CN117698752A - Target screening method and device - Google Patents

Target screening method and device Download PDF

Info

Publication number
CN117698752A
CN117698752A CN202410162172.2A CN202410162172A CN117698752A CN 117698752 A CN117698752 A CN 117698752A CN 202410162172 A CN202410162172 A CN 202410162172A CN 117698752 A CN117698752 A CN 117698752A
Authority
CN
China
Prior art keywords
vehicle
target
determining
candidate
targets
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202410162172.2A
Other languages
Chinese (zh)
Inventor
胡汇泽
杨唐涛
王邓江
殷旭梁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Wanji Iov Technology Co ltd
Original Assignee
Suzhou Wanji Iov Technology Co ltd
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 Suzhou Wanji Iov Technology Co ltd filed Critical Suzhou Wanji Iov Technology Co ltd
Priority to CN202410162172.2A priority Critical patent/CN117698752A/en
Publication of CN117698752A publication Critical patent/CN117698752A/en
Pending legal-status Critical Current

Links

Abstract

The application relates to a target screening method and device, wherein the method comprises the following steps: acquiring vehicle state information acquired by a vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located, determining a first candidate target in a preset interaction area of the vehicle and a second candidate target which is intersected with the vehicle according to the vehicle state information and the target state information, and determining a dangerous target of the vehicle from the targets according to the first candidate target and the second candidate target. The method improves the accuracy of screening dangerous targets in the intersections.

Description

Target screening method and device
Technical Field
The application relates to the technical field of intelligent driving, in particular to a target screening method and device.
Background
With the continuous development of intelligent driving technology, intelligent driving vehicles can realize higher-level automatic driving in various scenes.
Under the scene of higher requirement on the perception capability of the intelligent driving vehicle, the perception capability of the intelligent driving vehicle can be improved through a road side perception technology; for example, an intersection scene. However, because the working condition of the intersection is complex, the targets in the intersection area are more, and in order to reduce the receiving of the intelligent network-connected vehicles on the invalid targets, dangerous targets in the intersection can be screened to filter irrelevant targets.
However, the screening method for dangerous targets in the road mouth in the related art has the problem of low accuracy.
Disclosure of Invention
Based on the above, it is necessary to provide a target screening method and device, which can improve the accuracy of screening dangerous targets at intersections.
In a first aspect, the present application provides a target screening method, including:
acquiring vehicle state information acquired by a vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located;
determining a first candidate target in a preset interaction area of the vehicle and a second candidate target which is intersected with the vehicle according to the vehicle state information and the target state information;
and determining dangerous targets of the vehicle from the targets according to the first candidate target and the second candidate target.
In one embodiment, the vehicle status information includes a vehicle location; determining a first candidate target in a preset interaction area of the vehicle according to the vehicle state information and each target state information, wherein the method comprises the following steps:
determining a vehicle lane in which the vehicle is positioned according to the vehicle position;
determining a preset interaction area of the vehicle according to the lane of the vehicle;
And according to the state information of each target, determining the target in the preset interaction area in at least one target as a first candidate target.
In one embodiment, determining a vehicle lane in which a vehicle is located based on a vehicle location includes:
determining whether the vehicle is in a lane at the current moment according to the vehicle position;
if the vehicle is in the lane at the current moment, determining the lane in which the vehicle is positioned at the current moment as a vehicle lane;
if the vehicle is not in the lane at the current time, determining the vehicle history nearest lane of the vehicle as the vehicle lane.
In one embodiment, the preset interaction area comprises a first interaction area of a vehicle and a pedestrian, a second interaction area of the vehicle and a non-motor vehicle, and a third interaction area of the vehicle and the motor vehicle; according to the state information of each target, determining the target in the preset interaction area in at least one target as a first candidate target, wherein the method comprises the following steps:
according to the state information of each target, determining pedestrians, non-motor vehicles and motor vehicles in at least one target;
determining a pedestrian in a first interaction area as a first candidate target;
determining a non-motor vehicle in the second interaction region as a first candidate target;
And under the condition that the target is the motor vehicle, determining a first candidate target according to the third interaction area and the target lane where the motor vehicle is located.
In one embodiment, determining the first candidate object based on the third interaction region and the object lane in which the motor vehicle is located includes:
obtaining at least one candidate vehicle in a third interaction area from each vehicle;
for any candidate motor vehicle, if the candidate motor vehicle has a target lane, determining the candidate motor vehicle as a first candidate target;
and if the candidate motor vehicle does not have the target lane, determining the candidate motor vehicle with the distance from the vehicle being smaller than the preset motor vehicle distance threshold value as a first candidate target.
In one embodiment, the vehicle status information includes a vehicle location; determining a first candidate target in a preset interaction area of the vehicle according to the vehicle state information and each target state information, wherein the method comprises the following steps:
determining a preset interaction area of the vehicle according to the vehicle position and a preset distance threshold;
and determining the target in the preset interaction area in at least one target as a first candidate target according to the target state information of each target.
In one embodiment, the preset interaction region of the vehicle comprises a fourth interaction region of the vehicle and the pedestrian, and a fifth interaction region of the vehicle and the non-motor vehicle; the distance threshold includes a pedestrian distance threshold and a non-motor vehicle distance threshold; determining a preset interaction area of the vehicle according to the vehicle position and a preset distance threshold value comprises the following steps:
determining a fourth interaction area according to the vehicle position and the pedestrian distance threshold;
a fifth interaction zone is determined based on the vehicle location and the non-motor vehicle distance threshold.
In one embodiment, the vehicle status information includes a vehicle location; the target state information includes a target position; determining a second candidate object which is intersected with the vehicle according to the vehicle state information and the object state information, wherein the second candidate object comprises:
predicting vehicle running track information of the vehicle at the intersection according to the vehicle position and the vehicle lane of the vehicle;
predicting target driving track information of each target at the intersection according to the target position and the target lane of each target;
and determining a second candidate target according to the vehicle running track information and the target running track information.
In one embodiment, predicting vehicle travel track information of a vehicle at an intersection according to a vehicle position and a vehicle lane of the vehicle includes:
Determining a starting point of a vehicle entering the intersection and an ending point of the vehicle leaving the intersection according to the vehicle lanes and the vehicle positions;
and determining the vehicle running track information of the vehicle at the intersection according to the starting point and the ending point of the vehicle.
In one embodiment, predicting target driving track information of each target at an intersection according to a target position and a target lane of each target includes:
aiming at any one target, determining a starting point of the target entering the intersection and an ending point of the target leaving the intersection according to the target position and the target lane;
and determining target running track information of the target at the intersection according to the starting point and the ending point of the target.
In one embodiment, determining the second candidate target according to the vehicle driving track information and the target driving track information includes:
determining a reference target which is intersected with the vehicle and an intersection position of the vehicle and the reference target according to the vehicle running track information and the target running track information;
determining the vehicle time length for the vehicle to reach the intersection position according to the intersection position and the vehicle position;
determining the target time length of the reference target reaching the intersection position according to the intersection position and the target position of the reference target;
and determining a second candidate target according to the vehicle duration and the target duration.
In one embodiment, determining a second candidate target based on the vehicle duration and the target duration includes:
determining the intersection time difference between the vehicle and the reference target according to the vehicle duration and the target duration;
and determining a reference target with the absolute value of the intersection time difference smaller than or equal to a preset duration threshold as a second candidate target.
In one embodiment, determining a dangerous target for the vehicle from the targets based on the first candidate target and the second candidate target includes:
removing repeated targets in the first candidate target and the second candidate target to obtain total candidate targets;
if the number of the total candidate targets is smaller than or equal to a preset number threshold, determining that the total candidate targets are dangerous targets;
if the number of the total candidate targets is larger than the number threshold, determining dangerous targets according to the relation between the number of the repeated targets and the number threshold.
In one embodiment, determining the hazard target based on a relationship between the number of duplicate targets and a number threshold includes:
acquiring intersection time differences of the repeated targets and the vehicle under the condition that the number of the repeated targets is larger than or equal to a number threshold value, and determining the repeated targets with smaller absolute values of the number threshold value and the intersection time differences as dangerous targets;
If the number of the repeated targets is smaller than the number threshold, determining the repeated targets as dangerous targets, and determining the residual number of the dangerous targets according to the number threshold and the number of the repeated targets;
a remaining number of dangerous objects is determined from the remaining total candidate objects except for the duplicate objects.
In one embodiment, the vehicle status information includes a vehicle location; the target state information includes a target position and a target speed; determining a remaining number of dangerous objects from the remaining total candidate objects except for the duplicate object, comprising:
determining the real distance between the vehicle and the rest total candidate targets according to the vehicle position and the target positions of the rest total candidate targets;
determining target distances of the other total candidate targets according to a preset time threshold and target speeds of the other total candidate targets;
determining the risk evaluation value of the rest total candidate targets according to the real distance, the target distance, the preset speed weight and the distance weight;
and determining the rest total candidate targets with the rest number of risk evaluation values smaller as risk targets.
In a second aspect, the present application further provides a target screening apparatus, including:
the response module is used for responding to a target screening request sent by the vehicle and acquiring vehicle state information acquired by the vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located;
The determining module is used for determining a first candidate target in a preset interaction area of the vehicle and a second candidate target which is intersected with the vehicle according to the vehicle state information and the target state information;
and the screening module is used for determining dangerous targets of the vehicle from the targets according to the first candidate target and the second candidate target.
In a third aspect, embodiments of the present application provide a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of the method provided by any of the embodiments of the first aspect described above when the computer program is executed.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method provided by any of the embodiments of the first aspect described above.
In a fifth aspect, embodiments of the present application also provide a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method provided by any of the embodiments of the first aspect described above.
According to the target screening method and device, the vehicle state information collected by the vehicle at the current moment and the target state information of at least one target collected by the road side equipment in the intersection where the vehicle is located are obtained, a first candidate target in a preset interaction area of the vehicle and a second candidate target intersected with the vehicle are determined according to the vehicle state information and the target state information, and then dangerous targets of the vehicle are determined from the targets according to the first candidate target and the second candidate target. According to the method, a preset interaction area of the vehicle is determined according to the current state information of the vehicle and the current state information of at least one target, and a first candidate target is determined from the preset interaction area, so that the problems that more targets and omission of small targets are sent to the intelligent driving vehicle are solved; and judging whether the intersection target is intersected with the vehicle or not, and determining the intersection target which is possibly intersected with the vehicle as a second candidate target, thereby solving the problems of blind area and missing of a far-end target in the intersection; and finally, fusing the first candidate target and the second candidate target to determine dangerous targets of the vehicle, thereby improving the accuracy and reliability of screening dangerous targets in the intersection.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for a person having ordinary skill in the art.
FIG. 1 is a diagram of an application environment for a target screening method in one embodiment;
FIG. 2 is a flow chart of a method of screening targets in one embodiment;
FIG. 3 is a schematic diagram of a structure of a road junction according to an embodiment;
FIG. 4 is a flow chart of a method for screening targets according to another embodiment;
FIG. 5 is a flow chart of a method for screening targets according to another embodiment;
FIG. 6 is a schematic view of a road junction according to another embodiment;
FIG. 7 is a schematic view of a road junction according to another embodiment;
FIG. 8 is a flow chart of a method for screening targets according to another embodiment;
FIG. 9 is a flow chart of a method of screening targets according to another embodiment;
FIG. 10 is a flow chart of a method of screening targets according to another embodiment;
FIG. 11 is a flow chart of a method for screening targets according to another embodiment;
FIG. 12 is a flow chart of a method of screening targets according to another embodiment;
FIG. 13 is a flow chart of a method of screening targets according to another embodiment;
FIG. 14 is a flow chart of a method of screening targets according to another embodiment;
FIG. 15 is a schematic view of a road junction according to another embodiment;
FIG. 16 is a flow chart of a method of screening targets according to another embodiment;
FIG. 17 is a flow chart of a method of screening targets according to another embodiment;
FIG. 18 is a flow chart of a method of screening targets according to another embodiment;
FIG. 19 is a flow chart of a method of screening targets according to another embodiment;
FIG. 20 is a flow chart of a method of screening targets according to another embodiment;
FIG. 21 is a flow chart of a method of screening targets according to another embodiment;
FIG. 22 is a flow chart of a method of screening targets according to another embodiment;
FIG. 23 is a block diagram showing the construction of a target screening apparatus in one embodiment;
fig. 24 is an internal structural view of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The target screening method provided by the embodiment of the application can be applied to an application environment shown in fig. 1.
As shown in fig. 1, the vehicle 101 and the roadside device 102 each perform wired or wireless communication with the roadside system 103 through a network; the Road Side system may be integrated in a Road Side Unit (RSU)/a mobile edge computer (Mobile Edge Computing, MEC) on the Road Side, or may be integrated in a computer device, or may be placed on a cloud or other network server.
The road side system may be a road side computing unit/terminal/edge server, optionally, the road side system may also be a cloud server, a vehicle-mounted computing unit/terminal of a vehicle end, and the like. The sensing device may be a road side sensing device disposed at an intersection, for example, may be an intelligent base station (also called a road side base station) at the intersection, or may also be at least one of a millimeter wave radar sensor, a laser radar sensor, a camera, and the like, and the type of the road side device is not particularly limited herein; the roadside device may include a plurality of devices.
With the continuous development of intelligent driving technology, more and more scenes are paid attention to, wherein intersections are an important part of urban working conditions.
The intelligent driving vehicle runs in the intersection, and can not well solve the problems of blind area, remote target detection, inaccurate small target detection and the like due to the limited self-perception capability (range), so that the road side perception technology is introduced into the intersection to well solve the problems.
Meanwhile, as part of large-scale intersections are abnormal and complex in road conditions and have more interaction targets, dangerous targets in the intersections are screened, a plurality of irrelevant targets can be filtered, invalid information receiving of intelligent driving vehicles is reduced, and the calculated amount of a vehicle perception decision part is further reduced.
Based on the above, the object screening method is provided, and through screening by using a plurality of filtering screening methods such as the interaction area of the vehicle and the objects which possibly have intersection, dangerous interaction objects can be screened out better, so that the accuracy of screening dangerous objects in the intersection can be improved.
In an exemplary embodiment, as shown in fig. 2, there is provided a target screening method, which is illustrated by using the method applied to the road side system in fig. 1 as an example, and includes the following steps:
s201, acquiring vehicle state information acquired by a vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located.
In the road running process of the vehicle, the state sensor on the vehicle can acquire the vehicle state information in real time and send the vehicle state information to a road side system through the vehicle-mounted unit; wherein, the vehicle state information may include information such as a vehicle position and a vehicle speed; the vehicle position represents the position of the vehicle at the current time, and the vehicle speed represents the speed of the vehicle at the current time.
The road side equipment in the road side system can acquire the target state information of all targets in the perception range in real time; the road side equipment can be a sensing equipment and can be a smart base station, a millimeter wave radar sensor or a laser radar sensor and the like. The target may be any target within the perceived area of a roadside device at an intersection, including but not limited to, a car, truck, electric vehicle, bicycle, tricycle, scooter, pedestrian, and the like.
Taking an intersection as an example, as shown in fig. 3, fig. 3 is a schematic diagram of the intersection, for example, the intersection may include an entrance road, an exit road, a crosswalk, an intersection core area, and the like, the vehicle stop line is a boundary line between the vehicle entrance road and the crosswalk, the entrance road may be an area with a preset length (50 meters) from the lane stop line, that is, a lane entering the direction of the intersection, and the exit road is an area with a preset length (50 meters) away from the vehicle stop line, that is, a lane leaving the direction of the intersection; the area surrounded by the 4 crosswalks is regarded as the intersection.
The vehicle state information can comprise a vehicle position, after receiving the vehicle state information sent by the vehicle, the road side system can detect whether the vehicle is in an entrance way or an intersection, and when the vehicle is in the entrance way or the intersection, the road side system can acquire the vehicle state information sent by the vehicle at the current moment and the target state information of at least one target sent by road side equipment of the intersection so as to screen dangerous targets of the vehicle.
The vehicle-mounted unit of the vehicle can also detect whether the vehicle is in an entrance or an intersection according to the vehicle position acquired in real time, when the vehicle is in the entrance or the intersection, the vehicle-mounted unit sends a target screening request to the road side system, and the road side system responds to the target screening request sent by the vehicle to acquire vehicle state information sent by the vehicle at the current moment and target state information of at least one target sent by road side equipment of the intersection.
Optionally, the screening of dangerous targets for the vehicle is performed in the road side system, but after receiving the vehicle state information sent by the vehicle and the target state information of at least one target collected by the road side device, the road side system does not know which target collected by the road side device is the vehicle, so that the vehicle can be matched into the road side system according to the vehicle state information, i.e. the target matched with the vehicle is found in the at least one target collected by the road side device.
S202, determining a first candidate target in a preset interaction area of the vehicle and a second candidate target which is intersected with the vehicle according to the vehicle state information and the target state information.
The vehicle state information may include a vehicle position, the target state information may include a target position, and the target position may represent a position of the target at the current time; searching an interaction area corresponding to the vehicle position from the position interaction area relation according to the vehicle position, and determining the interaction area corresponding to the vehicle position in the position interaction relation as a preset interaction area of the vehicle; the position interaction area relation comprises a plurality of corresponding relations between positions and interaction areas.
The preset interaction area is an area which is interacted with the vehicle in the intersection; for example, objects travelling in a preset interaction area are at risk of collision with the vehicle. Therefore, the targets in the preset interaction area can be determined as the first candidate targets according to the target positions of the targets.
In addition, for the targets which are not partially in the preset interaction area, the intersection probability between each target acquired by the road side equipment and the vehicle can be calculated, and the target with the intersection probability greater than the preset probability threshold value with the vehicle is determined as the second candidate target. Wherein the presence of an intersection between the object and the vehicle may indicate the presence of a collision between the object and the vehicle.
For example, the vehicle state information includes a vehicle position, a vehicle speed, and a vehicle heading angle, the target state information includes a target position, a target speed, and a target heading angle, the target speed represents a speed of the target at a current time, and the target heading angle represents a heading angle of the target at the current time; for any one of the targets, the vehicle position, the vehicle speed and the vehicle course angle, and the target position, the target speed and the target course angle can be input into a pre-trained intersection model, the running track of the vehicle and the target is analyzed through the intersection model, the intersection probability between the target and the vehicle is determined, and the target with the intersection probability greater than the preset probability threshold value is determined as a second candidate target.
S203, determining dangerous targets of the vehicle from the targets according to the first candidate target and the second candidate target.
Dangerous objects of the vehicle may represent objects of the vehicle with a greater risk of collision; both the first candidate object and the second candidate object may be determined as dangerous objects of the vehicle.
In this embodiment, dangerous targets (pedestrians, non-vehicles and vehicles) having interaction risks with the vehicles are screened out in real time according to the full-flow track of the vehicles at the intersection.
In the target screening method provided by the embodiment of the application, the vehicle state information acquired by the vehicle at the current moment and the target state information of at least one target acquired by the road side equipment in the intersection where the vehicle is located are acquired, a first candidate target in a preset interaction area of the vehicle and a second candidate target intersected with the vehicle are determined according to the vehicle state information and the target state information, and then dangerous targets of the vehicle are determined from the targets according to the first candidate target and the second candidate target. According to the method, a preset interaction area of the vehicle is determined according to the current state information of the vehicle and the current state information of at least one target, and a first candidate target is determined from the preset interaction area, so that the problems that more targets and omission of small targets are sent to the intelligent driving vehicle are solved; and judging whether the intersection target is intersected with the vehicle or not, and determining the intersection target which is possibly intersected with the vehicle as a second candidate target, thereby solving the problems of blind area and missing of a far-end target in the intersection; and finally, fusing the first candidate target and the second candidate target to determine dangerous targets of the vehicle, thereby improving the accuracy and reliability of screening dangerous targets in the intersection.
In one exemplary embodiment, the vehicle status information includes a vehicle location; as shown in fig. 4, determining a first candidate object in a preset interaction area of a vehicle according to vehicle state information and object state information includes the steps of:
s401, determining a vehicle lane where the vehicle is located according to the vehicle position.
In one exemplary embodiment, as shown in FIG. 5, determining a vehicle lane in which a vehicle is located based on a vehicle location, comprises the steps of:
s501, determining whether the vehicle is in a lane at the current moment according to the vehicle position.
The high-precision map comprises a plurality of positions and lane numbers corresponding to the positions; if the position is not in the lane, the lane number is-1.
Therefore, the vehicle position can be matched with the position in the high-precision map, and the lane number corresponding to the vehicle position is determined as the lane number of the vehicle; it may then be determined whether the vehicle is in the lane at the current time based on the lane number.
S502, if the vehicle is in the lane at the current time, determining the lane in which the vehicle is in at the current time as the vehicle lane.
If the vehicle is in the lane at the current moment, determining the lane in which the vehicle is in at the current moment as the vehicle lane.
S503, if the vehicle is not in the lane at the current time, determining the vehicle history nearest lane of the vehicle as the vehicle lane.
If the vehicle is not in the lane at the current moment, the historical lane information of the vehicle can be acquired, wherein the historical lane information comprises at least one lane through which the vehicle runs; a vehicle history nearest lane in the history lane information may be determined as a vehicle lane; wherein the vehicle history nearest lane indicates the lane in which the vehicle has recently traveled.
If the vehicle is not in the lane at the current time, determining the history nearest vehicle lane as the vehicle lane in which the vehicle is currently located if the history nearest vehicle lane is only one lane; if the history recent lane of the vehicle includes a plurality of lanes, the plurality of lanes of the history recent vehicle lane of the vehicle may be all taken as the vehicle lanes of the vehicle.
If the history nearest lane of the vehicle includes a plurality of lanes, the vehicle lane of the vehicle may be determined according to the distance between the current vehicle position of the vehicle and the lane stop line of the history nearest vehicle lane and the second distance between the current vehicle position of the vehicle and the non-motor vehicle lane stop line on the left or right side of the same level with the history nearest vehicle lane.
For example, as shown in fig. 6, if the vehicle position of the vehicle is at the point a in the intersection, the history nearest lane of the vehicle is the south-entrance lane 2, and the south-entrance lane 2 is the straight-going lane and the right-turning lane, a first distance from the point a to the point A1 on the lane stop line of the south-entrance lane 2 and a second distance from the point a to the point A2 on the non-motor vehicle lane stop line on the right side of the south-entrance lane 2 may be obtained, and if the first distance is smaller than the second distance, it is determined that the vehicle lane is the straight-going lane, otherwise, the vehicle lane is the right-turning lane.
If the vehicle position of the vehicle is at the H point in the intersection, the history nearest lane of the vehicle is the south entrance lane 1, and the south entrance lane 1 is the straight lane and the left-turn lane, a first distance from the H point to the H1 point on the lane stop line of the south entrance lane 1 and a second distance from the H point to the H2 point of the non-motor vehicle lane stop line on the left side of the south entrance lane 1 can be obtained, if the first distance is smaller than the second distance, the vehicle lane is determined to be the straight lane, otherwise, the vehicle lane is determined to be the left-turn lane.
In the embodiment of the application, according to the vehicle position, whether the vehicle is in the lane at the current moment is determined, if the vehicle is in the lane at the current moment, the lane in which the vehicle is located at the current moment is determined as the vehicle lane, and if the vehicle is not in the lane at the current moment, the vehicle history nearest lane of the vehicle is determined as the vehicle lane. According to the method, the vehicle lane where the target vehicle is currently located is determined through the vehicle position, and the situation that the vehicle is not in the lane at the intersection is considered, so that the accuracy of determining the vehicle lane is improved.
S402, determining a preset interaction area of the vehicle according to the lane of the vehicle.
And storing the corresponding relation between the plurality of vehicle lanes and the interactive areas in the road side system, acquiring the interactive areas corresponding to the vehicle lanes from the corresponding relation between the plurality of vehicle lanes and the interactive areas, and determining the interactive areas corresponding to the vehicle lanes as preset interactive areas of the vehicle.
Taking the right turn of the vehicle in the south entrance roadway 2 as an example, as shown in fig. 7, fig. 7 shows a preset interaction area corresponding to the right turn of the vehicle, the preset interaction area of the vehicle may include a non-motor vehicle straight-going interaction area 1 of a left side branch, a non-motor vehicle left-hand-over interaction area 2 of a opposite side branch, and pedestrian of the present branch corresponding to the zebra crossing interaction area 3, the motor vehicle straight-going interaction area 4 and the motor vehicle left-hand-over interaction area 5.
Optionally, when the vehicle is in a right lane, the preset interaction area of the vehicle may further include a turning scene of the motor vehicle on the right branch, which is not illustrated in fig. 7.
If the vehicle is in the straight-going road, the preset interaction region of the vehicle may include a right-hand interaction region of the vehicle of the right-hand branch, a non-vehicle straight-going (red light running) region of the left-hand branch, and a non-vehicle left-hand (red light running) interaction region of the opposite-hand branch.
If the vehicle is in a left turn, the predetermined interaction zone of the vehicle may include a right turn of the motor vehicle facing the leg.
If the vehicle lane includes two or more directions (straight, right, and left), it is necessary to consider all the preset interaction areas in which the vehicle can travel.
For example, if the vehicle lane includes a straight lane and a right turn lane, the preset interaction region corresponding to the straight lane and the preset interaction region corresponding to the right turn lane may be determined as the preset interaction region of the vehicle.
S403, determining a target in a preset interaction area in at least one target as a first candidate target according to the state information of each target.
The preset interaction area comprises a first interaction area of the vehicle and the pedestrian, a second interaction area of the vehicle and the non-motor vehicle and a third interaction area of the vehicle and the motor vehicle; the preset interaction area where the vehicle and the pedestrian are likely to interact can be determined to be a first interaction area, the preset interaction area where the vehicle and the non-motor vehicle are likely to interact can be determined to be a second interaction area, and the preset interaction area where the vehicle and the motor vehicle are likely to interact can be determined to be a third interaction area.
With continued reference to fig. 7, the pedestrian of the present branch in fig. 5 may be a first interaction area in the corresponding zebra crossing interaction area 3, the non-motor vehicle straight-going interaction area 1 of the left branch and the non-motor vehicle left-hand interaction area 2 of the opposite branch may be a second interaction area, and the motor vehicle straight-going interaction area 4 and the motor vehicle left-hand interaction area 5 may be a third interaction area.
In an exemplary embodiment, as shown in fig. 8, according to each object state information, an object in a preset interaction area in at least one object is determined as a first candidate object, which includes the following steps:
s801, determining pedestrians, non-motor vehicles and motor vehicles in at least one target according to the state information of each target.
The target state information can comprise a target image, the target image can be detected, and the target type of the target can be determined; among the types of targets are pedestrians, non-vehicles, and vehicles.
S802, determining pedestrians in a first interaction area as a first candidate target, determining non-motor vehicles in a second interaction area as the first candidate target, and determining the first candidate target according to a third interaction area and a target lane where the motor vehicles are located when the targets are motor vehicles.
According to the pedestrian position of the pedestrian, determining the pedestrian with the pedestrian position in the first interaction area as a first candidate target; and determining the non-motor vehicle with the non-motor vehicle position in the second interaction area as a first candidate target according to the non-motor vehicle position of the non-motor vehicle.
For a motor vehicle, a motor vehicle whose position is in a third interaction region may also be determined as a first candidate target according to the target lane in which the motor vehicle is located.
According to the embodiment of the application, according to the state information of each target, determining pedestrians, non-motor vehicles and motor vehicles in at least one target, determining pedestrians in a first interaction area as a first candidate target, and determining non-motor vehicles in a second interaction area as the first candidate target; and determining a first candidate target according to the third interaction area and a target lane where the motor vehicle is located in the case that the target is the motor vehicle. In the method, because the constraints of pedestrians, non-motor vehicles and motor vehicles are different, different interaction areas can be determined according to different constraint conditions, and the first candidate target is determined according to the corresponding interaction areas, so that the reliability and the accuracy of the first candidate target are improved.
According to the target screening method, a vehicle lane where a vehicle is located is determined according to the position of the vehicle, a preset interaction area of the vehicle is determined according to the vehicle lane, and then a target in the preset interaction area in at least one target is determined to be a first candidate target according to state information of each target. According to the method, the whole flow path of the vehicle at the intersection can be predicted through the lane of the vehicle, so that the preset interaction area (dangerous area) of the vehicle is determined, the first candidate targets possibly having interaction risks for the vehicle are screened out in real time, the problem that more or less targets are sent to the intelligent driving vehicle is solved, and the accuracy of target screening is improved.
In the above embodiment it was proposed to determine the first candidate object from the third interaction zone and the object lane in which the motor vehicle is located, which will be described in more detail below by means of an embodiment. In one exemplary embodiment, as shown in fig. 9, determining a first candidate target according to a third interaction region and a target lane in which the motor vehicle is located, includes the steps of:
at least one candidate vehicle in a third interaction area is acquired from each vehicle S901.
At least one vehicle in a third interaction zone is acquired based on the vehicle position of each vehicle, and the at least one vehicle in the third interaction zone is determined as at least one candidate vehicle.
S902, for any one of the candidate vehicles, if the candidate vehicle has a target lane, the candidate vehicle is determined as a first candidate target.
For any one of the candidate vehicles, it may be determined first whether the candidate vehicle has a target lane, and a first candidate target may be determined from among the candidate vehicles. The lane in which the candidate vehicle is currently located may be determined as the target lane of the candidate vehicle, or if the candidate vehicle is not currently located in the lane, the historical nearest target lane of the candidate vehicle may be determined as the target lane of the candidate vehicle; if the candidate vehicle is not currently in the lane and the candidate vehicle does not have the historical closest target lane, determining that the candidate vehicle does not have the target lane.
If the candidate vehicle has a target lane, the candidate vehicle is determined as a first candidate target, i.e., the candidate vehicle having the target lane is determined as the first candidate target.
And S903, if the candidate motor vehicle does not have a target lane, determining the candidate motor vehicle with the distance from the vehicle being smaller than a preset motor vehicle distance threshold value as a first candidate target.
If a candidate vehicle does not have a target lane, indicating that the candidate vehicle was not previously collected by the roadside apparatus, the candidate vehicle may be a sudden vehicle.
The distance between the candidate motor vehicle without the target lane and the vehicle can be calculated, and the candidate motor vehicle with the distance smaller than the preset motor vehicle distance threshold value is determined as a first candidate target; wherein the motor vehicle distance threshold may be 12 meters.
In the target screening method provided by the embodiment of the application, at least one candidate motor vehicle in a third interaction area is obtained from each motor vehicle; for any candidate motor vehicle, if the candidate motor vehicle has a target lane, determining the candidate motor vehicle as a first candidate target; and if the candidate motor vehicle does not have the target lane, determining the candidate motor vehicle with the distance from the vehicle being smaller than the preset motor vehicle distance threshold value as a first candidate target. According to the method, whether the motor vehicles in the third interaction area exist or not is judged according to whether the motor vehicles in the third interaction area exist the lane information, whether the motor vehicles in the third interaction area are first candidate areas or not is determined according to the situation, all the motor vehicles in the third interaction area do not need to be used as second candidate targets, subsequent calculated amount is reduced, and accuracy of the first candidate targets is improved.
In addition to the above-described determination of the preset interaction zone from the vehicle lane, the preset interaction zone of the vehicle may also be determined by distance, which is described in detail below by way of one embodiment. In one exemplary embodiment, the vehicle status information includes a vehicle location; as shown in fig. 10, determining a first candidate object in a preset interaction area of a vehicle according to vehicle state information and object state information includes the steps of:
s1001, determining a preset interaction area of the vehicle according to the vehicle position and a preset distance threshold.
A circular area may be defined with the vehicle location as a center and the distance threshold as a radius, the circular area being determined as a preset interaction area of the vehicle.
Since the targets include pedestrians, non-vehicles and vehicles, and the first candidate target for distance screening of vehicles has been described in the above embodiments, only the pedestrians and the non-vehicles are described in the embodiments of the present application.
Since the pedestrian and the non-motor vehicle are different types of targets, the distance threshold may include a pedestrian distance threshold and a non-motor vehicle distance threshold, and then the preset interaction region of the vehicle should also include a fourth interaction region of the vehicle and the pedestrian and a fifth interaction region of the vehicle and the non-motor vehicle; then, in an exemplary embodiment, as shown in fig. 11, determining a preset interaction region of the vehicle according to the vehicle position and a preset distance threshold value includes the following steps:
S1101, determining a fourth interaction area according to the vehicle position and the pedestrian distance threshold.
A circular area is defined with the vehicle position as the center and the pedestrian distance threshold as the radius, and the circular area is determined as a fourth interaction area.
S1102, determining a fifth interaction area according to the vehicle position and the non-motor vehicle distance threshold.
A circular area is defined with the vehicle position as the center and the non-motor vehicle distance threshold as the radius, and the circular area is determined as a fifth interaction area.
In the embodiment of the application, the fourth interaction area is determined according to the vehicle position and the pedestrian distance threshold, and the fifth interaction area is determined according to the vehicle position and the non-motor vehicle distance threshold. And determining a fourth interaction area where the vehicle interacts with the pedestrian and a fifth interaction area where the vehicle interacts with the pedestrian according to the vehicle position, so that pedestrians and non-motor vehicles which are close to the vehicle can be screened out, and the accuracy of target screening is improved.
S1002, determining a target in a preset interaction area in at least one target as a first candidate target according to target state information of each target.
The target in the preset interaction area is determined as the first candidate target, and in the above embodiment, the preset interaction area is divided into the fourth interaction area of the vehicle and the pedestrian and the fifth interaction area of the vehicle and the non-vehicle, so that the first candidate target can be screened according to the interaction areas of different target types. In an exemplary embodiment, determining, according to the object state information of each object, an object in the preset interaction area in at least one object as a first candidate object includes: determining pedestrians and non-motor vehicles in at least one target according to the state information of each target; a pedestrian in the fourth interaction region is determined as a first candidate target, and a non-motor vehicle in the fifth interaction region is determined as a first candidate target.
The manner of determining pedestrians and non-vehicles in at least one target in the embodiment of the present application is the same as the manner of determining pedestrians and non-vehicles in the above embodiment, and the embodiment of the present application is not repeated here.
A pedestrian in the fourth interaction region is determined as a first candidate based on the pedestrian position of the pedestrian, i.e., it is indicated that a pedestrian having a distance to the vehicle less than the pedestrian distance threshold is determined as the first candidate.
The non-motor vehicle in the fifth interaction region is determined as the first candidate object according to the non-motor vehicle position of the non-motor vehicle, namely, the non-motor vehicle with the distance from the non-motor vehicle smaller than the non-motor vehicle distance threshold value is determined as the first candidate object.
The distance threshold value of the pedestrians is smaller than the distance threshold value of the non-motor vehicles, and the distance threshold value of the non-motor vehicles is smaller than the distance threshold value of the motor vehicles because the restriction of the pedestrians is strict; the pedestrian distance threshold may be 8 meters and the non-motor vehicle distance threshold may be 10 meters.
In the embodiment of the application, according to the state information of each target, the pedestrians and the non-motor vehicles in at least one target are determined, the pedestrians in the fourth interaction area are determined to be first candidate targets, and the non-motor vehicles in the fifth interaction area are determined to be first candidate targets. In the method, the fourth interaction area and the fifth interaction area are areas which are closer to the vehicle, so that pedestrians and non-motor vehicles in the fourth interaction area and the fifth interaction area are screened out respectively, the accuracy of target screening can be improved, and omission is avoided.
In the target screening method provided by the embodiment of the application, a preset interaction area of a vehicle is determined according to the position of the vehicle and a preset distance threshold, and a target in the preset interaction area in at least one target is determined to be a first candidate target according to target state information of each target. In the method, the preset interaction area is defined by the distance between the dangerous target and the vehicle, the first candidate target is determined from the preset interaction area, and the dangerous target is screened by the distance, so that the accuracy of target screening is improved.
It should be noted that, the first candidate target screened out according to the preset interaction area determined by the lane information of the vehicle and the first candidate target screened out according to the position of the vehicle and the lane information may have repeated targets, and the road side system may delete the repeated targets to obtain the first candidate target without the repeated targets; for example, the first candidate targets screened out according to the preset interaction area determined by the lane information of the vehicle include an A target and a B target, and the first candidate targets screened out according to the position of the vehicle and the lane information include a B target and a C target, one of the B targets is deleted, and the A target, the B target and the C target are used as the first candidate targets.
The above embodiments are all descriptions of how to determine the first candidate object, and how to determine the second candidate object is described below by way of one embodiment, in which the vehicle state information includes the vehicle position; the target state information includes a target position; as shown in fig. 12, determining a second candidate object that meets the vehicle based on the vehicle state information and the respective object state information, includes the steps of:
s1201 predicts vehicle travel track information of the vehicle at the intersection based on the vehicle position and the vehicle lane of the vehicle.
The track relation table between the positions and the lanes and the running track information can be pre-constructed, namely the track relation table comprises corresponding relations between various positions and various lanes and the running track information, and each position and each lane correspond to the running track information; the travel track information may include a travel track, i.e., a movement path, of the vehicle within the intersection.
Therefore, the vehicle position of the vehicle and the travel track information corresponding to the vehicle lane can be directly acquired from the track relation table, and the travel track information is determined as the vehicle travel track information of the vehicle at the intersection.
S1202, predicting target driving track information of each target at the intersection according to the target position and the target lane of each target.
And aiming at the target position and the target lane of the target, directly acquiring the driving track information corresponding to the target position and the target lane from the track relation table, and determining the driving track information as the target driving track information of the target at the intersection.
S1203 determines a second candidate target from the vehicle travel track information and the target travel track information.
The vehicle running track information comprises a vehicle running track, and the target running track information comprises a target running track. It is possible to determine whether or not the target travel locus of each target has an intersection with the vehicle travel locus, and determine the target having the intersection with the vehicle travel locus as the second candidate target.
For any target, the mode of judging whether the target running track of the target and the vehicle running track have a junction may be that the target running track and the vehicle running track of the target are respectively drawn on corresponding positions of a high-precision map, and whether the target running track and the vehicle running track have a junction is judged.
In the target screening method provided by the embodiment of the application, the vehicle running track information of the vehicle at the intersection is predicted according to the vehicle position and the vehicle lane of the vehicle, the target running track information of each target at the intersection is predicted according to the target position and the target lane of each target, and then the second candidate target is determined according to the vehicle running track information and the target running track information. According to the method, according to the predicted vehicle running track information of the vehicle at the intersection and the target running track information of the target at the intersection, the second candidate target interacted with the vehicle is determined, the problems of blind areas and omission of the target at the far end in the intersection are solved, and therefore the accuracy of target screening is improved.
In an exemplary embodiment, as shown in fig. 13, predicting vehicle travel track information of a vehicle at an intersection according to a vehicle position and a vehicle lane of the vehicle, includes the steps of:
s1301, determining the starting point of the vehicle entering the intersection and the ending point of the vehicle leaving the intersection according to the vehicle lanes and the vehicle positions.
In one exemplary embodiment, as shown in fig. 14, determining a start point of a vehicle entering an intersection and an end point of the vehicle exiting the intersection according to a vehicle lane and a vehicle position, comprises the steps of:
s1401, determining whether the vehicle is at an intersection according to the vehicle position.
Since the high-precision map refers to a digital map having high precision and detail, accurate position and attribute information of roads, buildings, terrains, traffic signs, traffic lights, lane lines, and other important landmarks are generally included in the high-precision map.
Therefore, the vehicle position can be mapped into the high-precision map, and it is determined whether the vehicle is at the intersection or not based on the position of the vehicle in the high-precision map.
S1402, if the vehicle is not at the intersection, determining a start point and an end point of the vehicle according to the vehicle lane.
If the vehicle is not at the intersection, which means that the vehicle has not entered the intersection, the coordinates of the point where the vehicle exits the entrance lane into the intersection region in the lane to which the vehicle belongs may be used as the start point of the vehicle, and the point where the vehicle exits the intersection region into the exit lane may be used as the end point of the vehicle.
As shown in fig. 15, when the vehicle is at the point A3, the starting point of the vehicle may be the point B; if the vehicle lane where the vehicle is located is a straight road, the end point of the vehicle may be point C; if the vehicle lane where the vehicle is located is a right turn, the end point of the vehicle may be the point D; if the vehicle lane where the vehicle is located is a left lane, the end point of the vehicle may be the E point; if the lane of the vehicle is a straight/right turn lane, the end point of the vehicle can be a point C and a point D; if the lane of the vehicle is a straight/left turn lane, the end points of the vehicle can be points C and E; if the lane of the vehicle is straight, left-turn, right-turn, then the vehicle may end at points C, D, and E.
It should be noted that, a certain deviation may exist between the selection of the points C, D and E, so that the midpoint of all stop lines of the exit road selected by the vehicle may be used as the point C, the point D or the point E, so that the influence on the intersection time difference may be ignored; for example, if the exit road has three lanes 1, 2, and 3, the middle lane 2 may be selected as the lane where the vehicle exits the intersection area, and the midpoint of the stop line of lane 2 may be selected as point C, point D, or point E; if there are two lanes 1, 2 in the exit, the point of the lane boundary common to the lanes 1, 2 is selected as point C, point D or point E. The other lane numbers are similar to this.
It should be noted that, when the vehicle lane is a turning lane, the start point and the end point of the vehicle are determined based on the same principle as the left-hand lane/right-hand lane.
S1403, if the vehicle is at the intersection, determining the vehicle position as the start point of the vehicle, and determining the end point of the vehicle from the vehicle lane.
If the vehicle is in the intersection area, the real-time vehicle position A point can be determined as the starting point of the vehicle; and determining the end point of the vehicle according to the lane of the vehicle.
The vehicle lane is the vehicle lane where the vehicle is located before entering the intersection area; the manner of determining the end point of the vehicle according to the vehicle lane is the same as that of the above embodiment, and this embodiment is not described here again.
In the embodiment of the application, whether the vehicle is at the intersection is determined according to the vehicle position, if the vehicle is not at the intersection, the starting point and the ending point of the vehicle are determined according to the vehicle lane, if the vehicle is at the intersection, the vehicle position is determined as the starting point of the vehicle, and the ending point of the vehicle is determined according to the vehicle lane. In the method, the starting point of the vehicle entering the intersection and the ending point of the vehicle leaving the intersection are determined according to the position of the vehicle in the intersection and the lane, so that the accuracy of determining the starting point and the ending point is improved.
S1302, determining vehicle running track information of the vehicle at the intersection according to the starting point and the ending point of the vehicle.
The vehicle travel track information may include a vehicle travel track equation of the vehicle.
If the vehicle lane is a straight road, the vehicle travel track can be regarded as approximately straight-line motion, and therefore, a straight-line track equation of the vehicle can be constructed according to the start point and the end point of the vehicle. The linear trajectory equation of the vehicle may be as shown in equation (1).
(1)
(2)
Wherein,can represent the origin coordinates>Indicating endpoint coordinates +.>Is a straight line slope, i.e.)>For the angle between the straight line and the x-axis, +.>Is the vehicle position coordinates of the vehicle.
If the vehicle lane is a right-hand lane/left-hand lane/turning around, the vehicle running track can be approximately regarded as circular motion, so that an arc track equation of the vehicle can be constructed according to the starting point and the ending point of the vehicle; the arc track equation at least needs the coordinates of a point on the circumference and the coordinates of a tangent point, and can take the starting point of the vehicle as the coordinates of the tangent point on the arc and the ending point of the vehicle as the coordinates of a point on the circumference. The arc trajectory equation of the vehicle can be constructed according to the formula (3) -the formula (6).
(3)
(4)
(5)
(6)
Thus, the above formulas (3) - (6) are combined to obtain
Therefore, the arc trajectory equation of the vehicle isWherein->Is any point in the circular arc track of the vehicle.
In the target screening method provided by the embodiment of the application, the starting point of the vehicle entering the intersection and the ending point of the vehicle leaving the intersection are determined according to the vehicle lane and the vehicle position, and the vehicle running track information of the vehicle at the intersection is determined according to the starting point and the ending point of the vehicle. According to the method, the vehicle running track information of the vehicle at the intersection can be constructed according to the preset starting point and the preset end point, and the reliability of the vehicle running track information is ensured.
In an exemplary embodiment, as shown in fig. 16, the target travel track information of each target at the intersection is predicted according to the target position and the target lane of each target, comprising the steps of:
s1601, for any one target, a start point of the target entering the intersection and an end point of the target leaving the intersection are determined according to the target position and the target lane.
In one embodiment, determining a start point of a target entering an intersection and an end point of the target exiting the intersection according to the target position and the target lane comprises: determining whether the target is at an intersection or not according to the target position; if the target is not at the intersection, determining a starting point and an ending point of the target according to the target lane; if the target is at the intersection, determining the target position as the starting point of the target, and determining the end point of the target according to the target lane.
First, whether the target is at the intersection may be determined based on the same principle of determining whether the vehicle is at the intersection. Therefore, the target position can be mapped into the high-precision map, and whether the target is in the intersection or not can be determined according to the position of the target in the high-precision map.
If the target is not at the intersection, which means that the target has not entered the intersection, the coordinates of the point at which the target exits the entrance lane into the intersection region in the lane to which the target belongs may be used as the start point of the target, and the point at which the vehicle exits the intersection region into the exit lane may be used as the end point of the target.
The method for determining the start point and the end point of the target according to the target lane is the same as the method for determining the start point and the end point of the vehicle according to the vehicle lane in the above embodiment, and the embodiments of the present application will not be described herein.
However, if the target is a non-motor vehicle, there is generally only one of the non-motor vehicle entrance and exit, and the midpoint of the stop line of the corresponding lane is directly selected as the start point and the end point.
If the target is in the intersection region, the target position can be determined as the starting point of the target, and the end point of the vehicle can be determined according to the target lane.
The target lane is the target lane where the vehicle is located before entering the intersection area; the method for determining the destination according to the destination lane is the same as the method for determining the destination according to the vehicle lane in the above embodiment, and this embodiment will not be described here again.
In the embodiment of the application, whether the target is at an intersection is determined according to the target position; if the target is not at the intersection, determining a starting point and an ending point of the target according to the target lane; if the target is at the intersection, determining the target position as the starting point of the target, and determining the end point of the target according to the target lane. In the method, the starting point of the target entering the intersection and the ending point of the target leaving the intersection are determined according to the position of the target in the intersection and the lanes, so that the accuracy of determining the starting point and the ending point is improved.
S1602, determining target driving track information of the target at the intersection according to the starting point and the ending point of the target.
The method for determining the target driving track information of the target at the intersection according to the start point and the end point of the target in the embodiment of the present application is the same as the principle of the method for determining the vehicle driving track information of the vehicle at the intersection according to the start point and the end point of the vehicle in the above embodiment, and the embodiments of the present application are not repeated here,
in the target screening method provided by the embodiment of the application, aiming at any target, determining a starting point of the target entering the intersection and an ending point of the target leaving the intersection according to the target position and the target lane; and determining target running track information of the target at the intersection according to the starting point and the ending point of the target. According to the method, the target running track information of the target at the intersection can be constructed according to the preset starting point and the preset end point, and the reliability of the target running track information is ensured.
The above-described embodiments are descriptions of the manner in which the vehicle travel track information and the target travel track information are determined, and the manner in which the second candidate target is determined from the vehicle travel track information and the respective target travel track information is described below by way of one embodiment. In an exemplary embodiment, as shown in fig. 17, determining the second candidate target based on the vehicle travel track information and the respective target travel track information includes the steps of:
s1701, determining a reference target which is intersected with the vehicle and an intersection position of the vehicle and the reference target according to the vehicle running track information and the target running track information.
The vehicle running track information comprises a vehicle running track equation of the vehicle, and the target running track information can comprise a target running track equation, so that the vehicle running track equation and the target running track equation can be combined, if a combined mode exists, interaction between the target and the vehicle is determined, and the combined equation is the intersection coordinate of the vehicle and the target.
And taking the object intersected with the vehicle as a reference object, and determining the corresponding intersection coordinate as an intersection position.
The form of intersection of the target and the vehicle can comprise intersection of a straight line and a straight line, intersection of a straight line and an arc, intersection of an arc and a straight line, intersection of an arc and an arc, and the intersection can be represented by an O point.
S1702, determining the vehicle duration of the vehicle reaching the intersection position according to the intersection position and the vehicle position.
Wherein the target duration represents a duration taken by the vehicle to reach the junction location from the vehicle location.
In one exemplary embodiment, the vehicle state information includes a vehicle speed; determining a vehicle duration for the vehicle to reach the intersection location based on the intersection location and the vehicle location, comprising: determining the vehicle distance of the vehicle reaching the intersection position according to the intersection position and the vehicle position; a vehicle duration is determined based on the vehicle range and the vehicle speed.
The vehicle distance represents the distance the vehicle needs from the vehicle location to the junction location; and calculating the vehicle distance between the vehicle position and the intersection position according to the vehicle position and the intersection position of the vehicle and the reference target.
For example, referring to fig. 15, if the vehicle is a straight running track, the linear distance BO from the start point B to the intersection point O of the straight line and the linear distance a between the current position A3 of the vehicle and the start point B can be calculated 3 B, determining the sum of the two linear distances as the vehicle distance for the vehicle to reach the intersection position; if the vehicle is in the circular arc running track, the circular arc distance BO between the point B and the point O and the straight line distance A between the point A3 of the vehicle position A3 and the starting point B of the vehicle can be calculated 3 B, the arc distance BO and the straight line distance A 3 B is determined as the distance of the vehicle to the junction.
Note that, referring to fig. 6, if the vehicle is currently at the intersection, the straight line distance/arc distance between the vehicle position a and the intersection O of the vehicle may be directly calculated, and the straight line distance/arc distance is determined as the vehicle path of the vehicle reaching the intersection.
Since the vehicle duration is the duration for which the vehicle reaches the intersection position from the vehicle position, the result of dividing the vehicle course by the vehicle speed is determined as the vehicle duration.
S1703, determining the target time length of the reference target reaching the intersection position according to the intersection position and the target position of the reference target.
In one exemplary embodiment, the target state information includes a target speed; determining a target duration of the reference target to the intersection position according to the intersection position and the target position of the reference target, including: determining a target distance for the target to reach the intersection position according to the intersection position and the target position of the reference target; and determining the target duration according to the target distance and the target speed.
Wherein the target distance represents the distance the target needs to reach the intersection location from the target location, and the target duration represents the duration the target takes to reach the intersection location from the target location.
The principle of determining the target distance for the target to reach the intersection position and determining the target duration in this embodiment is the same as that of determining the vehicle distance for the vehicle to reach the intersection position and determining the vehicle duration in the foregoing embodiments, and this embodiment of the present application will not be described herein.
S1704, determining a second candidate target according to the vehicle duration and the target duration.
The second candidate target may be a reference target whose time difference between the target duration and the vehicle duration is small. In one exemplary embodiment, as shown in fig. 18, determining the second candidate target according to the vehicle duration and the target duration includes the steps of:
s1801, determining a junction time difference between the vehicle and the reference target according to the vehicle duration and the target duration.
The difference between the vehicle duration and the target duration of the reference target is determined as the intersection time difference between the vehicle and the reference target.
S1802, determining a reference target with the absolute value of the intersection time difference smaller than or equal to a preset duration threshold as a second candidate target.
If the absolute value of the intersection time difference between the vehicle and the reference target is less than or equal to the duration threshold value, which indicates that the vehicle and the reference target have interaction risk, the reference target can be determined to be a second candidate target.
It should be noted that, the reference targets are targets that interact with the vehicle running track in the targets acquired by the roadside apparatus, and the reference targets include at least one, possibly a plurality of, reference targets, and therefore, among all the reference targets, the reference targets whose absolute values of the intersection time differences with the vehicle are less than or equal to the time threshold value are determined as second candidate targets, and the second candidate targets may include one or more targets.
The time length threshold can be selected according to the requirement on safety, and the larger the time length threshold is selected, the higher the requirement on safety is indicated; for example, the duration threshold may be 1 second, 2 seconds, 3 seconds, or the like.
According to the target screening method, a reference target which is intersected with a vehicle and an intersection position of the vehicle and the reference target are determined according to vehicle running track information and target running track information, vehicle duration of the vehicle reaching the intersection position is determined according to the intersection position and the vehicle position, target duration of the reference target reaching the intersection position is determined according to the intersection position and the target position of the reference target, and finally a second candidate target is determined according to the vehicle duration and the target duration. According to the method, the second candidate target is determined according to the time length of the vehicle reaching the intersection position and the time length of the target reaching the intersection position, so that the intersection time difference between the vehicle and the target is determined, the smaller the intersection time difference is, the easier the vehicle and the target collide, and therefore, the second candidate target is determined based on the principle, and the accuracy of the second candidate target is improved.
The above-described embodiment is an explanation of how to acquire the first candidate object and the second candidate object, and an explanation of how to select a dangerous object of the vehicle from the first candidate object and the second candidate object is provided below by way of an embodiment. In one exemplary embodiment, as shown in fig. 19, determining a dangerous object of a vehicle from among the objects according to the first candidate object and the second candidate object, includes the steps of:
s1901, eliminating repeated targets in the first candidate target and the second candidate target to obtain a total candidate target.
The first candidate target and the second candidate target may have the same target, one of the two same targets may be deleted, and one target may be reserved, so as to obtain the total candidate target.
For example, if there are P targets in the first candidate target, Q targets in the second candidate target, and S repeated targets in the first candidate target and the second candidate target, the total number of candidate targets (without repetition) actually screened according to the interaction area and the intersection time difference is (p+q-S), that is, there are (p+q-S) targets in the total candidate targets.
S1902, if the number of the total candidate targets is smaller than or equal to a preset number threshold, determining that the total candidate targets are dangerous targets.
Because the number of targets expected to be processed by the subsequent decision-making planning of the intelligent driving vehicle is limited, the processing number threshold dangerous targets can only be received, namely, the road side system needs to screen the number threshold dangerous targets from the total candidate targets, and the number threshold dangerous targets are sent to the intelligent driving vehicle. For example, the number threshold may be 16 or 32, etc.
Therefore, the number threshold is represented by T, and if the number of total candidate targets is smaller than or equal to the number threshold, which means that the number of targets screened according to the two modes of the preset interaction area and the intersection time difference is not more than T, the total candidate targets can be all used as dangerous targets.
S1903, if the number of the total candidate targets is larger than the number threshold, determining dangerous targets according to the relation between the number of the repeated targets and the number threshold.
If the number of the total candidate targets is larger than the number threshold, which means that the number of the targets screened according to the two modes of the preset interaction area and the intersection time difference is larger than T, T targets are required to be screened out from the total candidate targets screened according to the two modes of the preset interaction area and the intersection time difference to serve as dangerous targets.
Because the repeated targets are targets screened in the two modes of presetting interaction areas and intersection time differences, the risk of the repeated targets is larger, when the dangerous targets are selected from the total candidate targets, the repeated targets are preferentially considered, and the dangerous targets are preferentially selected from the repeated targets, so that the dangerous targets can be determined according to the relation between the number of the repeated targets and the number threshold.
In one exemplary embodiment, as shown in fig. 20, determining a dangerous target according to a relationship between the number of repetitive targets and a number threshold includes the steps of:
s2001, when the number of repetitive targets is greater than or equal to the number threshold, acquiring the intersection time differences between the repetitive targets and the vehicle, and determining the repetitive targets having smaller absolute values of the number threshold as dangerous targets.
Under the condition that the number of the repeated targets is larger than or equal to the number threshold, directly selecting a number threshold dangerous targets from the repeated targets; if the number of the repeated targets is equal to the number threshold, directly determining the repeated targets as dangerous targets; if the number of the repeated targets is greater than the number threshold, a number of dangerous targets can be selected from the repeated targets according to the absolute value of the intersection time difference of the repeated targets and the vehicle.
Since the smaller the absolute value of the intersection time difference is, the greater the possibility that the vehicle intersects with the target, the more dangerous the target is for the vehicle, therefore, the absolute values of the intersection time differences of the repetitive targets and the vehicle can be ordered in order from small to large, and the smallest T (number threshold) repetitive targets can be selected as dangerous targets.
S2002, in the case that the number of the repeated targets is smaller than the number threshold, determining the repeated targets as dangerous targets, and determining the residual number of the dangerous targets according to the number threshold and the number of the repeated targets.
If the number of the repeated targets is smaller than the number threshold, the number of the selectable dangerous targets is larger, and the repeated targets can be directly used as dangerous targets; an optional remaining number of dangerous objects is then determined.
For example, the number threshold is T, the repeat target is S, and the remaining number of dangerous targets is (T-S).
S2003, determining the remaining number of dangerous targets from the remaining total candidate targets except the repetitive targets.
Based on the determined remaining number of dangerous objects, the remaining number of dangerous objects may be selected from the remaining total candidate objects except for the repetitive object.
Continuing to take P targets in the first candidate targets, Q targets in the second candidate targets, S repeated targets in the first candidate targets and the second candidate targets, and describing that the total candidate targets are (P+Q-S) targets, the number threshold is T, the S repeated targets are determined to be dangerous targets, the residual number of dangerous targets is (T-S), and the number of the residual total candidate targets is (P+Q-S-S).
Thus, after S duplicate targets are identified as dangerous targets, there is a need to identify (T-S) dangerous targets from the (P+Q-S-S) remaining total candidate targets.
The remaining number of dangerous targets may be determined directly according to a preset dangerous evaluation model, for example, target state information and vehicle state information of the remaining total candidate targets and the remaining number are input into the dangerous evaluation model, each target state information and vehicle state information are analyzed through the dangerous evaluation model, the remaining number of target state information is output, and the remaining total candidate targets corresponding to the remaining number of target state information output by the dangerous evaluation model are determined as dangerous targets.
In the embodiment of the present application, under the condition that the number of the repeated targets is greater than or equal to the number threshold, acquiring the intersection time difference between the repeated targets and the vehicle, determining the repeated target with smaller absolute value of the number threshold number of the intersection time differences as the dangerous target, under the condition that the number of the repeated targets is smaller than the number threshold, determining the repeated target as the dangerous target, determining the remaining number of the dangerous target according to the number threshold number and the number of the repeated targets, and determining the remaining number of the dangerous targets from the remaining total candidate targets except the repeated targets. In the method, as the first candidate target and the second candidate target are targets screened in two ways, repeated targets in the first candidate target and the second candidate target are preferentially used as dangerous targets, and the accuracy of target screening can be ensured.
In the target screening method provided by the embodiment of the application, repeated targets in the first candidate target and the second candidate target are removed, so that total candidate targets are obtained; if the number of the total candidate targets is smaller than or equal to a preset number threshold, determining that the total candidate targets are dangerous targets; if the number of the total candidate targets is larger than the number threshold, determining dangerous targets according to the relation between the number of the repeated targets and the number threshold. In the method, the total candidate targets in the first candidate target and the second candidate target are determined, and two dangerous target modes are provided according to the relation between the total candidate targets and the number threshold value, so that the comprehensiveness and the reliability of target screening are ensured.
The remaining number of hazard targets may also be determined from the remaining total candidate targets by a preset hazard evaluation function, which is described in detail below by way of one embodiment, in which the vehicle status information includes vehicle position; the target state information includes a target position and a target speed; as shown in fig. 21, determining the remaining number of dangerous objects from the remaining total candidate objects except for the repetitive object includes the steps of:
s2101, determining the real distance between the vehicle and the rest total candidate targets according to the vehicle position and the target positions of the rest total candidate targets.
The manner in which the true distance of the vehicle from the remaining total candidate objects is determined for any remaining total candidate object may be as shown in equation (7).
(7)
Wherein,representing the true distance between the vehicle and the remaining total candidate objects, < >>Representing the vehicle position of the vehicle->Target position representing the remaining total candidate target, +.>May represent the abscissa of the vehicle, +.>May represent the ordinate of the vehicle, +.>The abscissa, +.>The ordinate of the remaining total candidate objects may be represented. />
S2102, determining target distances of the remaining total candidate targets according to a preset time threshold and target speeds of the remaining total candidate targets.
The product of the time threshold and the target speed of the remaining total candidate targets is determined as the target distance of the remaining total candidate targets.
S2103, determining danger evaluation values of the rest total candidate targets according to the real distance, the target distance, the preset speed weight and the distance weight.
And (3) carrying out weighted difference calculation on the real distance and the target distance, and determining the risk evaluation value of the rest total candidate targets, wherein the risk evaluation value is shown in a formula (8).
(8)
Wherein J represents the risk evaluation value of the remaining total candidate targets,、/>the weight coefficients of the speed and the distance are positive numbers respectively; / >The target speed for the remaining total candidate targets may be in m/s; />For a set time threshold, +.>;/>The units may be meters for the true distance of the vehicle from the remaining total candidate targets.
And S2104, determining the rest total candidate targets with the rest number of risk evaluation values smaller as risk targets.
And determining the remaining total candidate targets with the smallest risk evaluation value as dangerous targets, and determining the (T-S) remaining total candidate targets with the smallest risk evaluation value as dangerous targets.
In the target screening method provided by the embodiment of the application, the real distance between the vehicle and the other total candidate targets is determined according to the vehicle position and the target positions of the other total candidate targets, the target distance between the other total candidate targets is determined according to the preset time threshold and the target speed of the other total candidate targets, and then the risk evaluation value of the other total candidate targets is determined according to the real distance, the target distance, the preset speed weight and the preset distance weight; and determining the rest total candidate targets with the rest number of risk evaluation values smaller as risk targets. In the method, repeated targets are firstly determined to be dangerous targets, then the dangers of the rest total candidate targets are further judged through a dangerous evaluation function, and the rest number of dangerous targets are screened from the rest total candidate targets, so that the accuracy of screening dangerous targets is further improved.
In an exemplary embodiment, the embodiment of the present application further provides a target screening method, as shown in fig. 22, including the following steps:
s2201, acquiring vehicle state information acquired by a vehicle in response to a target screening request of the vehicle.
S2202, according to the vehicle state information, matching the vehicle with a plurality of candidate targets acquired by the road side equipment, and determining the candidate targets which are not matched with the vehicle as targets.
Wherein the target includes pedestrians, non-vehicles, and vehicles.
S2203 determines a lane in which the vehicle is located, based on the position of the vehicle on the high-precision map.
If the vehicle is currently in the lane, determining the lane in which the vehicle is currently located as the lane of the vehicle; if the vehicle is not currently in the lane, the historical nearest vehicle lane is determined as the lane in which the vehicle is currently located.
If the vehicle is not in the lane currently, if the history nearest lane is a lane, determining the history nearest vehicle lane as the lane in which the vehicle is currently located; if the historic nearest lane is a plurality of lanes, the lane of the vehicle can be determined according to the first distance between the vehicle and the lane stop line of the affiliated lane and the second distance between the vehicle and the non-motor lane stop line.
S2204, determining a preset interaction area according to the vehicle lane and the vehicle position, and determining the target in the preset interaction area as a first candidate target.
The preset interaction area comprises a first interaction area, a second interaction area, a third interaction area, a fourth interaction area and a fifth interaction area; according to the lane of the vehicle, a first interaction area of the vehicle and the pedestrian, a second interaction area of the vehicle and the non-motor vehicle and a third interaction area of the vehicle and the motor vehicle are determined.
And determining pedestrians in the first interaction area in each target as a first candidate target, and determining non-motor vehicles in the second interaction area in each target as the first candidate target.
A candidate vehicle in a third interaction region is acquired from each target, and a vehicle having a history of nearest target lanes is determined as a first candidate target, and a vehicle having no history of nearest target lanes and a distance from the vehicle of less than 12 meters is determined as a first candidate target.
Determining an area 8 meters away from the vehicle position as a fourth interaction area, and determining pedestrians in the fourth interaction area as first candidate targets; an area 10 meters from the vehicle position is determined as a fifth interaction area, and a non-motor vehicle in the fifth interaction area is determined as a first candidate object.
S2205 acquires a reference object that intersects with a vehicle travel locus of the vehicle, and calculates a difference between the intersection of the reference object and the vehicle.
S2206, determining the reference target with the absolute value of the difference between the intersections smaller than or equal to the preset duration threshold as a second candidate target.
Determining a starting point of the vehicle entering the intersection and a finishing point of the vehicle leaving the intersection according to the vehicle lanes of the vehicle, and constructing a vehicle track equation of the vehicle according to the starting point and the finishing point;
determining a starting point of the target entering the intersection and an ending point of the target leaving the intersection according to a target lane where the target is located, and constructing a target track equation of the target according to the starting point and the ending point;
and determining a reference target which is intersected with the vehicle and an intersection position of the reference target and the vehicle according to a vehicle track equation of the vehicle and a target track equation of the target.
For any reference target, calculating the vehicle distance of the vehicle to the intersection position, and calculating the vehicle duration according to the vehicle distance and the vehicle speed; calculating a target distance of the reference target to the intersection position, and calculating a target duration according to the target distance and the target speed; determining a junction time difference according to the vehicle duration and the target duration;
if the intersection time difference is greater than M seconds, determining that the corresponding reference target and the vehicle have no interaction risk; and if the intersection time difference is less than or equal to M seconds, determining that the corresponding reference target has interaction risk with the vehicle, and determining the corresponding reference target as a second candidate target.
S2207 determines a dangerous object of the vehicle from the first candidate object and the second candidate object.
If the vehicle can only accept T dangerous targets; if P first candidate targets are screened based on the preset interaction area, Q second candidate targets are screened based on the intersection time difference, and S identical targets in the preset interaction area and the intersection time difference are provided.
If P+Q-S is less than or equal to T, determining P+Q-S as a final dangerous target;
if P+Q-S is greater than T and S is greater than or equal to T, sorting the targets in S according to the intersection time difference, wherein the smaller the more dangerous the targets are, the more dangerous T are selected.
If P+Q-S is greater than T and S is less than T, determining S targets as dangerous targets, and screening (T-S) dangerous targets from the remaining targets according to an evaluation function.
The object screening method in the embodiment of the present application is performed in a road side system, but because the road side system does not know which object the vehicle is acquired by the road side device, it is necessary to match the vehicle to the object acquired by the road side device according to the vehicle state information of the vehicle, and the object which is not matched with the vehicle is taken as another object. The preset interaction area in the embodiment of the application is defined according to the size of the intersection, and is defined by combining channeling (straight-going, left-turning, right-turning, turning around and other scenes). According to the running direction of the vehicle, the first part of screening (screening based on dangerous areas) is carried out by combining with the canalization of other targets and whether the first part of screening is carried out in a corresponding preset interaction area or not; secondly, according to the states of the vehicle and other targets, obtaining an intersection time difference, and then screening a second part according to the intersection time difference (screening based on the intersection time difference); and finally, fusing the two screening methods to obtain a final screening result. The information such as coordinates and areas in the embodiments of the present application are derived from a high-precision map.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a target screening device for realizing the target screening method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitations in one or more embodiments of the target screening device provided below may be referred to above for the limitations of the target screening method, and will not be repeated here.
In an exemplary embodiment, as shown in fig. 23, there is provided a target screening apparatus 2300, comprising: a response module 2301, a determination module 2302 and a screening module 2303, wherein:
a response module 2301, configured to obtain vehicle state information collected by a vehicle at a current moment and target state information of at least one target collected by a roadside device at an intersection where the vehicle is located;
a determining module 2302 for determining a first candidate object in a preset interaction area of the vehicle and a second candidate object that intersects the vehicle based on the vehicle state information and the object state information;
a screening module 2303 is configured to determine a dangerous object for the vehicle from the objects based on the first candidate object and the second candidate object.
In one embodiment, the vehicle status information includes a vehicle location; the determination module 2302 includes:
the lane determining unit is used for determining a vehicle lane in which the vehicle is positioned according to the vehicle position;
the first area determining unit is used for determining a preset interaction area of the vehicle according to the lane of the vehicle;
and the first determining unit is used for determining the target in the preset interaction area in the at least one target as a first candidate target according to the state information of each target.
In one embodiment, the lane determining unit is specifically configured to determine whether the vehicle is in the lane at the current time according to the vehicle position; if the vehicle is in the lane at the current moment, determining the lane in which the vehicle is positioned at the current moment as a vehicle lane; if the vehicle is not in the lane at the current time, determining the vehicle history nearest lane of the vehicle as the vehicle lane.
In one embodiment, the preset interaction region comprises a first interaction region of a vehicle and a pedestrian, a second interaction region of the vehicle and a non-motor vehicle, and a third interaction region of the vehicle and the motor vehicle; the first determination unit includes:
a first determining subunit, configured to determine, according to the status information of each target, a pedestrian, a non-motor vehicle, and a motor vehicle in at least one target;
a second determining subunit configured to determine a pedestrian in the first interaction area as a first candidate target;
a third determination subunit configured to determine a non-motor vehicle in the second interaction region as a first candidate target;
and the fourth determination subunit is used for determining the first candidate target according to the third interaction area and the target lane where the motor vehicle is located in the case that the target is the motor vehicle.
In one embodiment, the fourth determination subunit is specifically configured to obtain, from each vehicle, at least one candidate vehicle in the third interaction region; for any candidate motor vehicle, if the candidate motor vehicle has a target lane, determining the candidate motor vehicle as a first candidate target; and if the candidate motor vehicle does not have the target lane, determining the candidate motor vehicle with the distance from the vehicle being smaller than the preset motor vehicle distance threshold value as a first candidate target.
In one embodiment, the vehicle status information includes a vehicle location; the determination module 2302 includes:
the second area determining unit is used for determining a preset interaction area of the vehicle according to the vehicle position and a preset distance threshold value;
and the second determining unit is used for determining the target in the preset interaction area in at least one target as a first candidate target according to the target state information of each target.
In one embodiment, the preset interaction region of the vehicle includes a fourth interaction region of the vehicle with a pedestrian and a fifth interaction region of the vehicle with a non-motor vehicle; the distance threshold includes a pedestrian distance threshold and a non-motor vehicle distance threshold; the second area determining unit is specifically configured to determine a fourth interaction area according to the vehicle position and the pedestrian distance threshold; a fifth interaction zone is determined based on the vehicle location and the non-motor vehicle distance threshold.
In one embodiment, the vehicle status information includes a vehicle location; the target state information includes a target position; the determination module 2302 includes:
the first prediction unit is used for predicting vehicle running track information of the vehicle at the intersection according to the vehicle position and the vehicle lane of the vehicle;
the second prediction unit is used for predicting target running track information of each target at the intersection according to the target position and the target lane of each target;
and a third determining unit for determining a second candidate target according to the vehicle running track information and the target running track information.
In one embodiment, the first prediction unit includes:
a fifth determining subunit, configured to determine, according to the vehicle lane and the vehicle position, a start point at which the vehicle enters the intersection and an end point at which the vehicle leaves the intersection;
and the sixth determination subunit is used for determining the vehicle running track information of the vehicle at the intersection according to the starting point and the ending point of the vehicle.
In one embodiment, the second prediction unit includes:
a seventh determining subunit, configured to determine, for any one of the targets, a start point at which the target enters the intersection and an end point at which the target leaves the intersection according to the target position and the target lane;
and the eighth determination subunit is used for determining target running track information of the target at the intersection according to the starting point and the ending point of the target.
In one embodiment, the third determining unit comprises:
a ninth determining subunit, configured to determine, according to the vehicle running track information and the target running track information, a reference target that intersects with the vehicle, and a position at which the vehicle intersects with the reference target;
a tenth determination subunit, configured to determine a vehicle duration for the vehicle to reach the intersection location according to the intersection location and the vehicle location;
an eleventh determination subunit, configured to determine, according to the intersection position and the target position of the reference target, a target duration for the reference target to reach the intersection position;
a twelfth determination subunit, configured to determine a second candidate target according to the vehicle duration and the target duration.
In one embodiment, the twelfth determining subunit is specifically configured to determine, according to the vehicle duration and the target duration, a junction time difference between the vehicle and the reference target; and determining a reference target with the intersection time difference smaller than an absolute value or equal to a preset duration threshold as a second candidate target.
In one embodiment, the screening module 2303 includes:
the obtaining unit is used for eliminating repeated targets in the first candidate target and the second candidate target to obtain total candidate targets;
a fourth determining unit, configured to determine that the total candidate target is a dangerous target if the number of total candidate targets is less than or equal to a preset number threshold;
And a fifth determining unit, configured to determine the dangerous target according to a relationship between the number of repeated targets and the number threshold if the number of total candidate targets is greater than the number threshold.
In one embodiment, the fifth determining unit includes:
a thirteenth determination subunit, configured to, when the number of the repeated targets is greater than or equal to the number threshold, obtain a junction time difference between the repeated targets and the vehicle, and determine, as a dangerous target, a repeated target with a smaller absolute value of the number threshold;
a fourteenth determination subunit, configured to determine, if the number of repeated targets is smaller than the number threshold, the repeated targets as dangerous targets, and determine the remaining number of dangerous targets according to the number threshold and the number of repeated targets;
a fifteenth determination subunit for determining a remaining number of dangerous objects from the remaining total candidate objects excluding the repetitive objects.
In one embodiment, the vehicle status information includes a vehicle location; the target state information includes a target position and a target speed; the fifteenth determination subunit is specifically configured to determine a real distance between the vehicle and the remaining total candidate targets according to the vehicle position and the target positions of the remaining total candidate targets; determining target distances of the other total candidate targets according to a preset time threshold and target speeds of the other total candidate targets; determining the risk evaluation value of the rest total candidate targets according to the real distance, the target distance, the preset speed weight and the distance weight; and determining the rest total candidate targets with the rest number of risk evaluation values smaller as risk targets.
The respective modules in the above-described target screening apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one exemplary embodiment, a computer device is provided, which may be a server, and the internal structure thereof may be as shown in fig. 24. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured 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, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing target screening data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a target screening method.
It will be appreciated by those skilled in the art that the structure shown in fig. 24 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application is applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.
The implementation principle and technical effects of each step implemented by the processor in this embodiment are similar to those of the above-mentioned target screening method, and are not described herein.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
The steps of the computer program implemented when executed by the processor in this embodiment realize the principle and technical effects similar to those of the above-described target screening method, and are not described herein again.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.
The steps of the computer program implemented when executed by the processor in this embodiment realize the principle and technical effects similar to those of the above-described target screening method, and are not described herein again.
It should be noted that, the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are all information and data authorized by the user or sufficiently authorized by each party, and the collection, use, and processing of the relevant data are required to meet the relevant regulations.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims (16)

1. A method of screening for a target, the method comprising:
acquiring vehicle state information acquired by a vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located;
determining a first candidate target in a preset interaction area of the vehicle and a second candidate target which is intersected with the vehicle according to the vehicle state information and each target state information;
And determining dangerous targets of the vehicle from the targets according to the first candidate target and the second candidate target.
2. The method of claim 1, wherein the vehicle status information includes a vehicle location; determining a first candidate target in a preset interaction area of the vehicle according to the vehicle state information and each target state information, wherein the first candidate target comprises:
determining a vehicle lane in which the vehicle is located according to the vehicle position;
determining a preset interaction area of the vehicle according to the vehicle lane;
and determining the target in the preset interaction area in the at least one target as the first candidate target according to the target state information.
3. The method of claim 2, wherein said determining a vehicle lane in which said vehicle is located based on said vehicle location comprises:
determining whether the vehicle is in a lane at the current moment according to the vehicle position;
if the vehicle is in the lane at the current moment, determining the lane in which the vehicle is positioned at the current moment as the vehicle lane;
and if the vehicle is not in the lane at the current moment, determining a vehicle history nearest lane of the vehicle as the vehicle lane.
4. The method of claim 2, wherein the pre-set interaction zone comprises a first interaction zone of the vehicle with a pedestrian, a second interaction zone of the vehicle with a non-motor vehicle, and a third interaction zone of the vehicle with a motor vehicle; the determining, according to each target state information, a target in the preset interaction area in the at least one target as the first candidate target includes:
determining pedestrians, non-motor vehicles and motor vehicles in the at least one target according to the target state information;
determining a pedestrian in the first interaction region as the first candidate target;
determining a non-motor vehicle in the second interaction region as the first candidate target;
and under the condition that the target is a motor vehicle, determining the first candidate target according to the third interaction area and a target lane where the motor vehicle is located.
5. The method of claim 4, wherein the determining the first candidate target based on the third interaction region and a target lane in which the motor vehicle is located comprises:
acquiring at least one candidate motor vehicle in a third interaction area from each motor vehicle;
For any one of the candidate vehicles, if the candidate vehicle has the target lane, determining the candidate vehicle as the first candidate target;
and if the target lane does not exist in the candidate motor vehicles, determining the candidate motor vehicles with the distance between the candidate motor vehicles and the vehicle being smaller than a preset motor vehicle distance threshold value as the first candidate target.
6. The method of any one of claims 1-5, wherein the vehicle status information includes a vehicle location; determining a first candidate target in a preset interaction area of the vehicle according to the vehicle state information and each target state information, wherein the first candidate target comprises:
determining a preset interaction area of the vehicle according to the vehicle position and a preset distance threshold;
and determining the target in the preset interaction area in the at least one target as the first candidate target according to the target state information of each target.
7. The method of claim 6, wherein the pre-set interaction zone of the vehicle comprises a fourth interaction zone of the vehicle with a pedestrian, and a fifth interaction zone of the vehicle with a non-motor vehicle; the distance threshold comprises a pedestrian distance threshold and a non-motor vehicle distance threshold; the determining the preset interaction area of the vehicle according to the vehicle position and the preset distance threshold value includes:
Determining the fourth interaction area according to the vehicle position and the pedestrian distance threshold;
and determining the fifth interaction area according to the vehicle position and the non-motor vehicle distance threshold value.
8. The method of any one of claims 1-5, wherein the vehicle status information includes a vehicle location; the target state information includes a target position; determining a second candidate object having an intersection with the vehicle based on the vehicle state information and each of the object state information, comprising:
predicting vehicle running track information of the vehicle at the intersection according to the vehicle position and the vehicle lane of the vehicle;
predicting target driving track information of each target at the intersection according to the target position and the target lane of each target;
and determining the second candidate target according to the vehicle running track information and the target running track information.
9. The method of claim 8, wherein predicting vehicle travel trajectory information for the vehicle at the intersection based on the vehicle location and the vehicle lane of the vehicle comprises:
determining a starting point of the vehicle entering the intersection and an ending point of the vehicle leaving the intersection according to the vehicle lane and the vehicle position;
And determining the vehicle running track information of the vehicle at the intersection according to the starting point and the ending point of the vehicle.
10. The method of claim 8, wherein predicting target travel track information for each of the targets at the intersection based on the target location and the target lane for each of the targets comprises:
determining a starting point of the target entering the intersection and an ending point of the target leaving the intersection according to the target position and the target lane aiming at any target;
and determining target driving track information of the target at the intersection according to the starting point and the ending point of the target.
11. The method of claim 8, wherein the determining the second candidate target based on the vehicle travel track information and each of the target travel track information comprises:
determining a reference target which is intersected with the vehicle and an intersection position of the vehicle and the reference target according to the vehicle running track information and the target running track information;
determining a vehicle duration of the vehicle reaching the intersection position according to the intersection position and the vehicle position;
Determining the target time length of the reference target to reach the intersection position according to the intersection position and the target position of the reference target;
and determining the second candidate target according to the vehicle duration and the target duration.
12. The method of claim 11, wherein the determining the second candidate target based on the vehicle duration and the target duration comprises:
determining a junction time difference between the vehicle and the reference target according to the vehicle duration and the target duration;
and determining a reference target with the absolute value of the intersection time difference smaller than or equal to a preset duration threshold as the second candidate target.
13. The method of any one of claims 1-5, wherein said determining a dangerous target for the vehicle from each of the targets based on the first candidate target and the second candidate target comprises:
removing repeated targets in the first candidate target and the second candidate target to obtain total candidate targets;
if the number of the total candidate targets is smaller than or equal to a preset number threshold, determining that the total candidate targets are dangerous targets;
And if the number of the total candidate targets is larger than the number threshold, determining the dangerous targets according to the relation between the number of the repeated targets and the number threshold.
14. The method of claim 13, wherein said determining the dangerous target based on a relationship between the number of repeating targets and the number threshold comprises:
acquiring the intersection time differences of the repeated targets and the vehicle under the condition that the number of the repeated targets is larger than or equal to the number threshold value, and determining the repeated targets with smaller absolute values of the number threshold value as the dangerous targets;
if the number of the repeated targets is smaller than the number threshold, determining the repeated targets as dangerous targets, and determining the residual number of the dangerous targets according to the number threshold and the number of the repeated targets;
determining the remaining number of dangerous objects from the remaining total candidate objects except for the duplicate object.
15. The method of claim 14, wherein the vehicle status information includes a vehicle location; the target state information comprises a target position and a target speed; the determining the remaining number of dangerous objects from the remaining total candidate objects except the repetitive object includes:
Determining the real distance between the vehicle and the rest total candidate targets according to the vehicle position and the target positions of the rest total candidate targets;
determining target distances of the remaining total candidate targets according to a preset time threshold and target speeds of the remaining total candidate targets;
determining the risk evaluation value of the rest total candidate targets according to the real distance, the target distance, the preset speed weight and the distance weight;
and determining the rest total candidate targets with the rest number of risk evaluation values smaller as the risk targets.
16. A target screening apparatus, the apparatus comprising:
the response module is used for responding to a target screening request sent by a vehicle and acquiring vehicle state information acquired by the vehicle at the current moment and target state information of at least one target acquired by road side equipment in an intersection where the vehicle is located;
a determining module, configured to determine, according to the vehicle state information and each of the target state information, a first candidate target in a preset interaction area of the vehicle, and a second candidate target that has an intersection with the vehicle;
and the screening module is used for determining dangerous targets of the vehicle from the targets according to the first candidate target and the second candidate target.
CN202410162172.2A 2024-02-05 2024-02-05 Target screening method and device Pending CN117698752A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410162172.2A CN117698752A (en) 2024-02-05 2024-02-05 Target screening method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410162172.2A CN117698752A (en) 2024-02-05 2024-02-05 Target screening method and device

Publications (1)

Publication Number Publication Date
CN117698752A true CN117698752A (en) 2024-03-15

Family

ID=90151987

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410162172.2A Pending CN117698752A (en) 2024-02-05 2024-02-05 Target screening method and device

Country Status (1)

Country Link
CN (1) CN117698752A (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017058774A (en) * 2015-09-14 2017-03-23 住友電気工業株式会社 Driving assistance device system, driving assistance device, and computer program
CN114475585A (en) * 2022-03-11 2022-05-13 阿波罗智能技术(北京)有限公司 Automatic driving method and device at intersection, electronic equipment and automatic driving vehicle
CN114475614A (en) * 2022-03-21 2022-05-13 中国第一汽车股份有限公司 Method, device, medium and equipment for screening dangerous targets
CN115009275A (en) * 2022-08-08 2022-09-06 北京理工大学前沿技术研究院 Vehicle track prediction method and system in urban scene and storage medium
CN116224317A (en) * 2022-12-29 2023-06-06 上海汽车工业(集团)有限公司 Method and device for predicting motion trail of non-motor vehicle and related equipment
CN116653963A (en) * 2023-07-31 2023-08-29 福思(杭州)智能科技有限公司 Vehicle lane change control method, system and intelligent driving domain controller
DE102023001318A1 (en) * 2022-04-13 2023-10-19 Mercedes-Benz Group AG Method and device for assisted driving of a vehicle at intersections
CN116935693A (en) * 2022-03-31 2023-10-24 中兴终端有限公司 Collision early warning method, vehicle-mounted terminal and storage medium
CN117253361A (en) * 2023-09-18 2023-12-19 北京京东振世信息技术有限公司 Method and device for generating vehicle running track information

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017058774A (en) * 2015-09-14 2017-03-23 住友電気工業株式会社 Driving assistance device system, driving assistance device, and computer program
CN114475585A (en) * 2022-03-11 2022-05-13 阿波罗智能技术(北京)有限公司 Automatic driving method and device at intersection, electronic equipment and automatic driving vehicle
CN114475614A (en) * 2022-03-21 2022-05-13 中国第一汽车股份有限公司 Method, device, medium and equipment for screening dangerous targets
CN116935693A (en) * 2022-03-31 2023-10-24 中兴终端有限公司 Collision early warning method, vehicle-mounted terminal and storage medium
DE102023001318A1 (en) * 2022-04-13 2023-10-19 Mercedes-Benz Group AG Method and device for assisted driving of a vehicle at intersections
CN115009275A (en) * 2022-08-08 2022-09-06 北京理工大学前沿技术研究院 Vehicle track prediction method and system in urban scene and storage medium
CN116224317A (en) * 2022-12-29 2023-06-06 上海汽车工业(集团)有限公司 Method and device for predicting motion trail of non-motor vehicle and related equipment
CN116653963A (en) * 2023-07-31 2023-08-29 福思(杭州)智能科技有限公司 Vehicle lane change control method, system and intelligent driving domain controller
CN117253361A (en) * 2023-09-18 2023-12-19 北京京东振世信息技术有限公司 Method and device for generating vehicle running track information

Similar Documents

Publication Publication Date Title
US11217012B2 (en) System and method for identifying travel way features for autonomous vehicle motion control
CN109801508B (en) Method and device for predicting movement track of obstacle at intersection
US11933627B2 (en) High-definition map building method and apparatus
EP3901782B1 (en) Lane-centric road network model for navigation
US20200255027A1 (en) Method for planning trajectory of vehicle
CN110562258B (en) Method for vehicle automatic lane change decision, vehicle-mounted equipment and storage medium
Yang et al. Generating lane-based intersection maps from crowdsourcing big trace data
WO2019079738A1 (en) Autonomous vehicle routing
US20220261601A1 (en) Multiple Stage Image Based Object Detection and Recognition
CN109643118B (en) Influencing a function of a vehicle based on function-related information about the environment of the vehicle
CN112106124A (en) System and method for using V2X and sensor data
CN114995451A (en) Control method, road side equipment and system for cooperative automatic driving of vehicle and road
KR20150128712A (en) Lane-level vehicle navigation for vehicle routing and traffic management
US11604075B2 (en) Systems and methods for deriving planned paths for vehicles using path priors
US11887324B2 (en) Cross-modality active learning for object detection
CN104875740B (en) For managing the method for following space, main vehicle and following space management unit
JP2022041923A (en) Vehicle path designation using connected data analysis platform
US11465620B1 (en) Lane generation
US20230159052A1 (en) Method for processing behavior data, method for controlling autonomous vehicle, and autonomous vehicle
US20210404841A1 (en) Systems and methods for inferring information about stationary elements based on semantic relationships
US11719553B2 (en) Spotfinder
CN117698752A (en) Target screening method and device
CN111337043B (en) Path planning method and device, storage medium and electronic equipment
CN115112138A (en) Trajectory planning information generation method and device, electronic equipment and storage medium
KR102302486B1 (en) Urban road speed processing method, urban road speed processing device, device and non-volatile computer storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination