数量小于原始车道组Gi+1的车道数量，因此在原始车道组Gi的左侧添加一条虚拟车道，从而与原始车道组Gi+1的车道数量保持一致，从 而得到图10右侧的经过重组的车道组。
又如图11所示，由于原始车道组Gi的车道数量小于原始车道组Gi+1的车道数量，因此在原始车道组Gi的右侧添加一条虚拟车道，从而与原始车道组Gi+1的车道数量保持一致，从而得到图11右侧的经过重组的车道组。
场景b，在原始车道组的中间添加虚拟车道。
在一个实施例中，当第一车道数量小于第二车道数量、且第一原始车道组内的中间车道路段分流为第二原始车道组内的至少两个车道路段时，终端则将第一原始车道组内的中间车道路段，转换为与第二原始车道组内的至少两个车道路段分别对齐的至少两个虚拟车道。
例如，如图12所示，由于原始车道组Gi的车道数量小于原始车道组Gi+1的车道数量，而且原始车道组Gi内的车道路段2分流为原始车道组Gi+1内的车道路段2～3，因此在原始车道组Gi的车道路段2转换为与原始车道组Gi+1内的车道路段2～3分别对齐的两个虚拟车道，从而与原始车道组Gi+1的车道数量保持一致，从而得到图12右侧的经过重组的车道组。
需要指出的是，上述场景a～b示例了Gi的车道数量小于Gi+1的车道数量的情况，此外，还可以存在Gi的车道数量大于Gi+1的车道数量的情况，此时可以进行如下操作：在Gi+1的左侧补上一个虚拟车道，或者在Gi+1的右侧补上一个虚拟车道，这两种情况分别与图11和图11相似，区别在于图10和图11中Gi的车道数量为3，Gi+1的车道数量为4，而在本实施例中，Gi的车道数量为4，Gi+1的车道数量为3；又或者，可以将Gi+1中间要拓展的车道替换成两个虚拟车道，这种情况与图12相似，区别在于图12中Gi的车道数量为3，Gi+1的车道数量为4，Gi中的车道2分流成Gi+1中的两个车道(如Gi+1中的车道2和车道3)，而在本实施例中，Gi的车道数量为4，Gi+1的车道数量为3，且Gi的车道2和车道3在Gi+1中合并为一个车道。
S906，在原始车道组数据中为虚拟车道配置车道特征，获得车道组数据。
其中，该车道特征包括车道宽度和对应的车道线类型。在配置车道特征时，可以在上述更新后的原始车道组数据中为虚拟车道配置车道特征。
对于上述场景a的情况，在添加虚拟车道之后，终端还可以在原始车道组数据中为虚拟车道配置第一虚拟车道宽度和第一虚拟车道线。其中，第一虚拟车道线是表征禁止变道的属于实线型的车道线，第一虚拟车道宽度等于目标宽度值。
在添加虚拟车道之后，为了确保该段行车道路整体的车道宽度不变，可以将该虚拟车道的车道宽度设置为0；此外，考虑到添加的虚拟车道在实际行车道路中不存在对应的车道，是无法在该虚拟车道上行驶、且无法变道至该虚拟车道，因此该虚拟车道对应的车道线设置为实线，即车道线类型为实线型。
对于上述场景b的情况，在添加完虚拟车道之后，终端还可以在至少两个虚拟车道之间配置虚线型的第二虚拟车道线；在原始车道组数据中为至少两个虚拟车道分别配置第二虚拟车道宽度；其中，至少两个虚拟车道的第二虚拟车道宽度之间的和值与第一原始车道组内的中间车道路段的车道宽度一致。
在添加虚拟车道之后，为了确保该段行车道路整体的车道宽度不变，可以将图12的这两条虚拟车道的车道宽度分别设置为原来车道的一半；此外，考虑这两条虚拟车道实际是一条车道，因此这两条虚拟车道之间设置虚线型的车道线。
上述实施例中，通过将未对齐的车道路段进行车道重组，以使重组后的各车道组之间的车道对齐，从而可以有利于在不同车道之间进行车辆变道成功率的预测，此外，在进行计算时，避免进行不同车道组中车道数量的逻辑判断，因此在自动驾驶(或辅助驾驶)时可以有 效地提高计算速度，提高计算效率；而且在开发阶段有效降低代码复杂度，以及提高代码的复用性。
方式2，将弯曲的原始车道组进行拉直。
在一个实施例中，车道组数据是对行车道路的原始车道组数据进行车道重组所得的，原始车道组数据是行车道路中各原始车道组对应的数据；因此，终端可以获取行车道路的原始车道组数据；当原始车道组数据中的原始车道组是存在曲率的车道组时，将原始车道组数据中的原始车道组转换为直线型的车道组。
其中，曲率可以指原始车道组的弯曲程度。例如，如图13所示，当原始车道组是弯曲的车道组时，将原本弯曲的原始车道组转换为直的车道组。
上述实施例中，通过原本弯曲的原始车道组转换为直的车道组，有利于变道所需耗时的计算，即在计算过程中无需考虑曲率，从而可以有利于在不同车道之间进行车辆变道成功率预测，还可以有效降低代码复杂度，以及提高代码的复用性。
在一个实施例中，车辆的车速包括相对于车道的平行速度和垂直速度；特征集序列包括第一车道与第二车道之间的特征集子序列，因此对于车辆在第一车道与第二车道之间的车辆变道成功率预测，如图14所示，具体方法可以包括：
S1402，依据垂直速度和车道宽度，确定跨越车道组内各车道的耗时。
其中，跨越车道组内各车道可以指车辆从车道组的一个车道变道至其它车道的耗时。
需要指出的是，在原始车道组中，各车道的车道宽度通常为3.5～3.75m，在后续实施例中，不妨令车道宽度为3.5m。由于重组后所得的车道组中包含虚拟车道，因此车道宽度可能为0、3.5/L和3.5。其中，n为在原始车道组的中间添加的虚拟车道数量；当车道宽度为0时，表示在原始车道组的边缘添加了虚拟车道；当该车道宽度3.5/L时，表示在原始车道组的中间添加了L个虚拟车道；当该车道宽度3.5时，表示并未在原始车道组内添加虚拟车道。
如图6所示，终端可以利用车辆的垂直速度V⊥和车道宽度，计算跨越车道组内各车道的耗时t。需要说明的是，在计算耗时t时，考虑到车辆是从一个车道变道至另一个车道，当车辆从车道2变道至车道3，此时耗时其中，w2为车道2的车道宽度，w3为车道3的车道宽度。
S1404，基于耗时和平行速度，确定车辆的行驶距离。
例如，在跨越车道的这个过程中，车辆同时会向前行驶，此时可以根据耗时t和平行车速V∥计算出车辆向前行驶的距离。
S1406，根据行驶距离和当前位置偏移量，确定车辆的前行位置偏移量。
在一个实施例中，终端计算行驶距离和当前位置偏移量之间的和值，然后将该和值作为车辆的前行位置偏移量。
S1408，依据前行位置偏移量、特征集子序列中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。
上述实施例中，确定车辆的行车位置在车道组中的当前位置偏移量，确定预设目标位置在车道组中的目标位置偏移量，该车道组是行车道路中的与车道组数据对应的车道集合，因此对于车辆变道成功率的预测，可以利用当前位置偏移量、目标位置偏移量、特征集子序列、车辆的车速和车道宽度进行综合计算，考虑到了车速、车道宽度、车道线类型以及当前位置偏移量和目标位置偏移量，可以准确地算出车辆能从第一车道跨越至第二车道的成功率，提高了车辆变道时变道成功率的准确性，而且覆盖的应用场景更广，有效提高交通安全。
考虑到车辆所处的第一车道可能位于第二车道的左侧，如图15的(a)图所示；此外，车辆所处的第一车道也可能位于第二车道的右侧，如图15的(b)图所示；车辆所处的第一车道也可能与第二车道位于相同车道(即车辆通过变道至其它车道，然后再变道回第一车道)，如图15的(c)图所示。因此，在进行车辆变道成功率预测时，可以划分以下几种场景进行说明，具体如下：
场景1，车辆所处的第一车道位于第二车道的左侧。
在一个实施例中，线段偏移量包括末端偏移量，特征集子序列的车道线数量(即特征集子序列中所有车道线段的数量)为n，n为不小于1的正整数；因此，终端在特征集子序列的第i个车道线中，沿车辆的行驶方向遍历各车道线段的特征子集，得到遍历结果；确定遍历结果中存在满足第二预设条件的目标特征子集；第二预设条件为前行位置偏移量不大于第i个车道线对应的末端偏移量、且不大于目标位置偏移量；按照预设步长对i进行自增处理，循环执行上述步骤(即本实施例中的上述各步骤)，直至当i＝n，且第i个车道线的遍历结果中存在满足第二预设条件的目标特征子集时，确定车辆能从第一车道变道至第二车道。此外，当确定遍历结果中不存在满足第二预设条件的目标特征子集时，确定车辆无法从第一车道变道至第二车道。因此，通过上述实施例，即使车辆所处的第一车道与预设目标位置所处的第二车道不属于相同车道，并且第一车道与第二车道是相邻车道还是间隔多个车道，也可以快速且准确地计算出车辆在第一车道和第二车道之间变道的成功率。
例如，如图15(a)所示，结合伪代码进行说明，具体如下：
初始化变量ForwardDis＝Da；
//对Pa和Pb之间的L根线进行遍历
for(int i＝1；i<＝L；i++){
//第i根线对应跨越第A+i-1个车道
1)用Wi表示第A+i-1个车道的车道宽度；
2)则跨越该车道的最短时间为
3)跨越该车道的时间内车辆前向运动的距离di＝V∥*ti；
4)此时ForwardDis＝ForwardDis+di；
5)正向遍历Linei中的所有三元组(start,end,type)，找到满足以下条件的车道线段：
1>start≦ForwardDis≦end，则ForWardDis保持不变，退出遍历过程；
2>ForwardDis≦start，则ForWardDis＝start，退出遍历过程；
如果没有满足1>和2>的三元组，则返回A-B不可通行，退出算法
}
if(ForwardDis≤Db){
返回A-B可通行
}else
{返回A-B不可通行}
其中，上述的A、B分别为Pa和Pb所处的车道。
场景2，车辆所处的第一车道位于第二车道的右侧。
在一个实施例中，终端在特征集子序列对应的第i个车道线中，沿车辆的行驶方向遍历各车道线段的特征子集，得到遍历结果；确定遍历结果中存在满足第二预设条件的目标特征 子集；第二预设条件为前行位置偏移量不大于第i个车道线对应的末端偏移量、且不大于目标位置偏移量；按照预设步长对i进行自减处理，循环执行上述步骤，直至当i＝1，且第i个车道线的遍历结果中存在满足第二预设条件的目标特征子集时，确定车辆能从第一车道变道至第二车道。此外，当确定遍历结果中不存在满足第二预设条件的目标特征子集时，确定车辆无法从第一车道变道至第二车道。因此，通过上述实施例，即使车辆所处的第一车道与预设目标位置所处的第二车道不属于相同车道，并且第一车道与第二车道是相邻车道还是间隔多个车道，也可以快速且准确地计算出车辆在第一车道和第二车道之间变道的成功率。
其中，对于车辆所处的第一车道位于第二车道的右侧的情况，可以参考上述场景1以及图15(b)。
场景3，车辆所处的第一车道与第二车道属于相同车道。
特别地，当第一车道与第二车道属于相同车道时，终端依据前行位置偏移量、特征集序列(或特征集子序列)中各车道线段的线段偏移量和目标位置偏移量，可以直接确定车辆可以从第一车道跨越至第二车道，从而可以快速计算出车辆变道的成功率。
作为一个示例，为了更加直观了解本申请的技术方案，这里结合图3b、图4-7、图10-13以及图16进行说明，具体如下所述：
(一)获取车辆的车速、起点位置和预设目标位置。
获取车辆的起点位置Pa(lon,lat)和预设目标位置Pb(lon,lat)，其中lon和lat分别表示经度和纬度。这里的Pa可以是车辆当前的定位位置，也可以是任意的一个需要判断的起点位置。
此外，还会获取车辆的车速V，该车速V可以是车辆真实的行驶速度，也可以是一个任意预设的速度，或根据车辆真实的行驶速度进行预测所得的速度。
(二)获取原始的车道组数据。
获取Pa和Pb之间的所有高精数据或车道级数据，如图3b所示；然后从该高精数据或车道级数据中获取包含Pa和Pb之内的所有车道组数据，该车道组数据包括：多个原始的车道组、每个车道组记录的车道数量、各车道左右的车道线类型及其对应区间、各车道上下游的车道级连通关系、车道级中心线形点和车道宽度等。
如果Pa和Pb之间缺失高精数据和车道级数据，则无法进行后续计算，此时车道跨越的通行性(可采用车辆变道成功率来衡量)判定流程终止。
(三)对原始的车道组进行转换。
由于Pa和Pb之间的车道组数据包含多个车道组，记为G＝{G1,G2,…,Gn}，每个车道组G的车道数量前后可能发生变化，不利于计算，因此将所有原始的车道组拼接成一个新的虚拟的车道组G’用于后续计算，具体步骤如下：
对所有的车道组G，对各相邻的车道组Gi和Gi+1进行重组，直至完成所有车道组G的重组，得到一个完整的车道组G’。
一般地，假设现在完成了G1至Gi的重组，得到了Gi′，接下来说明将Gi′和Gi+1的重组方法，对所有的可能进行分类讨论：
场景1，如图10所示，Gi+1的左侧拓展了一个车道，此时可以在Gi的左侧补上一个虚拟车道，该虚拟车道的左车道线类型为边缘实线，右侧为实线，从交规上该虚拟车道无法到达，只是一个占位车道，其车道宽度可以赋值为0(只是为了在逻辑上生成一个车道，保证道路宽度不变)，得到图10右侧的车道组G’。同理，对于左侧拓展n个车道的情况，在左侧补上n个虚拟车道。
场景2，如图11所示，Gi+1的右侧拓展了一个车道，此时可以在Gi的右侧补上一个虚拟车道，该虚拟车道的右车道线类型为边缘实线，左侧为实线。从交规上该虚拟车道无法到达，只是一个占位车道，其车道宽度可以赋值为0，得到图11右侧的车道组G’。同理，对于右侧拓展n个车道的情况，在左侧补上n个虚拟车道。
场景3，如图12所示，Gi+1中间拓展一个车道，此时可以将Gi中间要拓展的车道替换成两个虚拟车道(如图12左侧的车道2替换成右侧的虚拟车道2～3)，该虚拟车道的左右车道线类型均为虚线，表示这两车道可以连通。需要注意的是，这两个虚拟车道的车道宽度均在原来车道宽度的一半，得到图12右侧的车道组G’。同理，对于中间拓展m车道的情况，将该拓展车道替换成m个虚拟车道，每个虚拟车道的左右车道线均为虚线，每个虚拟车道的车道宽度设置为原来车道宽度的1/m。
需要指出的是，上述场景1～3示例了Gi的车道数量小于Gi+1的车道数量的情况，此外，还可以存在Gi的车道数量大于Gi+1的车道数量的情况，此时可以进行如下操作：在Gi+1的左侧补上一个虚拟车道，或者在Gi+1的右侧补上一个虚拟车道，这两种情况分别与图11和图11相似，区别在于图10和图11中Gi的车道数量为3，Gi+1的车道数量为4，而在本实施例中，Gi的车道数量为4，Gi+1的车道数量为3；又或者，可以将Gi+1中间要拓展的车道替换成两个虚拟车道，这种情况与图12相似，区别在于图12中Gi的车道数量为3，Gi+1的车道数量为4，Gi中的车道2分流成Gi+1中的两个车道(如Gi+1中的车道2和车道3)，而在本实施例中，Gi的车道数量为4，Gi+1的车道数量为3，且Gi的车道2和车道3在Gi+1中合并为一个车道。在对虚拟车道进行车道特征配置时，可以参考上述场景1～3。
根据上述重组方式，将Pa和Pb之间的所有G1,G2,…,Gn进行重组，得到一个新的车道组G’。
需要注意的是，如果是带有曲率的车道组，可以基于车道中心线的长度将其转换为直的车道组，如图13所示。
(四)Pa至Pb之间的变道成功率预测。
在获得新的车道组G’之后，车道组G’包含各个车道i的车道线特征集以及宽度信息w，其中每个车道线特征集可以包含每段车道线对应的三元组(start,end,type)，start和end分别表示每段车道线的起点和终点相较于车道组G’起点的偏移量(即起点偏移量和末端偏移量)，type记录线类型。如图4所示，可以得到：

如图15所示，图15展示了三种可能的Pa和Pb的相对关系。将Pa所在的车道号记为A，Pb所在的车道号记为B，则存在A>B、A<B和A＝B这三种情况。其中，A<B表示Pb在Pa的右侧，对应图15的(a)图；A＞B表示Pb在Pa的左侧，对应图15的(b)图；A＝B表示Pa和Pb处于相同车道，对应图15的(c)图。
(1)对于A＝B的情况，无需进一步判断，直接确定A～B可通行。
(2)对于A<B的情况
首先获取从车道A通行到车道B所需要跨越的所有L根车道线特征，其中L＝B-A，每根车道线包含三元组，对该三元组进行删除，只保留每根线type＝虚线(或者左实右虚的双线类型，表示可以跨越)的部分，得到特征集序列Line：
Line＝{Line1,…,LineL}
其中1到L为车辆从车道A行驶到车道B的车道线跨越顺序。如果存在某个Linei集合为空，则表示存在某车道线无法跨越，直接返回A～B不可通行(即车辆不能从车道A变道至车道B)。
若所有Linei均非空，则进行下一步判断：
车速V可以分解成平行于车道线的分量V∥(简称平行车速)以及垂直于车道线的分量V⊥(简称垂直速度)，如图6所示。为了保证交通安全，V⊥要尽量小，保证车辆平稳跨越车道，因此可以设置一个最大的垂直速度比如车道宽度为W，可以假设变1个车道最少需要2s，则可设置此时有:
接着，获得Pa和Pb在车道组G′内的位置偏移量，记为Da和Db，如图5所示，结合以下伪代码进行车辆变道成功率预测：
初始化变量ForwardDis＝Da；
//对Pa和Pb之间的L根线进行遍历
for(int i＝1；i<＝L；i++){
//第i根线对应跨越第A+i-1个车道；
1)用Wi表示第A+i-1个车道的车道宽度；
2)则跨越车道的最短时间为
3)跨越车道的这个时间内车辆向前运动的距离di＝V∥*ti；
4)此时ForwardDis＝ForwardDis+di；
5)正向遍历Linei中的所有三元组(start,end,type)，找到满足以下条件的车道线段：
1>start≦ForwardDis≦end，则ForWardDis保持不变，退出遍历过程；
2>ForwardDis≦start，则ForWardDis＝start，退出遍历过程；
如果没有满足1>和2>的三元组，则返回A-B不可通行，退出算法
}
if(ForwardDis≤Db){
返回A-B可通行(即车辆能从车道A变道至车道B)；
}else
{返回A-B不可通行}
(3)对于A>B的情况，可以参考上述方案(2)。
例如，如图16所示，当车辆需要在前方500米转入XX高速方向，采用本申请的方案，可以快速判断出当前位置或前方的某个位置是否可以变道至通向XX高速方向的车道，以便提示用户提前变道，或者在错过最后的变道时机时，可以及时为用户规划新的路径。
本申请的方案综合考虑了车速、车道宽度、跨越车道数、交通规则和交通安全等多方位因素来判断不同车道之间是否可以跨越通行，相较于传统跨越关系判断算法更具一般性和普适性，能应用于智能车控制技术领域、自动驾驶领域和高级辅助驾驶领域，对于车道级规划和车道级偏航等具体应用有帮助。此外，有了车道级规划，可以更高效地分配道路资源，能应用于智慧城市交通项目，缓解交通拥堵，合理分配车道资源使用，可以有效缓解交通拥堵、降低交通事故率、改善交通安全、降低能源消耗和环境污染等。
应该理解的是，虽然如上所述的各实施例所涉及的流程图中的各个步骤按照箭头的指示 依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，如上所述的各实施例所涉及的流程图中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
基于同样的发明构思，本申请实施例还提供了一种用于实现上述所涉及的车辆变道成功率的预测方法的车辆变道成功率的预测装置。该装置所提供的解决问题的实现方案与上述方法中所记载的实现方案相似，故下面所提供的一个或多个车辆变道成功率的预测装置实施例中的具体限定可以参见上文中对于车辆变道成功率的预测方法的限定，在此不再赘述。
在一个实施例中，如图17所示，提供了一种车辆变道成功率的预测装置，车辆处于行车道路中，行车道路包含多个车道组，每个车道组包括多条车道的车道路段，该装置包括：提取模块1702、处理模块1704、确定模块1706和判定模块1708，其中：
提取模块1702，用于获取多个车道组的车道组数据；从多个车道组的车道组数据确定行车道路的多条车道线分别对应的特征集，特征集分别包括对应于多个车道组的多个特征子集，每个特征子集包括对应的车道线类型和线段偏移量；
处理模块1704，用于将多个特征子集中满足第一预设条件的特征子集从其所对应的特征集去除掉，得到剩余特征集；
确定模块1706，用于在车辆行驶的过程中，确定车辆的行车位置在多个车道组中的当前位置偏移量，行车位置处于多条车道中的第一车道；确定车辆的预设目标位置在多个车道组中的目标位置偏移量，预设目标位置处于多条车道中不同于第一车道的第二车道；
判定模块1708，用于基于当前位置偏移量、目标位置偏移量、剩余特征集和车辆的行驶信息，预测车辆从第一车道变道至第二车道的成功率。
上述实施例中，对车辆所处的行车道路的车道组数据进行车道线特征提取，得到各车道线分别对应的特征集；将各特征集内满足第一预设条件的特征子集去除掉，得到不包含满足第一预设条件的特征子集的剩余特征集，从而根据剩余特征集是否为空集，并结合剩余特征集与车辆的行驶信息、当前位置偏移量和目标位置偏移量，可以快速准确地预测出车辆行驶时从不同车道之间进行变道的成功率；此外，由于获取到不包含满足第一预设条件的特征子集的剩余特征集，因此即使行车位置所处的第一车道与预设目标位置所处的第二车道不属于相同车道，也可以简单地利用剩余特征集中行车位置与预设目标位置之间车道线对应的特征集为空，即可快速确定出车辆无法从行车位置所处的第一车道至预设目标位置所处的第二车道；当剩余特征集中的特征集不为空时，在进行车辆变道成功率预测的过程中，考虑到了车辆的行驶信息、车道线类型以及车辆行车位置的当前位置偏移量和预设目标位置的目标位置偏移量，可以准确地预测出车辆从第一车道变道至第二车道的成功率，提高了车辆变道成功率预测的准确率，而且覆盖的应用场景更广，有效提高交通安全。
在其中的一个实施例中，提取模块1702，还用于从多个车道组的车道组数据中，读取各车道线的分别对应于多个车道组中的多个车道线段的车道线类型，并确定多条车道线段在各自处于的车道组中的线段偏移量；对每个车道线段，将其对应的车道线类型和线段偏移量进行组合，得到各车道线段分别对应的特征子集；对每条车道线，将其包含的车道线段对应的特征子集进行组合，得到各车道线分别对应的特征集。
在其中的一个实施例中，提取模块1702，还用于从多个车道组的车道组数据中确定多个车道线标识，多个车道线标识分别用于唯一标识多条车道线中的一条；从多个车道组的车道组数据中读取对应于相同车道线标识的各车道线段的车道线类型。
在其中的一个实施例中，车道组数据是对行车道路的原始车道组数据进行车道重组所得的，原始车道组数据是行车道路中各原始车道组对应的数据；
如图18所示，该装置还包括：第一重组模块1710，用于确定原始车道组数据中第一原始车道组的第一车道数量，以及确定原始车道组数据中第二原始车道组的第二车道数量，第一原始车道组和第二原始车道组是行车道路中互为相邻的车道组；当第一车道数量小于第二车道数量时，在第一原始车道组中添加虚拟车道以更新原始车道组数据，虚拟车道使得第一原始车道组与第二原始车道组达到车道对齐；可选地，车道重组还包括：在原始车道组数据中为虚拟车道配置车道特征，得到车道组数据。
在其中的一个实施例中，第一重组模块1710，还用于当第一车道数量小于第二车道数量、且第二原始车道组内的边缘车道路段与第一原始车道组内的边缘车道路段不在相同车道时，则在第一原始车道组中，添加与第二原始车道组的边缘车道路段对齐的虚拟车道。
在其中的一个实施例中，第一重组模块1710，还用于在原始车道组数据中为虚拟车道配置第一虚拟车道宽度和第一虚拟车道线；其中，第一虚拟车道线是表征禁止变道的属于实线型的车道线，第一虚拟车道宽度等于目标宽度值。
上述实施例中，通过将未对齐的车道路段进行车道重组，以使重组后的各车道组之间的车道对齐，从而可以有利于在不同车道之间进行车辆变道成功率的预测，此外，在进行计算时，避免进行不同车道组中车道数量的逻辑判断，因此在自动驾驶(或辅助驾驶)时可以有效地提高计算速度，提高计算效率；而且在开发阶段有效降低代码复杂度，以及提高代码的复用性。
在其中的一个实施例中，第一重组模块1710，还用于当第一车道数量小于第二车道数量、且第一原始车道组内的中间车道路段分流为第二原始车道组内的至少两个车道路段时，则将第一原始车道组内的中间车道路段，转换为与第二原始车道组内的至少两个车道路段分别对齐的至少两个虚拟车道。
在其中的一个实施例中，第一重组模块1710，还用于在至少两个虚拟车道之间配置虚线型的第二虚拟车道线；在原始车道组数据中为至少两个虚拟车道分别配置第二虚拟车道宽度；其中，至少两个虚拟车道的第二虚拟车道宽度之间的和值与中间车道路段的车道宽度一致。
上述实施例中，通过原本弯曲的原始车道组转换为直的车道组，有利于变道所需耗时的计算，即在计算过程中无需考虑曲率，从而可以有利于在不同车道之间进行车辆变道成功率的预测，还可以有效降低代码复杂度，以及提高代码的复用性。
在其中的一个实施例中，车道组数据是对行车道路的原始车道组数据进行车道重组所得的，原始车道组数据是行车道路中各原始车道组对应的数据；
如图18所示，该装置还包括：第二重组模块1712，用于获取行车道路的原始车道组数据；当原始车道组数据中的原始车道组是存在曲率的车道组时，将原始车道组数据中的原始车道组转换为直线型的车道组。
在其中的一个实施例中，特征集包括车道线中各车道线段的特征子集，特征子集包括车道线段的线段偏移量和车道线类型；
处理模块1704，还用于在各特征集内，查找满足第一预设条件的目标特征子集；在各特 征集中将目标特征子集去除掉，或目标特征子集内的线段偏移量和车道线类型去除掉。
在其中的一个实施例中，判定模块1708，还用于依据车辆的行驶信息确定车辆的行驶距离；根据行驶距离和当前位置偏移量，确定车辆的前行位置偏移量；依据前行位置偏移量、剩余特征集中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。
在其中的一个实施例中，车辆的行驶信息包括相对于车道的平行速度和垂直速度；剩余特征集包括第一车道与第二车道之间的特征集子序列；
判定模块1708，还用于依据垂直速度和车道宽度，确定跨越车道组内各车道的耗时；基于耗时和平行速度，确定车辆的行驶距离；根据行驶距离和行车位置的位置偏移量，确定车辆的前行位置偏移量；依据前行位置偏移量、特征集子序列中各车道线段的线段偏移量和目标位置偏移量，预测车辆从第一车道变道至第二车道的成功率。
在其中的一个实施例中，线段偏移量包括末端偏移量，特征集子序列对应的车道线数量为n，n为不小于1的正整数；
判定模块1708，还用于在特征集序列的第i个车道线中，沿车辆的行驶方向遍历各车道线段的特征子集，得到遍历结果；确定遍历结果中存在满足第二预设条件的目标特征子集；第二预设条件为前行位置偏移量不大于第i个车道线对应的末端偏移量、且不大于预设目标位置的位置偏移量；按照预设步长对i进行自增或自减处理，循环执行上述步骤，直至当i＝n或i＝1，且第i个车道线的遍历结果中存在满足第二预设条件的目标特征子集时，确定车辆能从第一车道变道至第二车道。
在其中的一个实施例中，判定模块1708，还用于当确定遍历结果中不存在满足第二预设条件的目标特征子集时，确定车辆无法从第一车道变道至第二车道。
上述实施例中，确定车辆的行车位置在车道组中的当前位置偏移量，确定预设目标位置在车道组中的目标位置偏移量，该车道组是行车道路中的与车道组数据对应的车道集合，因此对于车辆变道成功率的预测，可以利用当前位置偏移量、目标位置偏移量、特征集子序列、车辆的车速和车道宽度进行综合计算，考虑到了车速、车道宽度、车道线类型以及当前位置偏移量和目标位置偏移量，可以准确地算出车辆能从第一车道跨越至第二车道的成功率，提高了车辆变道时变道成功率的准确性，而且覆盖的应用场景更广，有效提高交通安全。
上述车辆变道成功率的预测装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在一个实施例中，提供了一种计算机设备，该计算机设备可以是终端，其内部结构图可以如图19所示。该计算机设备包括处理器、存储器、输入/输出接口、通信接口、显示单元和输入装置。其中，处理器、存储器和输入/输出接口通过系统总线连接，通信接口、显示单元和输入装置通过输入/输出接口连接到系统总线。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机程序。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的输入/输出接口用于处理器与外部设备之间交换信息。该计算机设备的通信接口用于与外部的终端进行有线或无线方式的通信，无线方式可通过WIFI、移动蜂窝网络、NFC(近场通信)或其他技术实现。该计算机程序被处理器执行时以实现一种车辆变道成功率的预测方法。该计算机设备的显示单元用于形成视觉可见的画面，可以是 显示屏、投影装置或虚拟现实成像装置，显示屏可以是液晶显示屏或电子墨水显示屏，该计算机设备的输入装置可以是显示屏上覆盖的触摸层，也可以是计算机设备外壳上设置的按键、轨迹球或触控板，还可以是外接的键盘、触控板或鼠标等。
本领域技术人员可以理解，图19中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
在一个实施例中，提供了一种计算机设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现上述车辆变道成功率的预测方法的步骤。
在一个实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时实现上述车辆变道成功率的预测方法的步骤。
在一个实施例中，提供了一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现上述车辆变道成功率的预测方法的步骤。
需要说明的是，本申请所涉及的用户信息(包括但不限于用户设备信息、用户个人信息等)和数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)，均为经用户授权或者经过各方充分授权的信息和数据，且相关数据的收集、使用和处理需要遵守相关国家和地区的相关法律法规和标准。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存、光存储器、高密度嵌入式非易失性存储器、阻变存储器(ReRAM)、磁变存储器(Magnetoresistive Random Access Memory，MRAM)、铁电存储器(Ferroelectric Random Access Memory，FRAM)、相变存储器(Phase Change Memory，PCM)、石墨烯存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器等。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic Random Access Memory，DRAM)等。本申请所提供的各实施例中所涉及的数据库可包括关系型数据库和非关系型数据库中至少一种。非关系型数据库可包括基于区块链的分布式数据库等，不限于此。本申请所提供的各实施例中所涉及的处理器可为通用处理器、中央处理器、图形处理器、数字信号处理器、可编程逻辑器、基于量子计算的数据处理逻辑器等，不限于此。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种车辆变道成功率的预测方法，所述车辆处于行车道路中，所述行车道路包含多个车道组，每个车道组包括多条车道的车道路段，所述方法由计算机设备执行，其中，所述方法包括：
    获取所述多个车道组的车道组数据；
    从所述多个车道组的车道组数据确定所述行车道路的多条车道线分别对应的特征集，所述特征集分别包括对应于所述多个车道组的多个特征子集，每个所述特征子集包括对应的车道线类型和线段偏移量；
    将所述多个特征子集中满足第一预设条件的特征子集从其所对应的所述特征集去除掉，得到剩余特征集；
    在所述车辆行驶的过程中，确定所述车辆的行车位置在所述多个车道组中的当前位置偏移量，所述行车位置处于所述多条车道中的第一车道；
    确定所述车辆的预设目标位置在所述多个车道组中的目标位置偏移量，所述预设目标位置处于所述多条车道中不同于所述第一车道的第二车道；
    基于所述当前位置偏移量、所述目标位置偏移量、所述剩余特征集和所述车辆的行驶信息，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
  2. 根据权利要求1所述的方法，其中，所述从所述多个车道组的车道组数据提取得到各车道线分别对应的特征集，包括：
    从所述多个车道组的车道组数据中，读取所述各车道线的分别对应于所述多个车道组中的多个车道线段的车道线类型，并确定所述多条车道线段在各自处于的所述车道组中的线段偏移量；
    对每个所述车道线段，将其对应的所述车道线类型和所述线段偏移量进行组合，得到各所述车道线段分别对应的所述特征子集；
    对每条所述车道线，将其包含的所述车道线段对应的所述特征子集进行组合，得到各所述车道线分别对应的所述特征集。
  3. 根据权利要求2所述的方法，其中，所述从所述多个车道组的车道组数据中，读取所述各车道线的分别对应于所述多个车道组中的多条车道线段的车道线类型，包括：
    从所述多个车道组的车道组数据中确定多个车道线标识，所述多个车道线标识分别用于唯一标识所述多条车道线中的一条；
    从所述多个车道组的车道组数据中读取对应于相同所述车道线标识的各所述车道线段的所述车道线类型。
  4. 根据权利要求1至3中的任一项所述的方法，其中，所述车道组数据是对所述行车道路的原始车道组数据进行车道重组所得的，所述原始车道组数据是所述行车道路中各原始车道组的数据；
    所述对所述行车道路的原始车道组数据进行车道重组包括：
    确定所述原始车道组数据中第一原始车道组的第一车道数量，以及确定所述原始车道组数据中第二原始车道组的第二车道数量，所述第一原始车道组和所述第二原始车道组是所述行车道路中互为相邻的车道组；
    当所述第一车道数量小于所述第二车道数量时，在所述第一原始车道组中添加虚拟车道以更新所述原始车道组数据，所述虚拟车道使得所述第一原始车道组与所述第二原始车道组 达到车道对齐；
    可选地，所述车道重组还包括：
    在所述原始车道组数据中为所述虚拟车道配置车道特征，得到所述车道组数据。
  5. 根据权利要求4所述的方法，其中，所述当所述第一车道数量小于所述第二车道数量时，在所述第一原始车道组中添加虚拟车道包括：
    当所述第一车道数量小于所述第二车道数量、且所述第二原始车道组内的边缘车道路段与所述第一原始车道组内的边缘车道路段不在相同车道时，则在所述第一原始车道组中，添加与所述第二原始车道组的边缘车道路段对齐的虚拟车道。
  6. 根据权利要求5所述的方法，其中，所述在所述原始车道组数据中为所述虚拟车道配置车道特征包括：
    在所述原始车道组数据中为所述虚拟车道配置第一虚拟车道宽度和第一虚拟车道线；
    其中，所述第一虚拟车道线是表征禁止变道的属于实线型的车道线，所述第一虚拟车道宽度等于目标宽度值。
  7. 根据权利要求4所述的方法，其中，所述当所述第一车道数量小于所述第二车道数量时，在所述第一原始车道组中添加虚拟车道包括：
    当所述第一车道数量小于所述第二车道数量、且所述第一原始车道组内的中间车道路段分流为所述第二原始车道组内的至少两个车道路段时，则将所述第一原始车道组内的中间车道路段，转换为与所述第二原始车道组内的至少两个车道路段分别对齐的至少两个虚拟车道。
  8. 根据权利要求7所述的方法，其中，所述在所述原始车道组数据中为所述虚拟车道配置车道特征包括：
    在所述至少两个虚拟车道之间配置虚线型的第二虚拟车道线；
    在所述原始车道组数据中为所述至少两个虚拟车道分别配置第二虚拟车道宽度；
    其中，所述至少两个虚拟车道的第二虚拟车道宽度之间的和值与所述中间车道路段的车道宽度一致。
  9. 根据权利要求1至8中的任一项所述的方法，其中，所述车道组数据是对所述行车道路的原始车道组数据进行车道重组所得的，所述原始车道组数据是所述行车道路中各原始车道组的数据；
    所述对所述行车道路的原始车道组数据进行车道重组包括：
    获取所述行车道路的原始车道组数据；
    当所述原始车道组数据中的原始车道组是存在曲率的车道组时，将所述原始车道组数据中的原始车道组转换为直线型的车道组。
  10. 根据权利要求1至9中的任一项所述的方法，其中，所述将满足第一预设条件的特征子集从对应的所述特征集去除掉包括：
    在各所述特征集内，查找满足第一预设条件的目标特征子集；
    在各所述特征集中将所述目标特征子集去除掉，或将所述目标特征子集内的线段偏移量和车道线类型去除掉。
  11. 根据权利要求1至9任一项所述的方法，其中，所述基于所述当前位置偏移量、所述目标位置偏移量、所述剩余特征集和所述车辆的行驶信息，预测所述车辆从所述第一车道变道至所述第二车道的成功率包括：
    依据所述车辆的行驶信息确定所述车辆的行驶距离；
    根据所述行驶距离和所述当前位置偏移量，确定所述车辆的前行位置偏移量；
    依据所述前行位置偏移量、所述剩余特征集中各车道线段的线段偏移量和所述目标位置偏移量，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
  12. 根据权利要求11所述的方法，其中，所述车辆的行驶信息包括相对于车道的平行速度和垂直速度，所述剩余特征集包括所述第一车道与所述第二车道之间的特征集子序列；
    所述依据所述车辆的行驶信息确定所述车辆的行驶距离包括：
    依据所述垂直速度和车道宽度，确定跨越所述车道组内各车道的耗时；
    基于所述耗时和所述平行速度，确定所述车辆的行驶距离；
    所述依据所述前行位置偏移量、所述剩余特征集中各车道线段的线段偏移量和所述目标位置偏移量，预测所述车辆从所述第一车道变道至所述第二车道的成功率包括：
    依据所述前行位置偏移量、所述特征集子序列中各车道线段的线段偏移量和所述目标位置偏移量，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
  13. 根据权利要求11所述的方法，其中，所述剩余特征集包括所述第一车道与所述第二车道之间的特征集子序列；所述线段偏移量包括末端偏移量，所述特征集子序列的车道线数量为n，n为不小于1的正整数；
    所述依据所述前行位置偏移量、所述剩余特征集中各车道线段的线段偏移量和所述目标位置偏移量，确定所述车辆是否能从所述第一车道变道至所述第二车道包括：
    在所述特征集子序列的第i个车道线中，沿所述车辆的行驶方向遍历各车道线段的特征子集，得到遍历结果；
    确定所述遍历结果中存在满足第二预设条件的目标特征子集；所述第二预设条件为所述前行位置偏移量不大于所述第i个车道线的末端偏移量、且不大于所述目标位置偏移量；
    按照预设步长对i进行自增或自减处理，循环执行上述步骤，直至当i＝n或i＝1，且第i个车道线的遍历结果中存在满足所述第二预设条件的目标特征子集时，确定所述车辆能从所述第一车道变道至所述第二车道。
  14. 根据权利要求13所述的方法，其中，所述方法还包括：
    当确定所述遍历结果中不存在满足所述第二预设条件的目标特征子集时，确定所述车辆无法从所述第一车道变道至所述第二车道。
  15. 根据权利要求1至14中的任一项所述的方法，其中，所述方法还包括：
    当所述车辆能从所述第一车道至所述第二车道时，在电子地图上显示变道指引，所述变道指引用于指示所述车辆按照所述变道指引进行变道。
  16. 根据权利要求1至15中的任一项所述的方法，其中，所述方法还包括：
    当所述车辆无法从所述第一车道至所述第二车道时，重新规划通向所述目标位置的路径；
    发出新路径提示，所述新路径提示用于指示所述车辆按照重新规划的所述路径进行行驶。
  17. 一种车辆变道成功率的预测装置，所述车辆处于行车道路中，所述行车道路包含多个车道组，每个车道组包括多条车道的车道路段，其中，所述装置包括：
    提取模块，用于获取所述多个车道组的车道组数据；从所述多个车道组的车道组数据确定所述行车道路的多条车道线分别对应的特征集，所述特征集分别包括对应于所述多个车道组的多个特征子集，每个所述特征子集包括对应的车道线类型和线段偏移量；
    处理模块，用于将所述多个特征子集中满足第一预设条件的特征子集从其所对应的所述特征集去除掉，得到剩余特征集；
    确定模块，用于在所述车辆行驶的过程中，确定所述车辆的行车位置在所述多个车道组中的当前位置偏移量，所述行车位置处于所述多条车道中的第一车道；确定所述车辆的预设目标位置在所述多个车道组中的目标位置偏移量，所述预设目标位置处于所述多条车道中不同于所述第一车道的第二车道；
    判定模块，用于基于所述当前位置偏移量、所述目标位置偏移量、所述剩余特征集和所述车辆的行驶信息，预测所述车辆从所述第一车道变道至所述第二车道的成功率。
  18. 一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，其中，所述处理器执行所述计算机程序时实现权利要求1至16中任一项所述的方法的步骤。
  19. 一种计算机可读存储介质，其上存储有计算机程序，其中，所述计算机程序被处理器执行时实现权利要求1至16中任一项所述的方法的步骤。
  20. 一种计算机程序产品，包括计算机程序，其中，该计算机程序被处理器执行时实现权利要求1至16中任一项所述的方法的步骤。
PCT/CN2023/134641 2022-11-30 2023-11-28 车辆变道成功率的预测方法、装置、计算机设备和存储介质 WO2024114620A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211514504.6 2022-11-30
CN202211514504.6A CN116977953A (zh) 2022-11-30 2022-11-30 车道的通行性判定方法、装置、计算机设备和存储介质

Publications (1)

Publication Number Publication Date
WO2024114620A1 true WO2024114620A1 (zh) 2024-06-06

Family

ID=88473702

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/134641 WO2024114620A1 (zh) 2022-11-30 2023-11-28 车辆变道成功率的预测方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN116977953A (zh)
WO (1) WO2024114620A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116977953A (zh) * 2022-11-30 2023-10-31 腾讯科技（深圳）有限公司 车道的通行性判定方法、装置、计算机设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180190122A1 (en) * 2016-12-30 2018-07-05 Stmicroelectronics S.R.L. Method and system for generating a lane departure warning in a vehicle
CN110103969A (zh) * 2019-05-16 2019-08-09 广州小鹏汽车科技有限公司 一种车辆控制方法、装置、系统及车辆
CN115158319A (zh) * 2022-07-22 2022-10-11 阿波罗智能技术(北京)有限公司 车辆变道方法、装置、电子设备和存储介质
CN116977953A (zh) * 2022-11-30 2023-10-31 腾讯科技（深圳）有限公司 车道的通行性判定方法、装置、计算机设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180190122A1 (en) * 2016-12-30 2018-07-05 Stmicroelectronics S.R.L. Method and system for generating a lane departure warning in a vehicle
CN110103969A (zh) * 2019-05-16 2019-08-09 广州小鹏汽车科技有限公司 一种车辆控制方法、装置、系统及车辆
CN115158319A (zh) * 2022-07-22 2022-10-11 阿波罗智能技术(北京)有限公司 车辆变道方法、装置、电子设备和存储介质
CN116977953A (zh) * 2022-11-30 2023-10-31 腾讯科技（深圳）有限公司 车道的通行性判定方法、装置、计算机设备和存储介质

Also Published As

Publication number Publication date
CN116977953A (zh) 2023-10-31

Similar Documents

Publication Publication Date Title
US9558660B1 (en) Method and apparatus for providing state classification for a travel segment with multi-modal speed profiles
WO2020034903A1 (zh) 一种基于拓扑图的智能导航的方法和系统
WO2024114620A1 (zh) 车辆变道成功率的预测方法、装置、计算机设备和存储介质
US11755553B2 (en) Traffic-aware route decoding using a probabilistic encoding data structure
US10030982B2 (en) Generalising topographical map data
CN113779430B (zh) 路网数据生成方法、装置、计算设备和存储介质
CN108053679A (zh) 一种基于众包数据的停车位引导方法及系统
CN116935656B (zh) 道路交通数据处理方法、装置、电子设备及存储介质
CN113177046B (zh) 路网拓扑图的生成方法、装置、设备及存储介质
Yuan et al. A map-matching algorithm with low-frequency floating car data based on matching path
CN111383444A (zh) 预测路况状态的方法、装置、服务器及存储介质
CN107588779B (zh) 一种基于任意两节点间行程时间的车辆智能导航方法
WO2023173831A1 (zh) 引导点的关联信息的更新方法和装置
CN112699202B (zh) 禁行道路的识别方法、装置、电子设备及存储介质
CN116226309A (zh) 地图资料处理方法、装置、电子设备及存储介质
EP4012336A2 (en) Method, apparatus, and system for identifying special areas and cleaning-up map data
CN115100231A (zh) 一种区域边界的确定方法和装置
CN116465394B (zh) 基于车辆轨迹数据的路网结构生成方法及装置
CN116562488B (zh) 导流岛生成方法、装置、计算机设备、介质和程序产品
CN116932929A (zh) 目标路段筛选方法、装置、计算机设备和存储介质
CN115493609A (zh) 车道级路径信息生成方法、装置、设备、介质和程序产品
CN118096934A (zh) 车道面生成方法、装置、设备及介质
CN117091620A (zh) 导航方法、装置、计算机设备和计算机可读存储介质
Jensen ‘Data-intensive routing in spatial networks
Liu et al. A Context-Aware Map Matching Method Based on Supported Degree