CN112277952B - Method for screening key obstacles under structured road - Google Patents

Method for screening key obstacles under structured road Download PDF

Info

Publication number
CN112277952B
CN112277952B CN202011220296.XA CN202011220296A CN112277952B CN 112277952 B CN112277952 B CN 112277952B CN 202011220296 A CN202011220296 A CN 202011220296A CN 112277952 B CN112277952 B CN 112277952B
Authority
CN
China
Prior art keywords
obj
main vehicle
lane
obstacle
matrix
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.)
Active
Application number
CN202011220296.XA
Other languages
Chinese (zh)
Other versions
CN112277952A (en
Inventor
刘俊
郭洪艳
高振海
胡云峰
戴启坤
梁德宝
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.)
Jilin University
Original Assignee
Jilin University
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 Jilin University filed Critical Jilin University
Priority to CN202011220296.XA priority Critical patent/CN112277952B/en
Publication of CN112277952A publication Critical patent/CN112277952A/en
Application granted granted Critical
Publication of CN112277952B publication Critical patent/CN112277952B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
    • B60W40/06—Road conditions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/50—Barriers

Abstract

The invention provides a method for screening key obstacles under a structured road, which is characterized in that a driving scene of a vehicle is judged based on a driving scene judgment mode, and key obstacles which have the most obvious influence on the driving of a main vehicle are screened out based on an obstacle list output by a sensing module; the method provides key barrier screening based on a target-level sensing fusion result, and can provide key barrier information for development of an intelligent driving auxiliary system and an unmanned driving system; the method designs a scene classification method based on the types of the left lane line and the right lane line adjacent to the main vehicle, and can realize the segmentation of the driving scene of the main vehicle under the structured road; the method provides a method for screening the key obstacles by positioning the obstacles on the lane and then comparing the relative longitudinal distance between the obstacles and the main vehicle, so that the accurate screening of the key obstacle information can be finally realized by traversing all obstacle information acquired by sensing.

Description

Method for screening key obstacles under structured road
Technical Field
The invention belongs to the technical field of intelligent automobiles, relates to a method for screening key obstacles, and particularly relates to a method for screening key obstacles under a structured road.
Background
With the improvement of the intelligent level of the automobile, the acceptance of users to the vehicle-mounted intelligent driving auxiliary system and the unmanned driving system is higher and higher, the development of the high-performance intelligent system depends on multiple sensing means, the final output results of many sensing fusion means are all barrier lists around the main automobile at present, but the development of the intelligent system only needs key barrier information which has the most obvious influence on the driving of the main automobile, the key barrier is the barrier of each lane closest to the main automobile, the influence of other non-key barriers on the driving of the automobile is weaker, and how to accurately screen the information of the key barrier becomes a key problem for developing the high-performance driving auxiliary system and the unmanned driving system.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a method for screening key obstacles under a structured road, which is characterized in that a driving scene of a vehicle is judged based on a driving scene judgment mode, and the key obstacles which have the most obvious influence on the driving of a main vehicle are screened based on an obstacle list output by a sensing module, wherein the key obstacles are the obstacles of each lane closest to the main vehicle, so that an intelligent driving auxiliary system and the unmanned driving technical function can be conveniently developed.
The invention is realized by adopting the following technical scheme:
a method for screening key obstacles under a structured road judges a driving scene of a main vehicle based on a driving scene judgment mode, screens out key obstacles which have the most obvious influence on the driving of the main vehicle based on an obstacle list output by a sensing module, wherein the key obstacles are the obstacles of each lane closest to the main vehicle, and facilitates the development of a main vehicle driving auxiliary strategy and a main vehicle unmanned system decision and control strategy, and the method comprises the following specific steps:
step one, defining required sensing information:
judging a driving scene and screening obstacles in a structured road environment, wherein a lane line on the structured road is clearly visible, and screening key obstacles is carried out in a vehicle coordinate system, so that a vehicle coordinate system is defined, the origin of coordinates of the vehicle coordinate system is at the center of mass o of the main vehicle, the x axis faces the advancing direction of the main vehicle, and the positive direction of the y axis is the direction of anticlockwise rotation of the x axis by 90 degrees;
the key obstacle screening is based on single or multi-sensing fused target obstacle sensing information and lane line information, so that the sensing information required by the key obstacle screening is defined, the total number of obstacles output by a sensing module is firstly defined as n, and the coordinate (x) of each obstacle relative to a main vehicle is defined asi,yi) I-1, …, n, where i represents the number of the obstacle, and assigns the information of all obstacles into an m × n matrix, the first three rows of which are sequentially the number i of the obstacle, the longitudinal coordinate x of the obstacle relative to the host vehicleiAnd lateral coordinate y of the obstacle relative to the host vehicleiThe fourth row of the matrix to the m-1 row of the matrix may be defined by a user to output other information of the obstacle, the last row of the matrix, i.e., the m-th row, is reserved as a marker for locating the obstacle in the lane, when the last row of the matrix before the obstacle is not located is assigned to be 0, the obstacle matrix is named as OBJ, and each column of the matrix represents all information of the corresponding obstacle, as shown in detail below:
Figure BDA0002761724010000021
defining lane line information acquired by the main vehicle, wherein the lane line information acquired by the main vehicle comprises four lane line identifications which are named as ll, l, r and rr from left to right in sequence, the lane line information acquired by the main vehicle is a cubic interpolation function, and the interpolation parameter information acquired by the lane line is dj,cj,bj,ajWhere the subscript j represents the parameters of the associated identified lane line, the lane line equation can be derived as shown in equation (2)
Figure BDA0002761724010000031
In the formula xjIs the longitudinal coordinate, y, of the lane line in the vehicle coordinate systemjThe lateral coordinates of the lane line under a vehicle coordinate system are obtained;
defining the type of each lane line acquired by the sensing module as sjJ is ll, l, r, rr, where sjTake 1 or 0, define s when the lane line type is a dotted linej1, when the lane line type is a non-dotted line sj＝0；
Step two, judging the driving scene of the main vehicle:
the driving scenes under the structured road can be divided into four types, the first type of driving scenes are three-lane driving scenes, the main vehicle runs on a middle lane, the lane of the main vehicle and the left and right adjacent lanes can run, and even if the number of the lanes is more than three, the influence of vehicles on the main vehicle on the other lanes except the lane where the main vehicle is located and the adjacent lane where the main vehicle is located is small and is not considered, so that the key obstacles closest to the main vehicle in each lane in the first type of driving scenes need to be screened; the second type of driving scene is a double-lane scene, the main vehicle runs on a left lane, only key obstacles of a lane where the main vehicle is located and an adjacent lane on the right side of the main vehicle need to be screened, and the left side of the lane where the main vehicle is located is considered to be an undrivable area; the third type of driving scene is a double-lane scene, the main vehicle runs on a right lane, only key obstacles of the lane where the main vehicle is located and the adjacent left lane of the main vehicle need to be screened, and the right side of the lane where the main vehicle is located is an undrivable area; the fourth type is a single-lane driving scene, only key obstacles of the lane need to be screened, and the left and right of the lane where the main vehicle is located are non-drivable areas;
before the key obstacles are screened, the driving scene of the main vehicle needs to be judged, the driving scene of the main vehicle is judged mainly according to the types of two lane lines next to the left side and the right side of the main vehicle, and s is obtained when the left lane line and the right lane line are both dotted linesl＝1,sr1, the driving scene of the main vehicle is the first driving scene; when only the right lane line is broken, i.e. sl＝0,sr1, the driving scene of the main vehicle is the second driving scene; when only the left lane line is broken, i.e. sl＝1,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the third driving scene; when the left and right sides are not broken lines, i.e. sl＝0,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the fourth driving scene;
step three, judging the lane where the obstacle is located:
on the basis of judging a driving scene, positioning all obstacles on lanes, and positioning the obstacles on the lanes mainly by comparing relative position relations between the obstacles and the lanes, firstly defining lane numbers under different scenes, wherein the lane number where the main vehicle is located is 1, the adjacent lane on the left side of the main vehicle is 2, the adjacent lane on the right side of the main vehicle is 3, when the obstacles do not belong to the lane of the main vehicle and the two adjacent lanes on the left and right of the main vehicle, the lane number where the obstacles belong is considered to be 0, the obstacles are not considered to represent the obstacles, and after the lane where the obstacles are located is judged, assigning the last line of a matrix OBJ as a corresponding lane number value;
1) when the vehicle is in a first type of driving scene, lane judgment needs to be carried out on n obstacles respectively, the number of the obstacles is i, and the longitudinal coordinate x of the obstacle is usediThe displacement of four lane lines from left to right at the longitudinal coordinate of the obstacle can be obtained by taking the formula (2) into the positionlli,yli,yri,yrriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
2) when the obstacles are in a second type of driving scene, the lane judgment of the n obstacles needs to be carried out respectively, and because the left sides of the adjacent left lane lines of the main vehicle are all non-drivable areas in the second type of driving scene, only whether the obstacles belong to the main vehicle lane or the adjacent right lane of the main vehicle lane needs to be judged, namely the obstacles numbered i, and the longitudinal coordinates x of the obstacles are used for judging whether the obstacles belong to the main vehicle lane or the adjacent right lane of the main vehicle laneiTaking in (2), and obtaining the lateral displacement of the lane line of the adjacent main car on the left side at the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the right side at the longitudinal coordinate of the obstacle, wherein the lateral displacement is yli,yri,yrriWhen y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
3) when the obstacles are in a third-row vehicle scene, the lane judgment of the n obstacles needs to be carried out respectively, and the right sides of adjacent right lane lines of the main vehicle are all not in the third-row vehicle sceneThe feasible area is determined by judging whether the obstacle belongs to the main lane or the adjacent left lane of the main lane, the obstacle with the number of i is used, and the longitudinal coordinate x of the obstacle is usediTaking in (2), and acquiring the lateral displacement of the lane line of the adjacent main car on the right side of the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the left side of the longitudinal coordinate of the obstacle, wherein the lateral displacement is ylli,yli,yriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
4) when the obstacles are in a fourth type driving scene, the lane judgment of the n obstacles needs to be carried out respectively, and because the right side of the adjacent right lane line of the main vehicle and the left side of the adjacent left lane line of the main vehicle are non-drivable areas in the fourth type driving scene, only the judgment of whether the obstacles belong to the lane of the main vehicle is needed, namely the obstacles numbered i, and the longitudinal coordinate x of the obstacles is used for judging whether the obstacles belong to the lane of the main vehicleiTaking in (2), and obtaining the lane line lateral displacement of the adjacent main car on the right side and the lane line lateral displacement of the adjacent main car on the left side at the longitudinal coordinate of the obstacle, wherein the lateral displacements are respectively yli,yriWhen y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
step four, screening key obstacles:
the key obstacles can directly influence the main vehicle driving auxiliary strategy and the main vehicle unmanned system decision and control strategy development, the lane where the obstacles are located is judged, and the key obstacles are screened on the basis of the step four;
1) when the vehicle is in the first type of driving scene, the lane where the main vehicle is located and the main vehicle need to be respectively matchedScreening key obstacles of left and right lanes adjacent to the vehicle, firstly extracting obstacles of left lanes adjacent to the main vehicle, namely extracting the weight of the last behavior 2 of the matrix OBJ to recombine into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is composed of a column with the last behavior 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent left lane of the main car, and the matrix OBJl+And OBJl-As shown in formula (3):
Figure BDA0002761724010000061
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in rear, and the barrier information closest to the main vehicle in front of the adjacent left lane of the main vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is the last row 1 and 2 obstacles of the matrix OBJThe longitudinal coordinates of the objects are positive columns which represent obstacles in the lane of the main vehicle and are positioned in front of the main vehicle, OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (4):
Figure BDA0002761724010000071
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent right lane of the main vehicle and is positioned in front of the main vehicle, and the OBJr-Is composed of a matrix OBJ with the last behavior 3 and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent right lane of the main car, and the matrix OBJr+And OBJr-As shown in(5) Shown in the figure:
Figure BDA0002761724010000081
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent right lane of the main vehicle; defining a key obstacle matrix as obj, wherein the key obstacle matrix represents a key obstacle in front of a left lane adjacent to the main vehicle, a key obstacle behind a left lane adjacent to the main vehicle, a key obstacle in front of a lane where the main vehicle is located, a key obstacle behind a lane where the main vehicle is located, a key obstacle in front of a right lane adjacent to the main vehicle, and key obstacle information behind a right lane adjacent to the main vehicle from left to right, and when the key obstacle at the corresponding position does not exist, the column corresponding to the matrix obj is zero, so that the key obstacle matrix obj [ obj ] finally output in the first type of driving scenel+ objl- objh+ objh- objr+objr-]；
2) When the vehicle is in a second type of driving scene, key obstacles of a lane where the main vehicle is located and a lane on the adjacent right side of the main vehicle need to be screened respectively, the left side of the lane where the main vehicle is located is an undrawable area, firstly, the last line of the matrix OBJ is extractedRecombining the columns of 1 into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (6):
Figure BDA0002761724010000091
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent right lane of the main vehicle and is positioned in front of the main vehicle, and the OBJr-Is formed by the last row 3 and the 2 nd row of obstacles of the matrix OBJThe array with negative coordinates represents an obstacle behind the host vehicle in the adjacent right lane of the host vehicle, and the matrix OBJr+And OBJr-As shown in formula (7):
Figure BDA0002761724010000101
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent right lane of the main vehicle; the final output key obstacle matrix obj ═ 00 obj in the second type of driving sceneh+ objh- objr+ objr-]；
3) When the vehicle is in a third driving scene, key obstacles of a lane where the main vehicle is located and a left lane adjacent to the main vehicle need to be screened respectively, firstly, obstacles of the left lane adjacent to the main vehicle are extracted, namely, the weight of the last behavior 2 of the matrix OBJ is extracted to be recombined into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is formed by the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negativeRepresents an obstacle behind the host vehicle in the adjacent left lane of the host vehicle, and a matrix OBJl+And OBJl-As shown in formula (8):
Figure BDA0002761724010000111
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in rear, and the barrier information closest to the main vehicle in front of the adjacent left lane of the main vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (9):
Figure BDA0002761724010000121
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ obj in the third type of driving scenel+ objl- objh+objh- 0 0]；
4) When the vehicle is in a fourth driving scene, only key obstacles of a lane where the main vehicle is located need to be screened, firstly, the weight of the last behavior 1 of the matrix OBJ is extracted, and two new matrices OBJ are formedh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in equation (10):
Figure BDA0002761724010000131
wherein n ish+Is a matrixNumber of columns, n, of which the last line of OBJ is 1 and the longitudinal coordinate of the obstacle in line 2 is positiveh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ 00 obj in the fourth driving sceneh+ objh- 0 0]；
Step five, outputting key obstacle information:
and outputting the key obstacle matrix obj information obtained in the fourth step, namely finishing the screening and outputting of the information of the key obstacles.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a method for screening key obstacles under a structured road, which provides key obstacle screening based on a target-level sensing fusion result, so that key obstacle information can be provided for the development of an intelligent driving auxiliary system and an unmanned driving system; because the method designs the scene classification method based on the types of the left lane line and the right lane line adjacent to the main vehicle, the division of the main vehicle driving scene under the structured road can be realized; because the method provides the method for screening the key obstacles by positioning the obstacles on the lane and then comparing the relative longitudinal distance between the obstacles and the main vehicle, the method can traverse all the obstacle information acquired by sensing and finally realize the accurate screening of the key obstacle information.
Drawings
The invention is further described with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of a method for screening key obstacles under a structured road according to the present invention;
FIG. 2 is a schematic view of a vehicle coordinate system according to the present invention;
FIG. 3 is a schematic view illustrating the classification of driving scenes according to the present invention;
Detailed Description
The invention is described in detail below with reference to the attached drawing figures:
the invention provides a method for screening key obstacles under a structured road, which is characterized in that a driving scene of a main vehicle is judged based on a driving scene judgment mode, and the key obstacles which have the most obvious influence on the driving of the main vehicle are screened out based on an obstacle list output by a sensing module, wherein the key obstacles are the obstacles of each lane closest to the main vehicle, so that the driving assistance strategy of the main vehicle and the decision and control strategy development of an unmanned system of the main vehicle are facilitated, and the method comprises the following specific steps:
step one, defining required sensing information:
judging a driving scene and screening obstacles in a structured road environment, wherein a lane line under the structured road is clearly visible, and screening key obstacles is carried out in a vehicle coordinate system, so that a vehicle coordinate system is defined, the origin of coordinates of the vehicle coordinate system is at the center of mass o of the main vehicle, the x axis faces the advancing direction of the main vehicle, and the positive direction of the y axis is the direction of anticlockwise rotation of the x axis by 90 degrees, as shown in FIG. 2;
the key obstacle screening is based on single or multi-sensing fused target obstacle sensing information and lane line information, so that the sensing information required by the key obstacle screening is defined, firstly, the total number of obstacles output by a sensing module is defined as n, and the coordinate (x) of each obstacle relative to a main vehicle is defined asi,yi) I-1, …, n, where i represents the number of the obstacle, and assigns the information of all obstacles into an m × n matrix, the first three rows of which are the number i of the obstacle and the longitudinal coordinate of the obstacle relative to the host vehiclexiAnd lateral coordinate y of the obstacle relative to the host vehicleiThe fourth row of the matrix to the m-1 row of the matrix may be defined by a user to output other information of the obstacle, the last row of the matrix, i.e., the m-th row, is reserved as a marker for locating the obstacle in the lane, when the last row of the matrix before the obstacle is not located is assigned to be 0, the obstacle matrix is named as OBJ, and each column of the matrix represents all information of the corresponding obstacle, as shown in detail below:
Figure BDA0002761724010000151
defining lane line information acquired by the main vehicle, wherein the lane line information acquired by the main vehicle comprises four lane line identifications which are named as ll, l, r and rr from left to right in sequence, the lane line information acquired by the main vehicle is a cubic interpolation function, and the interpolation parameter information acquired by the lane line is dj,cj,bj,ajWhere the subscript j represents the parameters of the associated identified lane line, the lane line equation can be derived as shown in equation (2):
Figure BDA0002761724010000161
in the formula xjIs the longitudinal coordinate, y, of the lane line in the vehicle coordinate systemjThe lateral coordinates of the lane line under a vehicle coordinate system are obtained;
defining the type of each lane line acquired by the sensing module as sjJ is ll, l, r, rr, where sjTake 1 or 0, define s when the lane line type is a dotted linej1, when the lane line type is a non-dotted line sj＝0；
Step two, judging the driving scene of the main vehicle:
the driving scenes under the structured roads can be divided into four types, as shown in fig. 3, the first type of driving scene is a three-lane driving scene, the main vehicle runs on a middle lane, the main vehicle lane and the left and right adjacent lanes can run, even if the number of the lanes is more than three, the influence of vehicles on the main vehicle on the other lanes except the lane where the main vehicle is located and the adjacent lane where the main vehicle is located is small, and the vehicles are not considered, so that the key obstacles closest to the main vehicle in each lane in the first type of driving scene need to be screened; the second type of driving scene is a double-lane scene, the main vehicle runs on a left lane, only key obstacles of a lane where the main vehicle is located and an adjacent lane on the right side of the main vehicle need to be screened, and the left side of the lane where the main vehicle is located is considered to be an undrivable area; the third type of driving scene is a double-lane scene, the main vehicle runs on a right lane, only key obstacles of the lane where the main vehicle is located and the adjacent left lane of the main vehicle need to be screened, and the right side of the lane where the main vehicle is located is an undrivable area; the fourth type is a single-lane driving scene, only key obstacles of the lane need to be screened, and the left and right of the lane where the main vehicle is located are non-drivable areas;
before the key obstacles are screened, the driving scene of the main vehicle needs to be judged, the driving scene of the main vehicle is judged mainly according to the types of two lane lines next to the left side and the right side of the main vehicle, and s is obtained when the left lane line and the right lane line are both dotted linesl＝1,sr1, the driving scene of the main vehicle is the first driving scene; when only the right lane line is broken, i.e. sl＝0,sr1, the driving scene of the main vehicle is the second driving scene; when only the left lane line is broken, i.e. sl＝1,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the third driving scene; when the left and right sides are not broken lines, i.e. sl＝0,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the fourth driving scene;
step three, judging the lane where the obstacle is located:
on the basis of judging a driving scene, positioning all obstacles on lanes, and positioning the obstacles on the lanes mainly by comparing relative position relations between the obstacles and the lanes, firstly defining lane numbers under different scenes, wherein the lane number where the main vehicle is located is 1, the adjacent lane on the left side of the main vehicle is 2, the adjacent lane on the right side of the main vehicle is 3, when the obstacles do not belong to the lane of the main vehicle and the two adjacent lanes on the left and right of the main vehicle, the lane number where the obstacles belong is considered to be 0, the obstacles are not considered to represent the obstacles, and after the lane where the obstacles are located is judged, assigning the last line of a matrix OBJ as a corresponding lane number value;
1) when the vehicle is in a first type of driving scene, lane judgment needs to be carried out on n obstacles respectively, the number of the obstacles is i, and the longitudinal coordinate x of the obstacle is usediThe displacement of four lane lines from left to right at the longitudinal coordinate of the obstacle can be obtained by taking the formula (2) into the positionlli,yli,yri,yrriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
2) when the obstacles are in a second type of driving scene, the lane judgment of the n obstacles needs to be carried out respectively, and because the left sides of the adjacent left lane lines of the main vehicle are all non-drivable areas in the second type of driving scene, only whether the obstacles belong to the main vehicle lane or the adjacent right lane of the main vehicle lane needs to be judged, namely the obstacles numbered i, and the longitudinal coordinates x of the obstacles are used for judging whether the obstacles belong to the main vehicle lane or the adjacent right lane of the main vehicle laneiTaking in (2), and obtaining the lateral displacement of the lane line of the adjacent main car on the left side at the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the right side at the longitudinal coordinate of the obstacle, wherein the lateral displacement is yli,yri,yrriWhen y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
3) when the obstacle is in a third-row vehicle scene, the lane where the n obstacles are respectively located needs to be judgedAnd determining that the right sides of the adjacent right lane lines of the main vehicle are all non-travelable areas in the third driving scene, so that only whether the obstacles belong to the main vehicle lane or the adjacent left lane of the main vehicle lane is judged, the obstacles with the numbers of i are numbered, and the longitudinal coordinates x of the obstacles are calculated according to the longitudinal coordinates x of the obstaclesiTaking in (2), and acquiring the lateral displacement of the lane line of the adjacent main car on the right side of the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the left side of the longitudinal coordinate of the obstacle, wherein the lateral displacement is ylli,yli,yriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
4) when the obstacles are in a fourth type driving scene, the lane judgment of the n obstacles needs to be carried out respectively, and because the right side of the adjacent right lane line of the main vehicle and the left side of the adjacent left lane line of the main vehicle are non-drivable areas in the fourth type driving scene, only the judgment of whether the obstacles belong to the lane of the main vehicle is needed, namely the obstacles numbered i, and the longitudinal coordinate x of the obstacles is used for judging whether the obstacles belong to the lane of the main vehicleiTaking in (2), and obtaining the lane line lateral displacement of the adjacent main car on the right side and the lane line lateral displacement of the adjacent main car on the left side at the longitudinal coordinate of the obstacle, wherein the lateral displacements are respectively yli,yriWhen y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
step four, screening key obstacles:
the key obstacles can directly influence the main vehicle driving auxiliary strategy and the main vehicle unmanned system decision and control strategy development, the lane where the obstacles are located is judged, and the key obstacles are screened on the basis of the step four;
1) when the vehicle is in a first type of driving scene, key obstacles of a lane where the main vehicle is located and adjacent left and right lanes of the main vehicle need to be screened respectively, firstly, obstacles of the adjacent left lane of the main vehicle are extracted, namely, the weight of the last behavior 2 of the matrix OBJ is extracted to be recombined into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is composed of a column with the last behavior 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent left lane of the main car, and the matrix OBJl+And OBJl-As shown in formula (3):
Figure BDA0002761724010000191
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in rear, and the barrier information closest to the main vehicle in front of the adjacent left lane of the main vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (4):
Figure BDA0002761724010000201
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent right lane of the main vehicle and is positioned in front of the main vehicle, and the OBJr-Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle of the 2 nd row as negative and represents that the main vehicle is in the adjacent right laneObstacles behind the main vehicle, matrix OBJr+And OBJr-As shown in formula (5):
Figure BDA0002761724010000211
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent right lane of the main vehicle; defining a key obstacle matrix as obj, wherein the key obstacle matrix represents a key obstacle in front of a left lane adjacent to the main vehicle, a key obstacle behind a left lane adjacent to the main vehicle, a key obstacle in front of a lane where the main vehicle is located, a key obstacle behind a lane where the main vehicle is located, a key obstacle in front of a right lane adjacent to the main vehicle, and key obstacle information behind a right lane adjacent to the main vehicle from left to right, and when the key obstacle at the corresponding position does not exist, the column corresponding to the matrix obj is zero, so that the key obstacle matrix obj [ obj ] finally output in the first type of driving scenel+ objl- objh+ objh- objr+objr-]；
2) When the vehicle is in a second type of driving scene, key obstacles of the lane where the main vehicle is located and the adjacent right lane of the main vehicle need to be screened respectively, and the main vehicle needs to be driven to move in the second type of driving sceneThe left side of the lane where the vehicle is located is an undroppable area, firstly, the column weight of the last behavior 1 of the matrix OBJ is extracted to be recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (6):
Figure BDA0002761724010000221
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle of the 2 nd row as positive, and represents the obstacle in the adjacent right lane of the main vehicle and in front of the main vehicleArticle, OBJr-Is composed of a matrix OBJ with the last behavior 3 and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent right lane of the main car, and the matrix OBJr+And OBJr-As shown in formula (7):
Figure BDA0002761724010000231
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent right lane of the main vehicle; the final output key obstacle matrix obj ═ 00 obj in the second type of driving sceneh+ objh- objr+ objr-]；
3) When the vehicle is in a third driving scene, key obstacles of a lane where the main vehicle is located and a left lane adjacent to the main vehicle need to be screened respectively, firstly, obstacles of the left lane adjacent to the main vehicle are extracted, namely, the weight of the last behavior 2 of the matrix OBJ is extracted to be recombined into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is composed of a column with the last behavior 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent left lane of the main car, and the matrix OBJl+And OBJl-As shown in formula (8):
Figure BDA0002761724010000241
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in rear, and the barrier information closest to the main vehicle in front of the adjacent left lane of the main vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (9):
Figure BDA0002761724010000251
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ obj in the third type of driving scenel+ objl- objh+objh- 0 0]；
4) When the vehicle is in a fourth driving scene, only key obstacles of a lane where the main vehicle is located need to be screened, firstly, the weight of the last behavior 1 of the matrix OBJ is extracted, and two new matrices OBJ are formedh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in equation (10):
Figure BDA0002761724010000261
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ 00 obj in the fourth driving sceneh+ objh- 0 0]；
Step five, outputting key obstacle information:
and outputting the key obstacle matrix obj information obtained in the fourth step, namely finishing the screening and outputting of the information of the key obstacles.

Claims (1)

1. A method for screening key obstacles under a structured road judges a driving scene of a main vehicle based on a driving scene judgment mode, screens out key obstacles which have the most obvious influence on the driving of the main vehicle based on an obstacle list output by a sensing module, wherein the key obstacles are the obstacles of each lane closest to the main vehicle, and are convenient for the development of a main vehicle driving auxiliary strategy and a main vehicle unmanned system decision and control strategy, and is characterized by comprising the following steps:
step one, defining required sensing information:
judging a driving scene and screening obstacles in a structured road environment, wherein a lane line on the structured road is clearly visible, and screening key obstacles is carried out in a vehicle coordinate system, so that a vehicle coordinate system is defined, the origin of coordinates of the vehicle coordinate system is at the center of mass o of the main vehicle, the x axis faces the advancing direction of the main vehicle, and the positive direction of the y axis is the direction of anticlockwise rotation of the x axis by 90 degrees;
the key obstacle screening is based on single or multi-sensing fused target obstacle sensing information and lane line information, so that the sensing information required by the key obstacle screening is defined, the total number of obstacles output by a sensing module is firstly defined as n, and the coordinate (x) of each obstacle relative to a main vehicle is defined asi,yi) I-1, …, n, where i represents the number of the obstacle, and assigns the information of all obstacles into an m × n matrix, the first three rows of which are sequentially the number i of the obstacle, the longitudinal coordinate x of the obstacle relative to the host vehicleiAnd lateral coordinate y of the obstacle relative to the host vehicleiThe fourth row of the matrix to the m-1 row of the matrix may be defined by a user to output other information of the obstacle, the last row of the matrix, i.e., the m-th row, is reserved as a marker for locating the obstacle in the lane, when the last row of the matrix before the obstacle is not located is assigned to be 0, the obstacle matrix is named as OBJ, and each column of the matrix represents all information of the corresponding obstacle, as shown in detail below:
Figure FDA0002761723000000021
defining lane line information acquired by the main vehicle, wherein the lane line information acquired by the main vehicle comprises four lane line identifications which are named as ll, l, r and rr from left to right in sequence, the lane line information acquired by the main vehicle is a cubic interpolation function, and the interpolation parameter information acquired by the lane line is dj,cj,bj,ajWhere the subscript j represents the parameters of the associated identified lane line, the lane line equation can be derived as shown in equation (2):
Figure FDA0002761723000000022
in the formula xjIs the longitudinal coordinate, y, of the lane line in the vehicle coordinate systemjThe lateral coordinates of the lane line under a vehicle coordinate system are obtained;
defining the type of each lane line acquired by the sensing module as sjJ is ll, l, r, rr, where sjTake 1 or 0, define s when the lane line type is a dotted linej1, when the lane line type is a non-dotted line sj＝0；
Step two, judging the driving scene of the main vehicle:
the driving scenes under the structured road can be divided into four types, the first type of driving scenes are three-lane driving scenes, the main vehicle runs on a middle lane, the lane of the main vehicle and the left and right adjacent lanes can run, and even if the number of the lanes is more than three, the influence of vehicles on the main vehicle on the other lanes except the lane where the main vehicle is located and the adjacent lane where the main vehicle is located is small and is not considered, so that the key obstacles closest to the main vehicle in each lane in the first type of driving scenes need to be screened; the second type of driving scene is a double-lane scene, the main vehicle runs on a left lane, only key obstacles of a lane where the main vehicle is located and an adjacent lane on the right side of the main vehicle need to be screened, and the left side of the lane where the main vehicle is located is considered to be an undrivable area; the third type of driving scene is a double-lane scene, the main vehicle runs on a right lane, only key obstacles of the lane where the main vehicle is located and the adjacent left lane of the main vehicle need to be screened, and the right side of the lane where the main vehicle is located is an undrivable area; the fourth type is a single-lane driving scene, only key obstacles of the lane need to be screened, and the left and right of the lane where the main vehicle is located are non-drivable areas;
before the key obstacles are screened, the driving scene of the main vehicle needs to be judged, the driving scene of the main vehicle is judged mainly according to the types of two lane lines next to the left side and the right side of the main vehicle, and s is obtained when the left lane line and the right lane line are both dotted linesl＝1,sr1, the driving scene of the main vehicle is the first driving scene; when only the right lane line is broken, i.e. sl＝0,sr1, the driving scene of the host vehicle is the second kindA driving scene; when only the left lane line is broken, i.e. sl＝1,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the third driving scene; when the left and right sides are not broken lines, i.e. sl＝0,srWhen the driving scene of the main vehicle is 0, the driving scene of the main vehicle is the fourth driving scene;
step three, judging the lane where the obstacle is located:
on the basis of judging a driving scene, positioning all obstacles on lanes, and positioning the obstacles on the lanes mainly by comparing relative position relations between the obstacles and the lanes, firstly defining lane numbers under different scenes, wherein the lane number where the main vehicle is located is 1, the adjacent lane on the left side of the main vehicle is 2, the adjacent lane on the right side of the main vehicle is 3, when the obstacles do not belong to the lane of the main vehicle and the two adjacent lanes on the left and right of the main vehicle, the lane number where the obstacles belong is considered to be 0, the obstacles are not considered to represent the obstacles, and after the lane where the obstacles are located is judged, assigning the last line of a matrix OBJ as a corresponding lane number value;
1) when the vehicle is in a first type of driving scene, lane judgment needs to be carried out on n obstacles respectively, the number of the obstacles is i, and the longitudinal coordinate x of the obstacle is usediThe displacement of four lane lines from left to right at the longitudinal coordinate of the obstacle can be obtained by taking the formula (2) into the positionlli,yli,yri,yrriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
2) when the obstacles are in a second type of driving scene, the lanes of the n obstacles need to be judged respectively, and because the left sides of the adjacent left lane lines of the main car are all non-drivable areas in the second type of driving scene, whether the obstacles belong to the main car lane or the main car lane is judged onlyAdjacent to the right lane, the number of the obstacle is i, and the longitudinal coordinate x of the obstacle is setiTaking in (2), and obtaining the lateral displacement of the lane line of the adjacent main car on the left side at the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the right side at the longitudinal coordinate of the obstacle, wherein the lateral displacement is yli,yri,yrriWhen y isli≥yi＞yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri≥yi≥yrriIf so, the value of the ith column and the mth row of the OBJ matrix is 3; when y isrri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
3) when the obstacle is in a third-row vehicle scene, the lane of the n obstacles needs to be judged respectively, and because the right sides of the adjacent right lane lines of the main vehicle are all non-drivable areas in the third-row vehicle scene, the obstacle only needs to be judged whether to belong to the main vehicle lane or the adjacent left lane of the main vehicle lane, namely the obstacle numbered i, and the longitudinal coordinate x of the obstacle is used for judging whether the obstacle belongs to the adjacent left lane of the main vehicle lane or not, namely the obstacle numbered iiTaking in (2), and acquiring the lateral displacement of the lane line of the adjacent main car on the right side of the longitudinal coordinate of the obstacle and the lateral displacement of the two lane lines on the left side of the longitudinal coordinate of the obstacle, wherein the lateral displacement is ylli,yli,yriWhen y islli≥yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is assigned to be 2; when y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞ylliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
4) when the obstacles are in a fourth type driving scene, the lanes of the n obstacles need to be judged respectively, and because the right side of the adjacent right lane line of the main vehicle and the left side of the adjacent left lane line of the main vehicle are non-drivable areas in the fourth type driving scene, only whether the obstacles belong to the lanes of the main vehicle or not needs to be judged, namely the obstacles numbered i, and the longitudinal coordinates x of the obstaclesiBrought into (2) to obtain the barrierThe lateral displacement of the lane line of the adjacent main car on the right side and the lateral displacement of the lane line of the adjacent main car on the left side at the longitudinal coordinate of the obstacle are respectively yli,yriWhen y isli≥yi≥yriIf so, the value of the ith column and the mth row of the OBJ matrix is 1; when y isri＞yiOr yi＞yliIf so, the value of the ith column and the mth row of the OBJ matrix is 0; sequentially judging the lanes of the n obstacles according to the mode;
step four, screening key obstacles:
the key obstacles can directly influence the main vehicle driving auxiliary strategy and the main vehicle unmanned system decision and control strategy development, the lane where the obstacles are located is judged, and the key obstacles are screened on the basis of the step four;
1) when the vehicle is in a first type of driving scene, key obstacles of a lane where the main vehicle is located and adjacent left and right lanes of the main vehicle need to be screened respectively, firstly, obstacles of the adjacent left lane of the main vehicle are extracted, namely, the weight of the last behavior 2 of the matrix OBJ is extracted to be recombined into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is composed of a column with the last behavior 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent left lane of the main car, and the matrix OBJl+And OBJl-As shown in formula (3):
Figure FDA0002761723000000051
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the main vehicleThe information of the obstacle closest to the main vehicle in front of the adjacent left lane of the vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (4):
Figure FDA0002761723000000061
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the list are 0, which represents that the main vehicle is not stored in the front of the laneAt the obstacle; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent right lane of the main vehicle and is positioned in front of the main vehicle, and the OBJr-Is composed of a matrix OBJ with the last behavior 3 and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent right lane of the main car, and the matrix OBJr+And OBJr-As shown in formula (5):
Figure FDA0002761723000000071
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the formula are 0 and represent the main formulaNo obstacle exists behind the adjacent right lane; defining a key obstacle matrix as obj, wherein the key obstacle matrix represents a key obstacle in front of a left lane adjacent to the main vehicle, a key obstacle behind a left lane adjacent to the main vehicle, a key obstacle in front of a lane where the main vehicle is located, a key obstacle behind a lane where the main vehicle is located, a key obstacle in front of a right lane adjacent to the main vehicle, and key obstacle information behind a right lane adjacent to the main vehicle from left to right, and when the key obstacle at the corresponding position does not exist, the column corresponding to the matrix obj is zero, so that the key obstacle matrix obj [ obj ] finally output in the first type of driving scenel+ objl- objh+ objh- objr+ objr-]；
2) When the vehicle is in a second driving scene, key obstacles of a lane where the main vehicle is located and key obstacles of a right lane adjacent to the main vehicle need to be screened respectively, the left side of the lane where the main vehicle is located is an undrawable area, firstly, the weight of the last behavior 1 of the matrix OBJ is extracted to be recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (6):
Figure FDA0002761723000000081
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Is a momentArray OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located;
the method for the key obstacle of the adjacent right lane of the main vehicle comprises the following steps: extracting the column of the last action 3 of the matrix OBJ to recombine into two new matrices OBJr+And OBJr-，OBJr+Is composed of a column with the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent right lane of the main vehicle and is positioned in front of the main vehicle, and the OBJr-Is composed of a matrix OBJ with the last behavior 3 and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent right lane of the main car, and the matrix OBJr+And OBJr-As shown in formula (7):
Figure FDA0002761723000000091
wherein n isr+The number of columns, n, being the last row 3 of the matrix OBJ and the longitudinal coordinate of the obstacle being positive in row 2r-The number of columns which are the last behavior 3 of the matrix OBJ and the longitudinal coordinate of the obstacle in the row 2 is negative; the key barriers of the adjacent right lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objr+，objr+Namely the matrix OBJr+Column corresponding to the minimum element in row 2, when OBJr+When it is an empty matrix objr+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent right lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent right lane of the main vehicle is objr-，objr-Namely the matrix OBJr-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJr-When it is an empty matrix objr-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent right lane of the main vehicle; the final output key obstacle matrix obj ═ 00 obj in the second type of driving sceneh+objh- objr+ objr-]；
3) When the vehicle is in a third driving scene, key obstacles of a lane where the main vehicle is located and a left lane adjacent to the main vehicle need to be screened respectively, firstly, obstacles of the left lane adjacent to the main vehicle are extracted, namely, the weight of the last behavior 2 of the matrix OBJ is extracted to be recombined into two new matrices OBJl+And OBJl-，OBJl+Is composed of a column with the last action 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as positive, represents the obstacle in the adjacent left lane of the main vehicle and is positioned in front of the main vehicle, and the OBJ-Is composed of a column with the last behavior 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the adjacent left lane of the main car, and the matrix OBJl+And OBJl-As shown in formula (8):
Figure FDA0002761723000000101
wherein n isl+The number of columns of which the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nl-The number of columns which are the last row 2 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in rear, and the barrier information closest to the main vehicle in front of the adjacent left lane of the main vehicle is objl+，objl+Namely the matrix OBJl+Column corresponding to the minimum value of longitudinal coordinates of obstacles in the middle 2 nd row, when OBJl+When it is an empty matrix objl+All elements in the main vehicle are 0, which represents that no barrier exists in front of the adjacent left lane of the main vehicle; the information of the obstacle nearest to the main vehicle behind the adjacent left lane of the main vehicle is objl-，objl-Namely the matrix OBJl-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJl-When it is an empty matrix objl-All elements in the main vehicle are 0, which represents that no barrier exists behind the adjacent left lane of the main vehicle;
the method for the key obstacle of the lane where the main vehicle is located comprises the following steps: the column of the last behavior 1 of the extracted matrix OBJ is recombined into two new matrices OBJh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in formula (9):
Figure FDA0002761723000000111
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ ob in the third type of driving scenejl+ objl- objh+ objh- 0 0]；
4) When the vehicle is in a fourth driving scene, only key obstacles of a lane where the main vehicle is located need to be screened, firstly, the weight of the last behavior 1 of the matrix OBJ is extracted, and two new matrices OBJ are formedh+And OBJh-，OBJh+Is composed of a column with the longitudinal coordinate of the obstacle of the last row 1 and the 2 nd row of the matrix OBJ as positive, which represents the obstacle in the lane where the main vehicle is positioned and positioned in front of the main vehicle, and the OBJh-Is composed of a column with the last behavior 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row as negative, represents the obstacle behind the main car in the lane where the main car is located, and the matrix OBJh+And OBJh-As shown in equation (10):
Figure FDA0002761723000000121
wherein n ish+The number of columns of which the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in row 2 are positive, nh-The number of columns which are the last row 1 of the matrix OBJ and the longitudinal coordinate of the obstacle in the 2 nd row is negative; the key barriers of the adjacent left lane of the main vehicle are respectively selected from the barrier closest to the main vehicle in front and the barrier closest to the main vehicle in back, and the information of the barrier closest to the main vehicle in front of the lane where the main vehicle is located is objh+，objh+Namely the matrix OBJh+Column corresponding to the minimum element in row 2, when OBJh+When it is an empty matrix objh+All elements in the main vehicle are 0, which represents that no barrier exists in front of the lane where the main vehicle is located; the information of the obstacle nearest to the main vehicle behind the lane where the main vehicle is located is objh-，objh-Namely the matrix OBJh-The column corresponding to the maximum value of the longitudinal coordinate of the obstacle in the middle 2 nd row is used as the OBJh-When it is an empty matrix objh-All elements in the main vehicle are 0, which represents that no barrier exists behind the lane where the main vehicle is located; the final output key obstacle matrix obj ═ 00 obj in the fourth driving sceneh+ objh- 0 0]；
Step five, outputting key obstacle information:
and outputting the key obstacle matrix obj information obtained in the fourth step, namely finishing the screening and outputting of the information of the key obstacles.
CN202011220296.XA 2020-11-05 2020-11-05 Method for screening key obstacles under structured road Active CN112277952B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011220296.XA CN112277952B (en) 2020-11-05 2020-11-05 Method for screening key obstacles under structured road

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011220296.XA CN112277952B (en) 2020-11-05 2020-11-05 Method for screening key obstacles under structured road

Publications (2)

Publication Number Publication Date
CN112277952A CN112277952A (en) 2021-01-29
CN112277952B true CN112277952B (en) 2021-07-23

Family

ID=74351611

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011220296.XA Active CN112277952B (en) 2020-11-05 2020-11-05 Method for screening key obstacles under structured road

Country Status (1)

Country Link
CN (1) CN112277952B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113335310B (en) * 2021-07-21 2021-11-30 新石器慧通（北京）科技有限公司 Decision-based exercise planning method and device, electronic equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101714273B1 (en) * 2015-12-11 2017-03-08 현대자동차주식회사 Method and apparatus for controlling path of autonomous driving system
CN107924624A (en) * 2015-07-28 2018-04-17 日产自动车株式会社 The control method and travel controlling system of travel controlling system
EP3330944A1 (en) * 2015-07-28 2018-06-06 Nissan Motor Co., Ltd. Method for controlling travel control device, and travel control device
CN109801508A (en) * 2019-02-26 2019-05-24 百度在线网络技术（北京）有限公司 The motion profile prediction technique and device of barrier at crossing
CN110614992A (en) * 2018-12-29 2019-12-27 长城汽车股份有限公司 Method and system for avoiding obstacle during automatic driving of vehicle and vehicle
CN110962847A (en) * 2019-11-26 2020-04-07 清华大学苏州汽车研究院（吴江） Lane centering auxiliary self-adaptive cruise trajectory planning method and system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107924624A (en) * 2015-07-28 2018-04-17 日产自动车株式会社 The control method and travel controlling system of travel controlling system
EP3330944A1 (en) * 2015-07-28 2018-06-06 Nissan Motor Co., Ltd. Method for controlling travel control device, and travel control device
KR101714273B1 (en) * 2015-12-11 2017-03-08 현대자동차주식회사 Method and apparatus for controlling path of autonomous driving system
CN110614992A (en) * 2018-12-29 2019-12-27 长城汽车股份有限公司 Method and system for avoiding obstacle during automatic driving of vehicle and vehicle
CN109801508A (en) * 2019-02-26 2019-05-24 百度在线网络技术（北京）有限公司 The motion profile prediction technique and device of barrier at crossing
CN110962847A (en) * 2019-11-26 2020-04-07 清华大学苏州汽车研究院（吴江） Lane centering auxiliary self-adaptive cruise trajectory planning method and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《自主驾驶车辆的预测避障控制》;余如、郭洪艳、陈虹;《信息与控制》;20150215;第44卷(第1期);全文 *

Also Published As

Publication number Publication date
CN112277952A (en) 2021-01-29

Similar Documents

Publication Publication Date Title
US10345822B1 (en) Cognitive mapping for vehicles
CN111442776B (en) Method and equipment for sequential ground scene image projection synthesis and complex scene reconstruction
CN102208036B (en) Vehicle position detection system
CN107150720A (en) Parking aid
CN111552284A (en) Method, device, equipment and medium for planning local path of unmanned vehicle
CN107444405A (en) Vehicle control system, control method for vehicle and wagon control program
DE102020115012A1 (en) User interface for autonomous vehicles with predicted trajectories
CN111301419A (en) Reinforcement learning based method for SAE4 level automated lane change
CN111976741A (en) Apparatus and method for controlling autonomous vehicle
CN102222236A (en) Image processing system and position measurement system
CN107111954A (en) Destination path generating means and travel controlling system
CN110979313B (en) Automatic parking positioning method and system based on space map
CN110726399A (en) Attitude estimation
CN113844451B (en) Driving device control method and device, electronic device and storage medium
DE102020115011A1 (en) User interface for autonomous vehicles with predicted trajectories
CN112277952B (en) Method for screening key obstacles under structured road
CN114566065A (en) Multi-vehicle cooperative lane changing method based on vehicle-road cooperation
US10599146B2 (en) Action-conditioned vehicle control
CN115230719A (en) Driving track planning method and device
US11301701B2 (en) Specific area detection device
JP4003586B2 (en) Inter-vehicle distance measuring device
CN114998886B (en) Vehicle tracking method and device based on radar vision fusion
CN115447607A (en) Method and device for planning a vehicle driving trajectory
DE102022106246A1 (en) DRIVING ASSISTANCE DEVICE
CN114764392A (en) Computer-implemented method and test unit for determining traffic scene similarity values

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
GR01 Patent grant
GR01 Patent grant