WO2022110740A1 - 一种横向规划约束确定方法及装置 - Google Patents

一种横向规划约束确定方法及装置 Download PDF

Info

Publication number
WO2022110740A1
WO2022110740A1 PCT/CN2021/097295 CN2021097295W WO2022110740A1 WO 2022110740 A1 WO2022110740 A1 WO 2022110740A1 CN 2021097295 W CN2021097295 W CN 2021097295W WO 2022110740 A1 WO2022110740 A1 WO 2022110740A1
Authority
WO
WIPO (PCT)
Prior art keywords
lateral
waypoint
vehicle
planning
constraint
Prior art date
Application number
PCT/CN2021/097295
Other languages
English (en)
French (fr)
Inventor
杨绍宇
王新宇
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP21896249.6A priority Critical patent/EP4250048A1/en
Publication of WO2022110740A1 publication Critical patent/WO2022110740A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
    • G08G1/096725—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information generates an automatic action on the vehicle control
    • 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
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B60W30/095—Predicting travel path or likelihood of collision
    • 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
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles
    • B60W60/001—Planning or execution of driving tasks
    • B60W60/0015—Planning or execution of driving tasks specially adapted for safety
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
    • 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
    • B60W2520/00—Input parameters relating to overall vehicle dynamics
    • B60W2520/10—Longitudinal speed
    • 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
    • B60W2552/53—Road markings, e.g. lane marker or crosswalk
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/20—Static objects
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4041—Position
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4042—Longitudinal speed
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4043—Lateral speed
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4044—Direction of movement, e.g. backwards
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/80—Spatial relation or speed relative to objects
    • B60W2554/804—Relative longitudinal speed
    • 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
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/80—Spatial relation or speed relative to objects
    • B60W2554/806—Relative heading
    • 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
    • B60W2720/00—Output or target parameters relating to overall vehicle dynamics
    • B60W2720/12—Lateral speed

Definitions

  • the present application relates to the field of automatic driving, and in particular, to a method and device for determining lateral planning constraints.
  • the development of artificial intelligence makes autonomous driving possible.
  • the key technologies for autonomous driving implementation include mapping and positioning, environmental perception, fusion prediction, decision-making, planning, and underlying control.
  • the planning is mainly for the longitudinal speed planning and the lateral path planning during the driving process of the autonomous vehicle.
  • the lateral path planning when the autonomous vehicle is driving on the real road, the lateral path planning module can plan the autonomous vehicle to avoid a small lane or cross the lane in a large amount according to the traffic environment around the autonomous vehicle and its own state.
  • the avoidance trajectory of avoidance The planning of the avoidance trajectory needs to be carried out within a certain lateral avoidance range, so that the avoidance process of the autonomous vehicle is reasonable and controllable. Therefore, during the driving process of the autonomous vehicle, a reasonable lateral avoidance range is required for lateral path planning.
  • the embodiments of the present application provide a method and device for determining lateral planning constraints, which can dynamically determine lateral planning constraints according to the state and road conditions obtained by the autonomous vehicle during driving.
  • an embodiment of the present application provides a method for determining lateral planning constraints, which is applied to an autonomous driving vehicle; the method includes: acquiring a driving reference trajectory of the autonomous driving vehicle on a first road, the driving reference trajectory including a plurality of waypoints; determining a static lateral planning constraint of the first side of the first waypoint according to the lane line of the first side of the first waypoint in the plurality of waypoints; the first When the actual driving trajectory of the autonomous driving vehicle does not deviate from the driving reference trajectory, according to the driving speed of the autonomous driving vehicle at the first waypoint, the static lateral planning Constraints, the body width of the self-driving vehicle, determine the initial lateral planning constraints of the first side of the self-driving vehicle when the self-driving vehicle is at the first waypoint; at least according to the relationship between the first object and the first waypoint and the initial lateral planning constraint to determine the actual lateral planning constraint of the first side of the autonomous vehicle when the autonomous vehicle is at the first
  • the lateral planning constraints of the waypoints can be initially determined according to the waypoints on the driving reference; the width of the vehicle and obstacles, shrink or release the lateral planning constraints for that waypoint.
  • the lane line on the first side is a changeable lane line;
  • the static lateral planning constraint includes a first soft static lateral planning constraint and a first hard static lateral planning constraint; wherein, the The first soft static lateral planning constraint is less than half the width of the lane where the first waypoint is located, and the first hard static lateral planning constraint is greater than half the width of the lane where the first waypoint is located;
  • the The initial lateral planning constraints include a first soft initial lateral planning constraint and a first hard initial lateral planning constraint; wherein the first soft initial lateral planning constraints are determined by the driving speed of the autonomous vehicle at the first waypoint, The first soft static lateral planning constraint and the body width of the autonomous vehicle are determined; the first hard initial lateral planning constraint is determined by the driving speed of the autonomous vehicle at the first waypoint, the first A hard static lateral planning constraint, the body width of the autonomous vehicle is determined.
  • two kinds of lateral planning constraints can be determined, wherein the cost of lateral displacement of the ego vehicle within the range of soft lateral planning constraints is smaller than the cost of lateral deviation within the range of hard lateral planning constraints.
  • the ego vehicle can flexibly perform lateral offsets according to soft lateral planning constraints and/or hard lateral planning constraints.
  • the self-vehicle can preferentially perform lateral offset within the range of soft lateral planning constraints; if lateral offset within the range of soft lateral planning constraints does not meet the actual driving requirements (for example, when obstacles cannot be avoided, or when unexpected road conditions cannot be dealt with) ), the lateral offset can be performed within the range of the hard lateral planning constraints.
  • the first soft static lateral planning constraint is obtained by subtracting a first preset value from a half width of the lane where the first waypoint is located, and the first hard static lateral planning constraint The constraint is obtained by adding the first soft static lateral planning constraint to the second preset value.
  • the soft static lateral planning constraint can be set to be less than half of the lane, thereby facilitating the self-vehicle to perform lateral offset at a small cost. Then, on the basis of the soft static lateral planning constraints, the hard static lateral planning constraints are obtained, so that the self-vehicle can have a larger avoidance space when evading.
  • the first waypoint is located in a first lane on the first road, the first object is located in a second lane adjacent to the first lane, and the first A second lane is located on the first side of the first lane;
  • the actual lateral planning constraints include hard actual lateral planning constraints;
  • the initial lateral planning constraints, determining the actual lateral planning constraints of the first side of the autonomous driving vehicle at the first waypoint includes: according to the first distance, the movement speed of the first object, the The movement direction of the first object, the movement speed of the self-driving vehicle, and the movement direction of the self-driving vehicle determine the collision risk between the first object and the self-driving vehicle; when the collision risk is When less than a preset safety threshold, reduce the first hard initial lateral planning constraint so that the first hard initial lateral planning constraint is less than or equal to half the width of the lane where the first waypoint is located; determine to reduce The latter first hard initial lateral planning constraint is the hard actual lateral planning constraint.
  • the width of the lateral planning constraint can be determined according to the collision risk between the own vehicle and the vehicle in the adjacent lane, so as to ensure the driving safety of the own vehicle.
  • the collision risk includes time to collision TTC and/or headway HWT.
  • the first object and the first waypoint are in the same lane, and the automatic driving vehicle gradually approaches the first object during the driving of the automatic driving vehicle;
  • the first The lane line on one side is a lane change line;
  • the first distance includes a first lateral offset between the first object and the first waypoint, and the first lateral offset is the first lateral offset the distance between the object and the first waypoint in a first direction;
  • the actual lateral planning constraints include hard actual lateral planning constraints, the first direction is perpendicular to the autonomous vehicle at the first waypoint
  • the driving direction when the autonomous driving vehicle is at the first waypoint is determined at least according to the first distance between the first object and the first waypoint and the initial lateral planning constraint
  • the actual lateral planning constraints on the first side include: adding a first difference and the width of the autonomous vehicle to obtain a first sum; the first difference is subtracted from the first lateral offset by the The first soft initial lateral planning constraint is obtained; when the first sum ⁇ the lateral expansion width of the first side, or
  • the self-vehicle can bypass the obstacle ahead. If it can be bypassed, the hard lateral planning constraints are expanded so that the ego vehicle can bypass the obstacles ahead by lateral offset.
  • the first object and the first waypoint are in the same lane, and the automatic driving vehicle gradually approaches the first object during the driving of the automatic driving vehicle;
  • the first The lane line on one side is a changeable lane line, and the lane line on the opposite side of the first side is a non-change lane line;
  • the first distance includes the first distance between the first object and the first waypoint.
  • the first lateral offset is the distance between the first object and the first waypoint in the first direction;
  • the actual lateral planning constraints include hard actual lateral planning constraints, the the first direction is perpendicular to the driving direction of the autonomous driving vehicle at the first waypoint; the said at least according to the first distance between the first object and the first waypoint and the initial lateral planning constraint,
  • Determining the actual lateral planning constraints on the first side of the autonomous vehicle when the autonomous vehicle is at the first waypoint includes: adding the first difference and the width of the autonomous vehicle to obtain a first sum; The first difference is obtained by subtracting the first soft initial lateral planning constraint from the first lateral offset; when the first sum > the lateral expansion width of the first side, or, the first When the sum of the sum and the preset first safety distance > the lateral expansion width, the first hard initial lateral planning constraint is determined as the hard actual lateral planning constraint.
  • the method further includes: determining a second distance between the first waypoint and the first object in a driving direction when the autonomous driving vehicle is located at the first waypoint , and determine the first length that the autonomous driving vehicle travels within the first duration; when the third distance is less than the first length, determine that the third distance is the autonomous driving vehicle at the first waypoint When the third distance is greater than the first length, the first length is determined as the longitudinal planning constraint when the autonomous vehicle is at the first waypoint; wherein, the The third distance is equal to the second distance, or the third distance is obtained by subtracting a preset second safety distance from the second distance; when the autonomous driving vehicle is located at the first waypoint, according to The road condition information within the longitudinal planning constraint range determines the driving strategy of the autonomous vehicle.
  • the longitudinal planning constraint range can be determined.
  • it can only consider the objects within the vertical planning constraints, which can avoid the phenomenon that the self-vehicle is still unable to pass the obstacle in front of it after evading by mistake or avoiding it.
  • the automatic driving is determined according to the traveling speed of the automatic driving vehicle at the first waypoint, the static lateral planning constraint, and the body width of the automatic driving vehicle
  • the initial lateral planning constraints on the first side of the vehicle when the vehicle is at the first waypoint includes: when the travel speed is ⁇ a preset first speed threshold, determining that the initial lateral planning constraints are equal to the static lateral planning constraints; when the traveling speed ⁇ a preset second speed threshold, determine that the initial lateral planning constraint is equal to one-half of the vehicle body width; when the first speed threshold ⁇ the traveling speed ⁇ the At the second speed threshold, the initial lateral planning constraint is determined according to the speed-lateral planning constraint curve according to the traveling speed; wherein, on the lateral planning constraint-speed curve, the size of the speed and the size of the lateral planning constraint negatively correlated.
  • the width of the lateral planning constraints can be narrowed when the vehicle speed is high, thereby improving the safety of vehicle driving; and the lateral planning constraints can also be expanded when the vehicle speed is low width, thereby increasing the lateral movement range of the vehicle.
  • the method further includes: when the autonomous driving vehicle performs an avoidance action, according to the initial lateral planning constraints of the first side of the second waypoint in the plurality of waypoints, An actual lateral planning constraint for the first side of the current position of the autonomous vehicle is determined; wherein the autonomous vehicle performs the avoidance action from the second waypoint.
  • the first side of the current position of the autonomous driving vehicle is determined according to the initial lateral planning constraints of the first side of the second waypoint in the plurality of waypoints
  • the actual lateral planning constraints include: when the current speed of the autonomous vehicle ⁇ the speed of the autonomous vehicle at the second waypoint, determining the actual lateral direction of the first side of the second waypoint
  • the planning constraint is the actual lateral planning constraint on the first side of the current position; when the current speed of the autonomous vehicle ⁇ the speed of the autonomous vehicle at the second waypoint, according to the current Speed, static lateral planning constraints on the first side of the second waypoint, and body width of the autonomous vehicle, determine the actual lateral planning constraints on the first side of the current location.
  • the width of the lateral planning constraint of the self-vehicle is locked so that it does not increase with the increase of the vehicle speed, thereby improving the driving safety of the self-vehicle;
  • the width of the lateral planning constraint of the ego car can be increased, thereby increasing the lateral movement space of the ego car.
  • the method further includes: when the autonomous driving vehicle performs a lane change action from the third lane to the fourth lane, changing the static state of the first side of the autonomous driving vehicle
  • the lateral planning constraint gradually changes from a first static lateral planning constraint to a second static lateral planning constraint; wherein the first static lateral planning constraint is determined by a lane line on the first side of the third lane, and the first static lateral planning constraint is Two static lateral planning constraints are determined by the lane lines on the first side of the fourth lane.
  • the lateral planning constraints of the ego vehicle can be gradually transitioned from the lateral planning constraints determined by the original lane to the lateral planning constraints determined by the target lane.
  • the method further includes: within the range of the actual lateral planning constraints, controlling the lateral displacement of the first side of the autonomous driving vehicle at the first waypoint,
  • the lateral displacement is a displacement in a first direction
  • the first direction is perpendicular to the driving direction of the autonomous vehicle at the first waypoint.
  • the self-vehicle can perform lateral displacement within the determined lateral planning constraints, which can increase the passability of the self-vehicle while ensuring the driving safety of the self-vehicle.
  • a device for determining lateral planning constraints which is configured in an autonomous vehicle; the device includes an acquisition unit configured to acquire a driving reference trajectory of the autonomous vehicle on a first road, the driving reference trajectory including a plurality of waypoints; a first determination unit configured to determine a static lateral direction of the first side of the first waypoint according to the lane line of the first side of the first waypoint in the plurality of waypoints planning constraints; the first side is the left side or the right side; a second determining unit is configured to, when the actual driving trajectory of the autonomous driving vehicle does not deviate from the driving reference trajectory, determine whether the autonomous driving vehicle The driving speed at the first waypoint, the static lateral planning constraint, the body width of the autonomous vehicle, and determining the initial lateral planning constraint of the first side of the autonomous vehicle at the first waypoint ; a third determining unit, configured to determine all the positions of the autonomous vehicle at the first waypoint at least according to the first distance between the first object and the first waypoint and the initial
  • the lane line on the first side is a changeable lane line;
  • the static lateral planning constraint includes a first soft static lateral planning constraint and a first hard static lateral planning constraint; wherein, the The first soft static lateral planning constraint is less than half the width of the lane where the first waypoint is located, and the first hard static lateral planning constraint is greater than half the width of the lane where the first waypoint is located;
  • the The initial lateral planning constraints include a first soft initial lateral planning constraint and a first hard initial lateral planning constraint; wherein the first soft initial lateral planning constraints are determined by the driving speed of the autonomous vehicle at the first waypoint, The first soft static lateral planning constraint and the body width of the autonomous vehicle are determined; the first hard initial lateral planning constraint is determined by the driving speed of the autonomous vehicle at the first waypoint, the first A hard static lateral planning constraint, the body width of the autonomous vehicle is determined.
  • the first soft static lateral planning constraint is obtained by subtracting a first preset value from a half width of the lane where the first waypoint is located, and the first hard static lateral planning constraint The constraint is obtained by adding the first soft static lateral planning constraint to the second preset value.
  • the first waypoint is located in a first lane on the first road, the first object is located in a second lane adjacent to the first lane, and the first The second lane is located on the first side of the first lane;
  • the actual lateral planning constraints include hard actual lateral planning constraints;
  • the third determining unit is further configured to: according to the first distance, the first object determine the collision risk between the first object and the autonomous vehicle; When the collision risk is less than a preset safety threshold, the first hard initial lateral planning constraint is reduced, so that the first hard initial lateral planning constraint is less than or equal to a half of the lane where the first waypoint is located one width; determine the reduced first hard initial lateral planning constraint, which is the hard actual lateral planning constraint.
  • the collision risk includes time to collision TTC and/or headway HWT.
  • the first object and the first waypoint are in the same lane, and the automatic driving vehicle gradually approaches the first object during the driving of the automatic driving vehicle;
  • the first The lane line on one side is a lane change line;
  • the first distance includes a first lateral offset between the first object and the first waypoint, and the first lateral offset is the first lateral offset the distance between the object and the first waypoint in a first direction;
  • the actual lateral planning constraints include hard actual lateral planning constraints, the first direction is perpendicular to the autonomous vehicle at the first waypoint
  • the third determining unit is further configured to: add the first difference and the width of the autonomous vehicle to obtain a first sum;
  • the first difference is determined by the first lateral deviation is obtained by subtracting the first soft initial lateral planning constraint; when the first sum ⁇ the lateral expansion width of the first side, or, the first sum and the preset first safety distance are in phase
  • the sum is ⁇ the lateral expansion width
  • the first soft initial lateral planning constraint and the lateral expansion width
  • the first object and the first waypoint are in the same lane, and the automatic driving vehicle gradually approaches the first object during the driving of the automatic driving vehicle;
  • the first The lane line on one side is a changeable lane line, and the lane line on the opposite side of the first side is a non-change lane line;
  • the first distance includes the first distance between the first object and the first waypoint.
  • the first lateral offset is the distance between the first object and the first waypoint in the first direction
  • the actual lateral planning constraints include hard actual lateral planning constraints
  • the first direction is perpendicular to the driving direction of the automatic driving vehicle at the first waypoint
  • the third determining unit is further configured to: add the first difference to the width of the automatic driving vehicle, The first sum is obtained; the first difference is obtained by subtracting the first soft initial lateral planning constraint from the first lateral offset; when the first sum > the lateral expansion width of the first side , or, when the sum of the first sum and the preset first safety distance > the lateral expansion width, the first hard initial lateral planning constraint is determined to be the hard actual lateral planning constraint.
  • the apparatus further includes: a fourth determination unit, configured to determine the relationship between the first waypoint and the first object when the autonomous driving vehicle is located at the first waypoint a second distance in the direction of travel, and determining a first length that the autonomous driving vehicle travels within a first duration; a fifth determining unit, configured to determine the third distance when the third distance is less than the first length The distance is the longitudinal planning constraint when the autonomous vehicle is at the first waypoint; or, when the third distance is greater than the first length, the first length is determined as the location where the autonomous vehicle is located.
  • the longitudinal planning constraint at the first waypoint wherein, the third distance is equal to the second distance, or the third distance is obtained by subtracting a preset second safety distance from the second distance; sixth and a determining unit, configured to determine a driving strategy of the automatic driving vehicle according to the road condition information within the longitudinal planning constraint range when the automatic driving vehicle is located at the first waypoint.
  • the second determining unit is further configured to: when the traveling speed ⁇ a preset first speed threshold, determine that the initial lateral planning constraint is equal to the static lateral planning constraint; When the traveling speed ⁇ the preset second speed threshold, it is determined that the initial lateral planning constraint is equal to one half of the vehicle body width; when the first speed threshold ⁇ the traveling speed ⁇ the second speed When the threshold value, the initial lateral planning constraint is determined according to the speed-lateral planning constraint curve according to the driving speed; wherein, on the lateral planning constraint-speed curve, the magnitude of the speed and the magnitude of the lateral planning constraint are negatively correlated .
  • the apparatus further includes: a seventh determination unit, configured to, when the automatic driving vehicle performs an avoidance action, initial lateral planning constraints on the side, and determining actual lateral planning constraints on the first side of the current position of the autonomous vehicle; wherein the autonomous vehicle starts to perform the avoidance action from the second waypoint.
  • a seventh determination unit configured to, when the automatic driving vehicle performs an avoidance action, initial lateral planning constraints on the side, and determining actual lateral planning constraints on the first side of the current position of the autonomous vehicle; wherein the autonomous vehicle starts to perform the avoidance action from the second waypoint.
  • the seventh determination is further used for: when the current speed of the autonomous driving vehicle is greater than or equal to the speed of the autonomous driving vehicle at the second waypoint, determining the second The actual lateral planning constraint of the first side of the waypoint is the actual lateral planning constraint of the first side of the current position; when the current speed of the autonomous vehicle ⁇ the autonomous vehicle at the second When the speed at the waypoint is used, according to the current speed, the static lateral planning constraints of the first side of the second waypoint, and the body width of the autonomous vehicle, determine the first position of the current position. The actual lateral planning constraints on the side.
  • the apparatus further includes an eighth determination unit, configured to, when the automatic driving vehicle performs a lane change action from the third lane to the fourth lane,
  • the static lateral planning constraints of the first side gradually change from the first static lateral planning constraints to the second static lateral planning constraints; wherein the first static lateral planning constraints are determined by the lanes on the first side of the third lane line determination, the second static lateral planning constraint is determined by the lane line of the first side of the fourth lane.
  • the apparatus further includes a control unit, configured to control the first side of the autonomous driving vehicle at the first waypoint within the range of the actual lateral planning constraints
  • the lateral displacement is a displacement in a first direction, and the first direction is perpendicular to the driving direction of the autonomous vehicle at the first waypoint.
  • an embodiment of the present application provides an apparatus for determining lateral planning constraints of an autonomous vehicle, including: a memory for storing a program; a processor for executing the program stored in the memory; When the program is executed, the processor executes the method described in the first aspect or any possible implementation manner of the first aspect.
  • an embodiment of the present application provides an automatic driving vehicle, including the apparatus for determining lateral planning constraints described in the second aspect.
  • embodiments of the present application provide a computer-readable storage medium, where the computer-readable medium stores instructions for execution by a computing device, and when the computing device executes the instructions, the first aspect or the first aspect is implemented.
  • the method described in any possible implementation manner In one aspect, the method described in any possible implementation manner.
  • the embodiments of the present application provide a computer program product including instructions, which, when the computer program product is run on a computing device, enables the computing device to execute the first aspect or any possible implementation of the first aspect. method described.
  • an embodiment of the present application provides a chip, the chip includes a processor and a data interface, the processor reads an instruction stored in a memory through the data interface, and executes the first aspect or the first aspect method in any possible implementation of .
  • the method and device for determining lateral planning constraints provided by the embodiments of the present application can dynamically release or reduce lateral planning constraints according to the traffic environment, thereby enhancing the passability of the own vehicle while ensuring the driving safety of the own vehicle.
  • FIG. 1 is a functional block diagram of an automatic driving vehicle provided by an embodiment of the present application
  • FIG. 2 is an architecture diagram of an automatic driving system provided by an embodiment of the present application
  • FIG. 3 is a flowchart of a method for determining a horizontal planning constraint provided by an embodiment of the present application
  • Fig. 4 is a kind of lateral planning constraint-velocity curve diagram provided by the embodiment of the present application.
  • FIG. 5 is a schematic diagram of a scenario where the method for determining a horizontal planning constraint provided by an embodiment of the present application can be applied;
  • FIG. 6 is a schematic diagram of a scenario where the method for determining lateral planning constraints provided by the embodiment of the present application can be applied;
  • FIG. 7 is a schematic diagram of a scenario where the method for determining a horizontal planning constraint provided by an embodiment of the present application can be applied;
  • FIG. 8 is a schematic diagram of a scenario where the method for determining a lateral planning constraint provided by an embodiment of the present application can be applied;
  • FIG. 9 is a schematic diagram of a scenario where the method for determining lateral planning constraints provided by the embodiment of the present application can be applied.
  • FIG. 10 is a schematic diagram of a scenario where the method for determining lateral planning constraints provided by the embodiment of the present application can be applied;
  • FIG. 11 is a schematic diagram of a scenario where the method for determining lateral planning constraints provided by the embodiment of the present application can be applied;
  • FIG. 12 is a flowchart of a method for determining a horizontal planning constraint provided by an embodiment of the present application
  • FIG. 13 is a flowchart of a method for determining a horizontal planning constraint provided by an embodiment of the present application.
  • FIG. 14 is a schematic structural diagram of a lateral planning constraint device provided by an embodiment of the present application.
  • FIG. 15 is a schematic structural diagram of a lateral planning constraint device according to an embodiment of the present application.
  • first and second are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature defined as “first” or “second” may expressly or implicitly include one or more of that feature.
  • the terms “including”, “including”, “having” and their variants mean “including but not limited to” unless specifically emphasized otherwise.
  • FIG. 1 shows a functional block diagram of a vehicle 100 provided by an embodiment of the present application.
  • the vehicle 100 may include a computing system 102, an interaction system 104, a propulsion system 106, a sensor system 108, a control system 110, a power source 112, and the like.
  • the structures illustrated in the embodiments of the present application do not constitute a specific limitation on the vehicle 100 .
  • the vehicle 100 may include more or fewer components than shown, or some components may be combined, or some components may be split, or a different arrangement of components.
  • the illustrated components may be implemented in hardware, software, or a combination of software and hardware.
  • the components of the vehicle 100 may be connected together by a system bus (eg, a controller area network bus, CAN bus), a network, and/or other connecting mechanisms so that the components may operate in an interconnected manner.
  • a system bus eg, a controller area network bus, CAN bus
  • Computing system 102 may include processor 1021, memory 1022, and the like.
  • the processor 1021 may include one or more processing units, for example, the processor 110 may include an application processor (application processor, AP), a modem processor, a graphics processor (graphics processing unit, GPU), an image signal processor (image signal processor, ISP), controller, video codec, digital signal processor (digital signal processor, DSP), baseband processor, and/or neural-network processing unit (neural-network processing unit, NPU), etc. Wherein, different processing units may be independent devices, or may be integrated in one or more processors.
  • application processor application processor, AP
  • modem processor graphics processor
  • image signal processor image signal processor
  • ISP image signal processor
  • controller video codec
  • digital signal processor digital signal processor
  • baseband processor baseband processor
  • neural-network processing unit neural-network processing unit
  • the memory 1022 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, universal flash storage (UFS), and the like.
  • Memory 1022 is generally used to store instructions that can be executed on processor 1021 to implement various operations.
  • the instructions include one or more software applications (behavior planner 10221, lateral planner 10222, speed planner 10223, controller 10224, as shown in FIG. 1 , High-precision map 10225, etc., which will be introduced in detail below).
  • Memory 1022 may also be used to store data used and/or generated by the one or more software applications.
  • the processor 1021 can perform various functions described below as well as data processing by executing instructions stored in the memory 1022 .
  • the computing system 102 can be implemented as an in-vehicle intelligent system or an automatic driving system, which can realize the automatic driving of the vehicle 100 (when the vehicle 100 is driving, the vehicle 100 drives completely autonomously without the driver's control or only the driver's small amount of control).
  • Semi-autonomous driving of the vehicle 100 can also be achieved (when the vehicle is running, the vehicle is not fully autonomous and requires moderate control by the driver).
  • the driver can also drive the vehicle 100 manually (the driver controls the vehicle 100 at a high level).
  • Auto-driving, semi-autonomous driving, and manually-controlled vehicles can be set to correspond to different levels of automation.
  • the interaction system 104 may include a wireless communication system 1041, a display screen 1042, a microphone 1043, a speaker 1044, and the like.
  • Wireless communication system 1041 may include one or more antennas, modems, baseband processors, etc., and may communicate with other vehicles and other communication entities.
  • wireless communication systems can be configured to communicate according to one or more communication technologies, such as mobile communication technologies such as 2G/3G/4G/5G, and wireless local area networks (WLAN) (such as wireless True (wireless fidelity, Wi-Fi) network), Bluetooth (bluetooth, BT), global navigation satellite system (global navigation satellite system, GNSS), frequency modulation (frequency modulation, FM), near field communication technology (near field communication, NFC), infrared technology (infrared, IR) and other wireless communication technologies, as well as other communication technologies, will not be listed here.
  • mobile communication technologies such as 2G/3G/4G/5G
  • WLAN wireless local area networks
  • WLAN wireless True (wireless fidelity, Wi-Fi) network
  • Bluetooth blue (wireless fidelity, Wi-Fi) network
  • global navigation satellite system global navigation satellite system
  • frequency modulation frequency modulation, FM
  • NFC
  • the display screen 1042 is used to display images, videos, and the like.
  • Display screen 1042 includes a display panel.
  • the display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode or an active-matrix organic light-emitting diode (active-matrix organic light).
  • LED organic light-emitting diode
  • AMOLED organic light-emitting diode
  • FLED flexible light-emitting diode
  • Miniled MicroLed, Micro-oLed, quantum dot light-emitting diode (quantum dot light emitting diodes, QLED) and so on.
  • the display panel may be covered with a touch panel, and when the touch panel detects a touch operation on or near it, the touch operation may be transmitted to the processor 1021 to determine the type of touch event.
  • Visual output related to touch operations may be provided through display screen 1042 .
  • the touch panel may be located at a different location than the display screen 1042 .
  • the microphone 1043 also called “microphone” or “microphone” is used to convert sound signals into electrical signals.
  • the user wants to control the vehicle 100 by voice, the user can make a voice close to the microphone 1043 through the human mouth, and input the voice command into the microphone 1043 .
  • the vehicle 100 may be provided with at least one microphone 1043 .
  • the vehicle 100 may be provided with two microphones 1043, which can implement a noise reduction function in addition to collecting sound signals.
  • the electronic device 100 may further be provided with three, four or more microphones 1043 to collect sound signals, reduce noise, identify sound sources, and implement directional recording functions.
  • Speakers 1044 also referred to as “speakers”, are used to convert audio electrical signals into sound signals.
  • the vehicle 100 can listen to music through the speaker 1044, or listen to prompt information.
  • Propulsion system 106 may include power components 1061, energy components 1062, transmission components 1063, actuation components 1064, and the like.
  • the power component 1061 may be an engine, which may be any one or a combination of engines such as a gasoline engine, an electric motor of an electric vehicle, a diesel engine, and a hybrid engine, or may be other forms of engines.
  • Energy component 1062 may be a source of energy that powers power component 1061 in whole or in part. That is, the power component 1061 may be configured to convert the energy provided by the energy component 1062 into mechanical energy. Energy sources that can be provided by the energy component 1062 include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electrical power. Energy components 1062 may also include any combination of fuel tanks, batteries, capacitors, and/or flywheels. In some embodiments, energy component 1062 may also provide energy to other systems of vehicle 100 .
  • Transmission components 1063 may include gearboxes, clutches, differentials, propshafts, and other components. When configured, transmission member 1063 can transmit mechanical energy from power member 1061 to actuation member 1064 .
  • Actuating members 1064 may include wheels, tires, and the like. Wheels can be configured in a variety of styles, including unicycle, two-wheeler/motorcycle, tricycle, or four-wheeled car/truck. The tires may be attached to wheels, which may be attached to the transmission member 1063, which may rotate in response to the mechanical power transmitted by the transmission member 1063 to drive the vehicle 100 in action.
  • Sensor system 108 may include positioning components 1081, cameras 1082, inertial measurement units 1083, radar 1084, and the like.
  • the positioning component 1081 may be configured to estimate the position of the vehicle 100 .
  • the positioning component 1081 may include a transceiver configured to estimate the position of the vehicle 100 relative to the earth based on satellite positioning data.
  • the computing system 102 may be configured to use the location component 1081 in conjunction with map data to estimate the road on which the vehicle 100 is likely to travel and the location of the vehicle 100 on the road.
  • the positioning component 1081 may include a global positioning system (global positioning system, GPS) module, may also include a Beidou satellite navigation system (beidou navigation satellite system, BDS), or may include a Galileo satellite navigation system (galileo satellite navigation system), and many more.
  • Cameras 1082 may include exterior cameras configured to capture the environment outside the vehicle 100 , and may include interior cameras configured to capture the environment inside the vehicle 100 .
  • Camera 1082 may be a camera that detects visible light, or light from other parts of the spectrum (infrared or ultraviolet, etc.). Camera 1082 is used to capture two-dimensional images and may also be used to capture depth images.
  • Inertial measurement unit (IMU) 1083 is configured as any combination of sensors that sense changes in position and orientation of vehicle 100 based on inertial acceleration.
  • inertial measurement unit 1083 may include one or more accelerometers and gyroscopes.
  • Radar 1084 may include sensors configured to use light or sound waves to sense or detect objects in the environment in which vehicle 100 is located.
  • the radar 1084 may include laser radar, millimeter-wave radar, or ultrasonic radar.
  • Computing system 102 may determine sensory information of vehicle 100 based on data collected by various components in sensor system 108 .
  • computing system 102 may utilize computer vision computing and sensor fusion techniques to determine sensory information for vehicle 100 based on data collected by one or more components of sensor system 108 .
  • the control system 110 may include a steering component 1101, a braking component 1102, a throttle 1103, and the like.
  • Steering component 1101 may be a component configured to adjust the direction of travel of vehicle 100 in response to driver manipulation or computer commands.
  • the steering component 1101 may include a steering wheel.
  • the throttle valve 1102 may be a component configured to control the operating speed and acceleration of the power component 1061 and thus the speed and acceleration of the vehicle 100 .
  • the braking component 1103 may be a component configured to reduce the travel speed of the vehicle 100 .
  • the braking member 1103 may use friction to slow down the rotational speed of the wheels in the actuating member 1064 .
  • the accelerator 1104 may be a component configured to increase the travel speed of the vehicle 100 .
  • Each component in the control system 110 can perform corresponding operations in response to the control instruction issued by the control module 1024 .
  • the steering wheel in the steering component 1101 can be rotated by a specific angle in response to the control command issued by the control module 1024 .
  • the braking component 1103 may perform braking in response to the control instruction issued by the control module 1024 . Wait, I won't list them one by one here.
  • the power supply 112 may be configured to provide electrical power to some or all of the components of the vehicle 100 .
  • the power source 112 may include a rechargeable lithium-ion battery or a lead-acid battery.
  • the power source 112 may include one or more battery packs.
  • the power source 112 and the energy source part 1062 can be implemented together, and the chemical energy provided by the power source 112 can be converted into the mechanical energy of the motor through the power source part 1061 and transmitted to the actuating part 1064 through the transmission part 1063 to realize the vehicle 100 of driving.
  • an embodiment of the present application provides a method for determining lateral planning constraints.
  • the vehicle 100 can dynamically release or reduce lateral planning constraints according to the vehicle's own information and road condition information during its driving.
  • the vehicle 100 may also be referred to as a self-vehicle.
  • Lateral planning constraints can also be called lateral constraints, which refer to the lateral avoidance range of the ego vehicle. That is to say, the ego vehicle needs to perform lateral displacement within the range or constraints of lateral planning constraints.
  • the lateral direction here refers to the direction perpendicular to or approximately perpendicular to the running direction of the vehicle.
  • the processor 1021 of the vehicle 100 may implement the lateral planning constraint determination method provided by the embodiment of the present application by running the instructions of one or more applications stored in the memory 1022 .
  • the application of the one or more applications will be mainly described. It can be understood that when it is mentioned that an application performs a certain step or has a certain function, it means that the processor 1021 executes the step or realizes the function by running the instructions of the application.
  • the behavior planning module 10221 is responsible for high-level decisions, such as changing lanes or keeping lanes.
  • the horizontal planning module 10222 is responsible for planning the driving trajectory, and the vertical planning module 10223 is responsible for planning the driving speed.
  • the control module 10224 operates the steering component 1101 , the throttle valve 1102 , the braking component 1103 or the accelerator 1104 according to the planned driving trajectory and speed, so that the vehicle can travel according to the planned driving trajectory and speed.
  • the lateral planning module 10222 obtains a decision instruction (eg, keeping a lane) issued by the behavior planning module 10221 , it can perform lateral path planning, and transmit the planned lateral path to the vertical planning module 10223 .
  • the vertical planning module 10223 can perform speed planning according to the lateral path.
  • the control module 10224 may issue control commands to the steering component 1101 , the throttle valve 1102 , the braking component 1103 or the accelerator 1104 according to the planning result of the lateral planning module 10222 and the planning result of the longitudinal planning module 10223 .
  • the lateral planning module 10222 may further include a reference path planning module (reference path planner), an obstacle avoidance module (obstacle avoidance planner), a trajectory smoothing module (trajectory smoother) and a collision detection module.
  • the collision detection module may specifically be a model based collision detection module (model based collision checker).
  • the reference trajectory planning module can determine the driving reference trajectory of the self-vehicle in the time period T1 according to the decision-making instruction issued by the behavior planning module 10221, self-vehicle perception information, map information, etc.
  • the time period T1 may include the current time.
  • the time period T1 is composed of a duration of N seconds before the current moment and a duration of M seconds after the current moment.
  • the time period T1, the value of N, and the value of M can be preset.
  • the driving reference trajectory before the current time is the actual driving trajectory of the self-vehicle or the historical driving trajectory.
  • the driving reference trajectory after the current time is the planned or predicted trajectory.
  • the driving reference trajectory may include a plurality of waypoints located on the driving reference trajectory.
  • the driving reference trajectory consists of connecting lines between adjacent waypoints among the plurality of waypoints.
  • the driving reference trajectory may coincide with the middle line of the lane where the vehicle is located.
  • the reference trajectory planning module may execute the method for determining lateral planning constraints provided by the embodiments of the present application, and output the determined lateral planning constraints to the obstacle avoidance module, so that the obstacle avoidance module, the trajectory smoothing module, and the collision detection module perform a lateral path within the lateral planning constraints. planning.
  • the embodiments of the present application provide solutions for determining the horizontal planning constraints. Therefore, the following will specifically introduce the process of determining the horizontal planning constraints.
  • the process of lateral path planning within lateral planning constraints reference may be made to the introduction of the prior art.
  • lateral refers to a direction perpendicular to the running direction of the vehicle.
  • Longitudinal refers to the direction in which the vehicle travels.
  • the lateral planning constraint determination method may include the steps shown in FIG. 3 . details as follows.
  • the reference trajectory planning module may execute step 301 to obtain the driving reference trajectory B1 of the self-vehicle on the road A1.
  • the reference trajectory planning module can determine the driving reference trajectory B1 according to map information (for example, the map information can be obtained from the high-precision map 10225).
  • the driving reference trajectory B1 may be determined by other applications or components, and then the reference trajectory planning module may obtain the driving reference trajectory B1 from the other applications or components.
  • map information for example, the map information can be obtained from the high-precision map 10225.
  • the driving reference trajectory B1 may be determined by other applications or components, and then the reference trajectory planning module may obtain the driving reference trajectory B1 from the other applications or components.
  • the driving reference track B1 has multiple waypoints, or in other words, the driving reference track is obtained by connecting adjacent waypoints among the multiple waypoints.
  • the reference trajectory planning module may determine lateral planning constraints for waypoints among the plurality of waypoints on the travel trajectory B1. Taking the waypoint B11 among the multiple waypoints as an example, as shown in FIG. 3 , the reference trajectory planning module may execute step 302 to determine the waypoint B11 according to the lane line on the C1 side of the waypoint B11 on the driving reference trajectory B1 The static lateral planning constraint D1 on the C1 side. The C1 side can be left or right. It can be understood that the waypoint B11 is located in the lane, or is located in the virtual lane (for example, the waypoint B11 is located at the intersection, the reference trajectory planning module can determine the virtual lane according to the map information).
  • the lane line on the C1 side of the waypoint B11 refers to a vertical line from the waypoint B11 to the lane line on the C1 side of the lane where the waypoint B11 is located, and the intersection of the vertical line and the lane line on the C1 side of the lane is the waypoint B11 The lane line on the C1 side.
  • the lane lines on both sides of the waypoint B11 may be dashed by default.
  • the dashed line may also be referred to as a lane change line.
  • the self-vehicle can change lanes at the dotted line.
  • the solid line Opposite the dashed line is the solid line, which can also be called the non-replaceable line. According to traffic laws, your own vehicle cannot change lanes at the solid line.
  • static lateral planning constraints D1 may include soft static lateral planning constraints (soft margins) D11 and hard static lateral planning constraints (hard margins) D12. Among them, the width of the hard static lateral planning constraints is greater than or equal to the soft static lateral planning constraints.
  • the soft static lateral planning constraint D11 is less than half the width of the lane where the waypoint B11 is located.
  • the center line of the lane can divide the lane into two single-sided lanes, and the width of each single-sided lane is half the width of the lane. Therefore, in this embodiment of the present application, half of the width of the lane is equal to the width of the one-sided lane, and one-half of the lane may also be referred to as a one-sided lane.
  • the width W1 can be subtracted from the width of the one-sided lane (or half the width of the lane) of the lane where the waypoint B11 is located to obtain the soft static lateral planning constraint D11.
  • the width W1 may be preset, for example, may be 0.3 meters, or 0.4 meters, or 0.5 meters, etc., which will not be listed one by one here.
  • the width of the hard static lateral planning constraint D12 is equal to the width of the soft static lateral planning constraint D12.
  • the soft static lateral planning constraint D11 is less than half the width of the lane where the waypoint B11 is located.
  • the width W1 can be subtracted from the width of the one-sided lane (or half the width of the lane) of the lane where the waypoint B11 is located to obtain the soft static lateral planning constraint D11.
  • the width W1 reference may be made to the above description, and details are not repeated here.
  • the width of the hard static lateral planning constraint D12 is greater than the width of the soft static lateral planning constraint D12.
  • the soft static lateral planning constraint D11 can be added to the width W2 to obtain the hard static lateral planning constraint D12.
  • the width W1 may be preset, for example, may be 0.4 meters, or 0.8 meters, or 1.2 meters, etc., which will not be listed one by one here.
  • the width W2 may be greater than the width W1.
  • the lane line on the C1 side of the waypoint B11 is a dashed line, that is, when the own vehicle can cross or cross the lane line on the C1 side of the waypoint B11, the hard line on the C1 side of the waypoint B11 can be set.
  • the width of the static lateral planning constraint is larger, so that the self-vehicle can borrow the lane on the C1 side of the waypoint B11 to perform lateral displacement, which increases the adjustable range of the lateral displacement.
  • the self-vehicle can be set to drive normally in the rightmost lane, and the current lane width is 3.6 meters.
  • the left side of the lane is a solid line, and lane changes are not allowed.
  • the lane line on the right side of the lane where the vehicle is currently located is a dashed line, and lane changes can be made. From 80m to 120m in front of the vehicle, the dotted line becomes a solid line, and it turns into an area where lane changes cannot be made.
  • the intersection area is 120m ahead of the vehicle.
  • the length of the driving reference trajectory is up to 150 meters.
  • front refers to the direction in which the front of the own vehicle points
  • rear refers to the direction that the rear of the own vehicle points to.
  • the reference trajectory planning module may execute step 303 to determine whether the vehicle is in a lane changing state.
  • the reference trajectory planning module may determine whether the self-vehicle is in a lane-changing state according to the decision instruction it recently received from the behavior planning module 10221 . If the most recent decision instruction received from the behavior planning module 10221 is to change lanes, it can be determined that the vehicle is in a state of lane change. If the decision instruction recently received from the behavior planning module 10221 is to keep the lane, it can be determined that the ego vehicle is not in a lane changing state.
  • the judgment result in step 303 can be set as No, that is, the self-vehicle is not in the lane-changing state.
  • the reference trajectory planning module may also execute step 304 to determine whether the vehicle is in an avoidance state.
  • the driving reference trajectory B1 is not the actual driving trajectory of the own vehicle.
  • the self-vehicle may deviate from the driving reference trajectory B1 due to reasons such as avoiding obstacles. It can be determined whether the closest distance between the current actual position of the vehicle and the driving reference trajectory B1 is greater than the preset threshold Y1, thereby determining whether the actual driving route of the vehicle deviates from the driving reference trajectory B1.
  • the closest distance between the current actual position of the vehicle and the driving reference trajectory B1 is greater than the preset threshold Y1, it can be considered that the actual driving route of the vehicle deviates from the driving reference trajectory B1, and the vehicle is performing the avoidance action, that is, the vehicle is in the avoidance state .
  • the closest distance between the current actual position of the vehicle and the driving reference trajectory B1 is less than or equal to the preset threshold Y1
  • a vertical line can be drawn from the actual position of the own vehicle to the driving reference trajectory B1 to obtain the intersection point between the vertical line and the driving reference trajectory B1, and then the distance between the actual position of the own vehicle and the intersection point can be obtained.
  • the threshold value Y1 may be a preset value, for example, may be 0.2 meters, or 0.3 meters, etc., which will not be listed one by one here.
  • the lateral deviation between the final driving trajectory recently output by the lateral planning module 10222 (eg, output to the longitudinal planning module 10223 ) and the driving reference trajectory B1 can be determined. It can be understood that the lateral planning module 10222 can output the final driving trajectory according to a preset period. For one cycle, the final travel trajectory output by the lateral planning module 10222 is the final travel obtained by further processing the obstacle avoidance module, the trajectory smoothing module and the collision detection module on the basis of the reference trajectory planning module to determine the travel reference trajectory and lateral planning constraints. trajectory.
  • the reference trajectory planning module can acquire the final driving trajectory recently output by the lateral planning module 10222, and compare the lateral deviation between the final driving trajectory and the driving reference trajectory B1.
  • a vertical line can be drawn from any point b1 on the final driving trajectory to the driving reference trajectory B1 to obtain the intersection of the vertical line and the driving reference trajectory B1, and then the distance between the intersection and the point b1 is calculated.
  • the distance from each point on the final driving trajectory to the driving reference trajectory B1 can be obtained. From the distances from each point on the final driving trajectory to the driving reference trajectory B1, the maximum distance is determined, and the maximum distance is used as the lateral deviation between the final driving trajectory and the driving reference trajectory B1.
  • the threshold value Y2 may be a preset value, for example, may be 0.2 meters, or 0.3 meters, etc., which will not be listed one by one here.
  • step 304 it can be determined whether the outgoing vehicle is in an avoidance state.
  • the judgment result in step 304 can be set as No, that is, the own vehicle is not in the avoidance state.
  • the reference trajectory planning module may also perform step 305 to determine the initial lateral planning on the C1 side of the ego vehicle when it is at the waypoint B11 according to the running speed V1 of the ego vehicle at the waypoint B11, the static lateral planning constraint D1, and the body width of the ego car Constrain E1.
  • step 305 when step 305 is executed, the self-vehicle may not actually drive to the waypoint B11.
  • the traveling speed V1 of the self-vehicle at the waypoint B11 in step 305 may be a planned or predicted speed.
  • the longitudinal planning module 10223 can predict the speed of the self-vehicle when it travels to the waypoint B11 according to the perception information of the vehicle, the final driving trajectory recently output by the lateral planning module 10222, and the traffic condition information on the road A1, and is represented by: This results in the travel speed V1.
  • the specific scheme for predicting the speed of the self-vehicle at a certain waypoint may refer to the introduction of the prior art, and details are not repeated here.
  • the travel speed V1 may be inertial filtered to obtain the filtered travel speed V1.
  • the inertial filter may be a first-order inertial filter.
  • the traveling speed V2 of the waypoint B12 on the traveling reference trajectory B1 that is adjacent to the waypoint B11 and located before the waypoint B11 can be acquired.
  • the traveling speed V2 may be the speed of the ego vehicle when it travels to the waypoint B12 predicted by the longitudinal planning module 10223, or may be the actual traveling speed of the ego vehicle at the waypoint B12.
  • inertial filtering may be performed on the traveling speed V1 according to the traveling speed V2 and the inertial filter coefficient.
  • the inertial filter coefficient when the ego vehicle is accelerating can be set to 0.97
  • the inertial filter coefficient when the ego car decelerates can be set to 0.90.
  • travel speed V1 (or filtered travel speed V1 ) may be compared to a preset speed threshold v1 . If the travel speed V1 (or the filtered travel speed V1 ) is less than or equal to the speed threshold v1 , it may be determined that the initial lateral planning constraint E1 is equal to the static lateral planning constraint D1 .
  • the static lateral planning constraints D1 include soft static lateral planning constraints D11 and hard static lateral planning constraints D12.
  • the initial lateral planning constraint E1 includes a soft initial lateral planning constraint E11 and a hard initial lateral planning constraint E12. The initial lateral planning constraint E1 is equal to the static lateral planning constraint D1.
  • the soft initial lateral planning constraint E11 is equal to the soft static lateral planning constraint D11
  • the hard initial lateral planning constraint E12 is equal to the hard static lateral planning constraint D12.
  • the speed threshold v1 may be 20 km/h, or 25 km/h, or 30 km/h, and so on. During specific implementation, the size of the speed threshold v1 can be set according to experience or experiments.
  • travel speed V1 (or filtered travel speed V1 ) may be compared to a preset speed threshold v2. If the traveling speed V1 (or the filtered traveling speed V1 ) is greater than or equal to the speed threshold v2, it can be determined that the initial lateral planning constraint E1 is equal to half of the vehicle body width.
  • the initial lateral planning constraints E1 include soft initial lateral planning constraints E11 and hard initial lateral planning constraints E12.
  • the initial lateral planning constraint E1 is equal to half of the body width of the ego vehicle.
  • the soft initial lateral planning constraint E11 and the hard initial lateral planning constraint E12 are both equal to half the body width of the ego vehicle.
  • the speed threshold v2 is greater than the speed threshold v1.
  • the speed threshold v2 may be 70 km/h, or 75 km/h, 80 km/h, and so on.
  • the size of the speed threshold v2 can be set according to experience or experiments.
  • the initial lateral planning constraint E1 is between the static lateral planning constraint D1 and half the ego vehicle body width , and is negatively correlated with the travel speed V1 (or the filtered travel speed V1).
  • the negative correlation is specifically a negative correlation of a first-order linear change.
  • the initial lateral planning constraint E1 can be determined using the lateral planning constraint-velocity curve shown in FIG. 4 .
  • a coordinate point on the lateral planning constraint-velocity curve whose ordinate is the travel speed V1 (or the filtered travel speed V1) can be determined, and then the abscissa of the coordinate point is used as the initial lateral planning constraint E1.
  • the initial lateral planning constraints E1 include soft initial lateral planning constraints E11 and hard initial lateral planning constraints E12. If the traveling speed V1 (or the filtered traveling speed V1) is greater than the speed threshold v1 and less than the speed threshold v2, the soft initial lateral planning constraint E11 is between the soft static lateral planning constraint D11 and half the width of the vehicle body, and is equal to The travel speed V1 (or the filtered travel speed V1) is negatively correlated; and the hard initial lateral planning constraint E12 is between the hard static lateral planning constraint D12 and half the ego vehicle body width, and is related to the travel speed V1 (or the filtered travel speed V1). The travel speed V1) is negatively correlated.
  • the width of the road A1 can be set to be 3.6 meters, the width W1 to be 0.4 meters, the width W2 to be 1.2 meters, and the width of the self-vehicle body to be 2 meters.
  • the left side of waypoint B11 is a solid line, and the hard static lateral planning constraint 502 to the left of waypoint B11 is also 1.4 meters.
  • the traveling speed V2 is 26.0 km/h
  • the filtered traveling speed V1 is 25.2 km/h. It can be seen that the vehicle is in an accelerating state at the waypoint B1.
  • the inertia filter coefficient when the ego vehicle is accelerating can be set to 0.97.
  • the speed threshold v1 can be set to 20km/h
  • the speed threshold v2 can be set to 70km/h.
  • the first-order linear change is performed, and the soft initial lateral planning constraints and the hard initial lateral planning constraints on the left side of the waypoint B11 are both 1.36 meters, and can be obtained
  • the soft initial lateral planning constraint on the right side of waypoint B11 is 1.36 meters
  • the hard initial lateral planning constraint on the right side is 2.36 meters.
  • the reference trajectory planning module may also execute step 306, at least according to the distance L1 between the object F1 and the waypoint and the initial lateral planning constraint E1, determine when the vehicle is at the waypoint B11
  • the initial lateral planning constraints E1 may include soft initial lateral planning constraints E11 and hard initial lateral planning constraints E12.
  • the actual lateral planning constraint H1 may include a soft actual lateral planning constraint H11 and a hard actual lateral planning constraint H12.
  • the object F1 can be an object that affects the driving process of the vehicle when the vehicle reaches the waypoint B11. Therefore, the vehicle needs to pay attention to the object F1 and adjust the driving strategy, such as the driving speed and the driving route, according to the object F1.
  • the object F1 can also be understood as an obstacle, or in other words, there is a risk of collision between the object F1 and the ego vehicle.
  • the object F1 may specifically be an object that needs attention when the self-vehicle travels to the waypoint B11. That is to say, the object F1 may be an object that needs to be considered when the vehicle travels to the waypoint B11 to avoid collision with the object F1.
  • step 305 is described with examples in combination with different application scenarios.
  • FIG. 6 shows an application scenario, in which the ego vehicle, namely the vehicle 100, is driving on the lane A11 of the road A1.
  • the road A1 also includes a lane A12 located on the right side of the lane A11 and adjacent to the lane A11.
  • the vehicle 200 is traveling on the lane A12. It can be understood that if the vehicle 100 crosses or crosses the lane, the vehicle 100 and the vehicle 200 are at risk of collision, and thus, the lateral planning constraint on the right side of the vehicle 100 needs to be contracted.
  • the distance L1 between the waypoint B11 and the vehicle 200 , the moving speed V3 of the vehicle 200 , the moving direction M1 of the vehicle 200 , the moving speed V4 of the vehicle 100 , and the moving direction M2 of the vehicle 100 can be determined. Then, the collision risk between the vehicle 100 and the vehicle 200 is determined according to the distance L1, the moving speed V3 of the vehicle 200, the moving direction M1 of the vehicle 200, the moving speed V4 of the vehicle 100, and the moving direction M2 of the vehicle 100.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may be reduced so that the reduced hard initial lateral planning constraint E12 is less than or equal to the lane where the waypoint B11 is located (ie half the width of lane A11).
  • the width of the hard initial lateral planning constraint E12 can be reduced to the soft initial lateral planning constraint E11.
  • the reduced hard initial lateral planning constraint E12 is used as the hard actual lateral planning constraint H12.
  • the moving speed V3 of the vehicle 200 may be the predicted moving speed of the vehicle 200 when the vehicle 100 travels to the waypoint B11.
  • the moving direction M1 of the vehicle 200 may be the predicted moving direction of the vehicle 200 when the vehicle 100 travels to the waypoint B11.
  • the moving speed V4 of the vehicle 100 may be the predicted moving speed of the vehicle 100 when the vehicle 100 travels to the waypoint B11.
  • the moving direction M2 may be the predicted moving direction of the vehicle 100 when the vehicle 100 travels to the waypoint B11.
  • the determined collision risk may be time to collision (TTC).
  • TTC time to collision
  • it may be determined that the vehicle 100 and the vehicle 200 are gradually approaching according to the difference between the moving speed V3 and the moving speed V4, and the moving direction of the vehicle 200 and the moving direction of the vehicle 100 .
  • the difference between the distance L1, the moving speed V3 and the moving speed V4 it can be determined how long the vehicle 100 and the vehicle 200 will collide under the condition that the moving direction and the moving speed of the vehicle 100 and the vehicle 200 are kept unchanged. , get the collision time.
  • the collision time between the vehicle 100 and the vehicle 200 may also be determined according to the solution in the prior art, which will not be repeated here.
  • the safety threshold may include a time threshold T1.
  • the time threshold T1 may be referred to as a time-to-collision safety threshold.
  • the time threshold T1 may be 4.5 seconds, or 5 seconds, or 6 seconds, and so on.
  • the developer can set the time threshold T1.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may be reduced.
  • the time to collision is greater than or equal to the time threshold T1
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may not be reduced.
  • the determined collision risk may be head way time (HWT).
  • HWT head way time
  • the lead time distance of the vehicle 100 and the vehicle 200 may be determined according to the distance L1, the movement speed V3, and the movement speed V4.
  • the head-vehicle time distance between the vehicle 100 and the vehicle 200 may be determined according to the solution in the prior art, which will not be repeated here.
  • the safety threshold may include a time threshold T2.
  • the time threshold T2 may be referred to as the lead vehicle headway safety threshold.
  • the time threshold T2 may be 0.4 seconds, or 0.5 seconds, or 0.6 seconds, and so on.
  • the developer can set the time threshold T2.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may be reduced.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may not be reduced.
  • the determined risk of collision may include time to collision and lead time.
  • the safety threshold may include a time threshold T1 and a time threshold T1.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may be reduced when the time to collision is less than the time threshold T1 and/or the lead vehicle time distance is less than the time threshold T2.
  • the hard initial lateral planning constraint E12 on the right side of the vehicle 100 may not be reduced.
  • the right initial hard lateral planning constraint 602 of waypoint B11 is recovered to the width position of the right soft initial lateral planning constraint 601 .
  • the width of the soft initial lateral planning constraint on the right is 1.36 meters
  • the width of the hard initial lateral planning constraint on the right of waypoint B11 is reduced to 1.36 meters.
  • Figure 7 shows another application scenario.
  • the object F1 is in the same lane as the waypoint B11, or the object F1 blocks the lane where the waypoint B11 is located.
  • the vehicle 100 gradually approaches the object F1 during travel. For this purpose, the vehicle 100 needs to plan how to avoid the object F1.
  • the object F1 may be a single obstacle.
  • the object F1 may be an obstacle group composed of multiple obstacles.
  • the multiple obstacles cannot satisfy the kinematic constraints of the vehicle 100, and the vehicle 100 cannot find a trajectory that can pass through the gaps between the multiple obstacles.
  • the plurality of obstacles can be treated as a single obstacle.
  • the lane line on the C1 side of waypoint B11 can be set as a dashed line, which can change lanes.
  • the lateral offset P1 between the object F1 and the waypoint B11 can be determined.
  • the lateral offset P1 is the offset or distance of the C1-side boundary of the object F1 relative to the waypoint B11 in the direction J1.
  • the direction J1 is perpendicular to the direction of travel of the vehicle 100 when the vehicle 100 is located at the waypoint B11.
  • the lateral offset P1 may be the offset or distance of the boundary C1 of the object F1 relative to the waypoint B11 in the direction J1.
  • the boundary of the C1 side may be the farthest point of the C1 side, or the most C1 side (the C1 side is right or left).
  • the actual lateral planning constraint H1 includes a soft actual lateral planning constraint H11 and a hard actual lateral planning constraint H12. Among them, in order to avoid the object F1, the vehicle 100 needs to determine the hard actual lateral planning constraint H12.
  • the plan is as follows.
  • the lateral offset P1 may be subtracted from the soft initial lateral planning constraint on the C1 side of the waypoint B11, that is, the lateral offset P1 may be subtracted from the soft initial lateral planning constraint E11 to obtain the difference Q1.
  • the difference Q1 is added to the body width of the vehicle 100 to obtain the sum S1.
  • the lateral expansion width may refer to the lateral distance that a vehicle can cross the lane line and intrude into the adjacent lane within the scope permitted by the traffic laws.
  • the lateral expansion width may be preset, for example, may be 2.8 meters, or 3 meters, or 3.2 meters, etc., which will not be listed here. If the sum S1 is less than or equal to the lateral expansion width, the soft initial lateral planning constraint on the C1 side of waypoint B11 (ie, the soft initial lateral planning constraint E11) can be added to the lateral expansion width, and the obtained sum can be used as the hard actual Lateral planning constraint H12.
  • the lateral safety distance may be preset.
  • the lateral safety distance represents the allowable lateral error when the vehicle passes through the obstacle, so that the vehicle does not collide with the obstacle when the lateral error within the lateral safety distance occurs.
  • the lateral safety distance represents the distance that the vehicle should keep from the obstacle when it safely passes the obstacle.
  • the safe distance may be 0.5 meters, or 0.6 meters, etc., which will not be listed here. In this example, it can be determined whether the sum of the sum S1 and the safety distance is less than or equal to the lateral expansion width.
  • the soft initial lateral planning constraint on the C1 side of the waypoint B11 ie, the soft initial lateral planning constraint E11
  • the additive sum of can be used as a hard practical lateral planning constraint H12.
  • the soft initial lateral planning constraints on the C1 side of B11 ie, the soft initial lateral planning constraints E11
  • the soft initial lateral planning constraints E11 can be directly used as the soft actual lateral planning constraints.
  • the obstacles that are in front of the vehicle 100 and whose speed is faster than that of the vehicle 100 can be eliminated; the obstacles whose collision time with the vehicle 100 at the waypoint B11 is greater than 4.5 seconds can be eliminated; the driving track and the lane where the vehicle 100 is located Obstacle culling without interference; other vehicles with lateral velocity culling.
  • the remaining obstacles are regarded as obstacles of interest or objects of interest. Calculate the coordinates of the objects of interest on the frenet coordinate system, and sort the objects of interest according to the longitudinal distance from the waypoint B11 from near to farthest. The obstacles can then be clustered according to the kinematic constraints of the vehicle 100 . As shown in FIG.
  • the longitudinal distance between all obstacles in the obstacle group and the waypoint B11 can be set within 20-30 meters, and according to the kinematic constraints of the vehicle 100, the vehicle 100 cannot find a path that can pass through the obstacle group.
  • All the obstacles in the obstacle group shown in Figure 7 are clustered.
  • the clustered obstacles are treated as a single obstacle. It can be set that the starting point of the clustered obstacles relative to the longitudinal coordinates of the waypoint B11 is 23 meters away from the waypoint B11 in the longitudinal direction, and the end point is 28 meters away from the waypoint B11 in the longitudinal direction.
  • the horizontal coordinate starting point of the clustered obstacles relative to the waypoint B11 is -1.5 meters away from the waypoint B11 in the horizontal direction, and the soft initial horizontal planning constraint on the right beyond this time is 0.14 meters (right); the horizontal coordinate The end point of is 1.6 meters (left) from waypoint B11 in the lateral direction.
  • the left side of the lane where waypoint B11 is located is a solid line, and the right side is a dotted line. Since the left side is a solid line area, the vehicle 100 cannot pass on the left side of the clustered obstacles. It can be set that when the vehicle 100 can safely pass through the obstacle, the safety distance between the vehicle 100 and the obstacle laterally needs to be kept at 0.5 meters, and the vehicle body width of the vehicle 100 is 2 meters.
  • the maximum value of the expansion hard lateral planning constraints of the vehicle 100 may be set to 2.8 meters. That is, the lateral expansion width is 2.8 meters.
  • Figure 8 shows yet another application scenario.
  • the object F1 and the waypoint B11 are in the same lane, or the object F1 blocks the lane where the waypoint B11 is located.
  • the vehicle 100 gradually approaches the object F1 during travel.
  • the vehicle 100 needs to plan how to avoid the object F1.
  • the lane line on the C1 side of waypoint B11 can be set as a dashed line, which can change lanes.
  • the other side refers to the opposite side of C1. Specifically, the C1 side is the right side, and the other side is the left side. The C1 side is the left side, then the other side is the right side.
  • the lateral offset P1 between the object F1 and the waypoint B11 can be determined. For details, reference may be made to the introduction of the embodiment shown in FIG. 7 , which will not be repeated here.
  • the actual lateral planning constraint H1 includes a soft actual lateral planning constraint H11 and a hard actual lateral planning constraint H12. Among them, in order to avoid the object F1, the vehicle 100 needs to determine the hard actual lateral planning constraint H12.
  • the plan is as follows.
  • the lateral offset P1 may be subtracted from the soft initial lateral planning constraint on the C1 side of the waypoint B11, that is, the lateral offset P1 may be subtracted from the soft initial lateral planning constraint E11 to obtain the difference Q1.
  • the difference Q1 is added to the body width of the vehicle 100 to obtain the sum S1.
  • the hard initial lateral planning constraint on the C1 side of the waypoint B11 ie, the hard initial lateral planning constraint E12 ) is used as the hard actual lateral planning constraint H12 .
  • the safety distance may be preset. For the safety distance, reference may be made to the above description of the embodiment shown in FIG. 7 , which will not be repeated here. In this example, it can be determined whether the sum of the sum S1 and the safety distance is less than or equal to the lateral expansion width. If the sum of the sum S1 and the safety distance is greater than the lateral expansion width, the hard initial lateral planning constraint on the C1 side of waypoint B11 (ie, the hard initial lateral planning constraint E12) is used as the hard actual lateral planning constraint H12.
  • the soft initial lateral planning constraints on the C1 side of B11 ie, the soft initial lateral planning constraints E11
  • the soft initial lateral planning constraints E11 can be directly used as the soft actual lateral planning constraints.
  • the vehicle 100 can avoid the trajectory of the object F1 when it is located at the waypoint B11 where it cannot be planned.
  • the object F1 directly blocks the traveling direction of the vehicle 100 when it is located at the waypoint B11.
  • the initial lateral planning constraints of the waypoint B11 can be directly output, so that the subsequent planning module no longer considers how to avoid the object F1 when the vehicle 100 is located at the waypoint B11.
  • the vehicle 100 may also plan or determine the longitudinal planning constraints of the waypoint B11 in the case that the object F1 directly blocks the driving direction of the vehicle 100 when it is located at the waypoint B11 .
  • Vertical planning constraints can also be called vertical avoidable length constraints.
  • the driving strategy may include driving trajectory and/or driving speed.
  • the longitudinal distance L2 between the waypoint B11 and the object F1 can be determined.
  • the longitudinal distance L2 is specifically the distance in the travel direction when the vehicle travels to the waypoint B11.
  • the longitudinal distance L2 can be obtained by the frenet coordinate system.
  • the longitudinal distance L2 may be the distance from the ordinate starting point of the object F1 in the frenet coordinate system to the waypoint B11 in the longitudinal direction.
  • the longitudinal direction is the travel direction when the vehicle travels to the waypoint B11.
  • the longitudinal distance L2 can be obtained by the flood filling method.
  • the flood filling method will be introduced in detail below, and will not be repeated here.
  • the travelable length L3 within the time period T3 can be determined.
  • longitudinal distance L2 and length L3 may be compared. If the longitudinal distance L2 is smaller than the length L3, the longitudinal distance L2 may be determined as the longitudinal planning constraint of the waypoint B11. If the longitudinal distance L2 is greater than the length L3, the length L3 may be determined as the longitudinal planning constraint of the waypoint B11.
  • a longitudinal safety distance may be set.
  • the longitudinal safety distance can be preset, for example, it can be 2 meters, 2.5 meters, and the like.
  • the longitudinal distance L4 can be obtained by subtracting the longitudinal safety distance from the longitudinal distance L2.
  • the longitudinal distance L4 and the length L3 can be compared. If the longitudinal distance L4 is smaller than the length L3, the longitudinal distance L4 may be determined as the longitudinal planning constraint of the waypoint B11. If the longitudinal distance L4 is greater than the length L3, the length L3 may be determined as the longitudinal planning constraint of the waypoint B11.
  • the obstacles that are in front of the vehicle 100 and whose speed is faster than that of the vehicle 100 can be eliminated; the obstacles whose collision time with the vehicle 100 at the waypoint B11 is greater than 4.5 seconds can be eliminated; the driving track and the lane where the vehicle 100 is located Obstacle culling without interference; other vehicles with lateral velocity culling.
  • the remaining obstacles are regarded as obstacles of interest or objects of interest. Calculate the coordinates of the objects of interest on the frenet coordinate system, and sort the objects of interest according to the longitudinal distance from the waypoint B11 from near to farthest. The obstacles can then be clustered according to the kinematic constraints of the vehicle 100 . As shown in FIG.
  • the longitudinal distance between all obstacles in the obstacle group and the waypoint B11 can be set within 20-30 meters, and according to the kinematic constraints of the vehicle 100, the vehicle 100 cannot find a path that can pass through the obstacle group.
  • All the obstacles in the obstacle group shown in Figure 7 are clustered.
  • the clustered obstacles are treated as a single obstacle. It can be set that the starting point of the clustered obstacles relative to the longitudinal coordinates of the waypoint B11 is 23 meters away from the waypoint B11 in the longitudinal direction, and the end point is 28 meters away from the waypoint B11 in the longitudinal direction.
  • the horizontal coordinate starting point of the clustered obstacles relative to the waypoint B11 is -2.5 meters away from the waypoint B11 in the lateral direction, and the soft initial lateral planning constraint on the right beyond this time is 1.14 meters (right); clustering
  • the end point of the lateral coordinates of the rear obstacle with respect to the waypoint B11 is 1.6 meters (left side) from the waypoint B11 in the lateral direction.
  • the left side of the lane where waypoint B11 is located is a solid line
  • the right side is a dashed line. Since the left side is a solid line area, the vehicle 100 cannot pass on the left side of the clustered obstacles.
  • the maximum value of the expansion hard lateral planning constraint can be set to 2.8 meters, that is, the lateral expansion width is 2.8 meters. Then, when the hard lateral planning constraints on the right side are expanded, the vehicle 100 cannot plan to pass through the clustered obstacles ahead. In this case, take the hard initial lateral planning constraint on the right as the hard actual lateral planning constraint on the right. That is, the width of the hard actual lateral planning constraint on the right is 2.36 meters.
  • the object F1 can be set as a stationary obstacle, or the obstacles included in the object F1 are stationary obstacles.
  • the hard real lateral planning constraints and stationary obstacles to the right of waypoint B11 can be projected into an occupancy grid map (OGM).
  • OGM occupancy grid map
  • the flood filling method can be used to fill the map edge 60 meters ahead, and the farthest distance that can be filled forward is OGM the maximum value (not shown).
  • the hard lateral planning constraint on the right side of waypoint B11 is not extended, and the flood filling method can be used to fill in the position of the starting point of the clustered obstacle in front of waypoint B11, namely is 23 meters. That is, the determined longitudinal distance L2 is 23 meters.
  • the longitudinal safety threshold can be set to 2 meters, and the maximum distance that can be filled forward is 21 meters. That is, the determined longitudinal distance L4 is 21 meters.
  • the furthest distance expected to avoid stationary obstacles can be calculated from the speed of the vehicle 100 at waypoint B11.
  • the longitudinal planning constraint of the waypoint B11 is taken as small as 21 meters between the longitudinal distance L4 and the length L3.
  • the longitudinal planning constraint for waypoint B11 ends before the stationary obstacle.
  • obstacles outside the longitudinal planning constraints may be ignored, and only obstacles within the longitudinal planning constraints are considered. That is to say, when the vehicle 100 is located at the waypoint 11, the driving strategy of the vehicle 100 is determined according to the road condition information within the longitudinal planning constraint range.
  • the above-mentioned solution for determining the longitudinal planning constraints may be specifically executed by the reference trajectory planning module of the vehicle 100 .
  • the reference trajectory planning module can directly determine the initial lateral planning constraint as the actual lateral planning constraint, and output the obstacle avoidance module.
  • step 303 The solution for determining the horizontal planning constraints is described above when the judgment results of step 303 and step 304 are both negative. Next, the solution for determining the lateral planning constraints in the case that the judgment result in step 303 is yes is introduced.
  • the reference trajectory planning module may execute step 308 to determine the static lateral planning constraint D2 according to the lane line on the C1 side of the lane change starting waypoint, and according to The lane line on the C1 side of the lane-changing end waypoint determines the static lateral planning constraint D3; wherein, during the lane change of the ego vehicle, the static lateral planning constraint on the C1 side of the ego vehicle gradually changes from the static lateral planning constraint D2 to the static lateral planning Constrain D3.
  • the self-vehicle (ie, the vehicle 100 ) can be set to be in a lane-changing state from lane A1 to lane A3 , that is, the self-vehicle can change lanes from lane A1 to lane A3 .
  • the static lateral planning constraints on the C1 side of lane A1 can be determined based on the lane lines on the C1 side of lane A1.
  • the static lateral planning constraints of C1 of lane A3 can also be determined according to the lane lines on the C1 side of lane A3.
  • the way to determine the static lateral planning constraints according to the lane lines can be referred to the above introduction, and will not be repeated here.
  • the static lateral planning constraints on the C1 side of the ego vehicle are gradually changed from the static lateral planning constraints on the C1 side of the lane A1 to the static lateral planning constraints on the C1 side of the lane A3.
  • the reference trajectory planning module can respond to the lane change instruction issued by the behavior planning module 10221, and plan the lane change trajectory from the lane A1 to the lane A3 according to the self-vehicle perception information, map information, etc.
  • the first waypoint deviating from the centerline of lane A1 on the lane-changing trajectory can be determined as the lane-changing start waypoint, and the first waypoint returning to the centerline of lane A3 on the lane-changing trajectory can be determined as the lane-changing end waypoint.
  • the waypoint deviating from the centerline of the lane A1 may refer to a waypoint whose distance from the centerline of the lane A1 is greater than a preset threshold.
  • the preset threshold may be 0.2 meters, 0.25 meters, etc., and may be preset during specific implementation.
  • the waypoint returning to the centerline of lane A3 may mean that the distance from the centerline of lane A3 is less than or equal to the preset threshold.
  • the distance from the waypoint to the centerline may refer to the closest distance from the waypoint to the centerline. For example, you can make a perpendicular line through the waypoint to the centerline, and then use the distance between the waypoint and the foot of the foot as the distance from the waypoint to the centerline.
  • the static lateral planning constraints on the C1 side of the lane-change starting waypoint may be determined according to the lane line on the C1 side of the lane-changing starting waypoint.
  • the static lateral planning constraints for C1 of the lane-change end waypoint can be determined according to the lane line on the C1 side of the lane-change end waypoint. It can be determined that at least a waypoint on the lane-changing trajectory is located between the lane-changing starting waypoint and the lane-changing ending waypoint, and each waypoint in the at least one waypoint is arranged in order from the lane-changing trajectory.
  • the static lateral planning constraints on the C1 side of the starting waypoint are changed to the static lateral planning constraints of C1 at the lane change ending waypoint.
  • the change from the static lateral planning constraint on the C1 side of the starting waypoint to the static lateral planning constraint on the C1 side of the lane change ending waypoint may be a first-order linear change.
  • the width of the lane A1 can be set to be 3.6 meters. From all the waypoints of the lane change trajectory, determine the lane change start waypoint and the lane change end waypoint.
  • the lateral planning constraint is 1.4 m
  • the soft static lateral planning constraints on the left and right sides are both 1.4 meters.
  • the hard static lateral planning constraints on the left gradually shrink from 2.6 m to 1.4 m, and the hard static lateral planning constraints on the right gradually increase from 1.4 m to 2.6 m. Specifically, as shown in Figure 11.
  • the initial lateral planning constraints of the waypoints and the actual lateral planning constraints of the waypoints can be determined according to the static lateral planning constraints of the waypoints.
  • the solution for determining the initial lateral planning constraints and the solution for determining the actual lateral planning constraints can be referred to the above introduction, which will not be repeated here.
  • step 303 is NO
  • step 304 is YES
  • the reference trajectory planning module may also execute step 309 to determine the self-direction according to the initial lateral planning constraints of the avoidance starting waypoint. The actual lateral planning constraints of the vehicle in the avoidance state.
  • the details can be introduced above, and are not repeated here.
  • the actual lateral planning constraints on the C1 side of the ego vehicle in the avoidance state can be determined according to the initial lateral planning constraints on the C1 side of the last waypoint B13 before the ego vehicle enters the avoidance state.
  • the waypoint B13 is a waypoint on the driving reference track B1.
  • the waypoint B13 may be referred to as an avoidance starting waypoint. That is, the own vehicle executes the avoidance action from the waypoint B13.
  • For the determination process of the initial lateral planning constraints on the C1 side of the waypoint B13 reference may be made to the above description of the determination process of the initial lateral planning constraints on the C1 side of the waypoint B11.
  • the initial lateral planning constraint on the C1 side of the waypoint B13 is C1 of the current position of the ego vehicle.
  • the actual lateral planning constraints on the side that is to say, during the avoidance period, if the ego vehicle accelerates, the lateral planning constraints are kept unchanged, so that the lateral planning constraints do not decrease with the acceleration of the vehicle speed.
  • the ego vehicle is determined according to the current speed of the ego vehicle, the static lateral planning constraints of C1 of waypoint B13, and the body width of the ego vehicle.
  • steps 301-306 and steps 308 and 309 are executed by the reference trajectory planning module, and step 307 is executed by the obstacle avoidance module as examples
  • the method for determining the lateral planning constraints provided by the embodiment of the present application is used as an example. An example is given, but the execution subject of each step in the lateral planning constraint determination method is not limited.
  • the steps in the lateral planning constraint determination method may be performed by one or more other applications or components.
  • the lateral planning constraint determination module may be set up separately, and the lateral planning constraint determination module executes each step in the lateral planning constraint determination method.
  • the developer can set the execution body of each step in the horizontal planning constraint determination method provided by the embodiment of the present application.
  • the lateral planning constraints of the vehicle in the normal driving state of the lane can be determined, the lateral planning constraints in the lane-keeping state but in the avoidance state can also be determined, and the lateral planning constraints in the lane-changing state can also be determined.
  • the lateral planning constraint determination method may include the steps shown in FIG. 12 . details as follows.
  • Step S1 obtain surrounding obstacle information, lane line information, current ego vehicle speed and current ego vehicle position information as input, and simultaneously obtain position information of all waypoints of ego vehicle driving reference trajectory.
  • Step S2 according to different current driving states (driving in lane, driving in lane change, driving in intersection), generate soft static lateral planning constraints within the lane of each waypoint on the reference trajectory and hard static lateral constraints that can cross lanes.
  • the waypoint of the reference trajectory is located in the lane-changing area (the measured lane line is a dashed line)
  • the soft-static lateral planning constraint width is smaller than the current lane width
  • the hard-static lateral constraint width on the lane-changing side is greater than the soft-static lateral constraint width of the measurement. planning constraints.
  • the width of the soft static lateral planning constraint is smaller than the current lane width, and the width of the hard static lateral constraint is equal to the soft static lateral planning constraint.
  • the width of the soft static lateral planning constraint is smaller than the current lane width, and the width of the hard static lateral constraint is larger than the soft static lateral planning constraint.
  • the soft-static lateral planning constraint width gradually transitions from the width of the current lane to the width of the target lane, and the hard-static lateral planning constraint is based on the left and right side alignments of the current lane.
  • the rule changes gradually when the reference trajectory waypoint is in the lane-changeable area or the non-lane-changeable area.
  • step S3 according to the current speed of the vehicle, the hard static lateral planning constraints and the soft static lateral constraints are changed with the speed.
  • the hard static lateral planning constraints and soft static lateral constraints are slowly reduced when the vehicle speed increases, and increase rapidly when the vehicle speed decreases, resulting in a hysteresis effect.
  • step S4 the running state of the ego vehicle is judged at this time. If the ego vehicle is evading, then step S5 is performed. If the ego vehicle currently travels according to the reference trajectory without lateral displacement, step S5 is skipped and step S6 is directly executed.
  • Step S5 when the ego vehicle leaves the reference trajectory and starts to avoid or avoid the process, during the acceleration process of the ego vehicle, the width of the locked hard static lateral planning constraint and the soft static lateral constraint is kept unchanged, and the hard static state is enlarged during the deceleration process of the ego vehicle.
  • the width of lateral planning constraints and soft static lateral constraints can only be relaxed but not narrowed, so as to ensure the self-vehicle avoidance space.
  • Step S6 Screen out the social vehicles that are normally traveling in the adjacent lanes of the vehicle. For example, if the TTC from the social vehicle to the current position of the vehicle and the HWT is less than a certain threshold, the side of the social vehicle on each waypoint of the reference trajectory is shrunk.
  • the width of the planning constraints to the soft lateral planning constraints reduces the avoidable range of the ego vehicle.
  • the obstacles of interest that greatly invade the current lane of the ego car are screened, and the hard lateral planning constraints are expanded for the obstacles of interest, so that the ego car can have enough space to plan and pass.
  • the screening of the obstacles of interest is obtained by removing irrelevant obstacles from the obstacles around the vehicle.
  • the types of irrelevant obstacles are, obstacles that are too far behind or in front of the ego vehicle or do not need to be avoided in normal driving, far away from the ego vehicle lane, disjoint obstacles, line-pressing obstacles that cross or block at a large angle, line-pressing obstacles Driving, but at an obstacle that is faster than your vehicle or is further away from your vehicle.
  • the remaining obstacles of interest are sorted from near to far away from the ego vehicle along the SL coordinate system.
  • the obstacles are clustered according to the ego kinematic constraints. Calculate the width of the encroaching lane on both sides of the obstacle group closest to the ego vehicle after clustering. When the expansion of the hard lateral planning constraints can make the ego car plan to pass and there is no collision risk on this side, each road on the reference trajectory is calculated. Point to expand the hard lateral planning constraints on that side.
  • Step S7 project the hard lateral planning constraints and the completely stationary obstacles into the OGM, and use the flood filling method to obtain the longest distance that the ego vehicle can stably avoid.
  • the longest distance to avoid a completely stationary obstacle is calculated, and the distance between this distance and the distance obtained by the flood filling method is taken as the final avoidable length output.
  • the ego vehicle is planning laterally, it will ignore the obstacles after the avoidable length, and only avoid the obstacles before the avoidable length.
  • Step S8 output the reference trajectory and avoidable length with hard lateral planning constraints and soft lateral planning constraints. Give follow-up avoidance planning modules.
  • the method for determining lateral planning constraints can dynamically release or contract lateral planning constraints according to the environment, and when the ego vehicle is running at high speed or the surrounding environment is at risk, the lateral planning constraints are contracted so that the ego car drives in the lane. When a substantial avoidance is required, the lateral planning constraints are released, and the self-vehicle is supported to avoid substantially across the lane. Thereby, the passability of the system can be enhanced while ensuring the driving safety of the self-vehicle.
  • the method for determining lateral planning constraints can also generate a longitudinal avoidable length to deal with static obstacles.
  • the road ahead is blocked by single or multiple static obstacles and the ego vehicle cannot pass, the remaining width Insufficient self-vehicle safety is also outdated, so that the planning module does not avoid stationary obstacles ahead. It can make the lateral planning module ignore the obstacles after the avoidable length, and avoid the phenomenon that the ego vehicle still cannot pass the obstacles in front of it after erroneous avoidance or avoidance.
  • the method for determining the horizontal planning constraints can divide the avoidance constraints into horizontal planning constraints and vertical planning constraints.
  • the lateral constraint controls the avoidance range, which is divided into two parts: the cross-lane hard lateral planning constraint and the inner-lane soft lateral planning constraint.
  • Longitudinal constraints control the look-ahead time to avoid completely static obstacles.
  • the method for determining lateral planning constraints can generate static lateral planning constraints according to road line information, intersection information, and the like.
  • the static lateral planning constraints can be adjusted with the speed of the ego vehicle to avoid the middle lateral planning constraints.
  • the lateral planning constraints decrease slowly with the increase of the ego vehicle speed, and expand rapidly as the ego car speed decreases.
  • the hard lateral planning constraints and the soft lateral planning constraints are locked so that they only expand and do not decrease.
  • the method for determining lateral planning constraints can dynamically reduce lateral constraints for side obstacles, and when a vehicle in an adjacent lane moves toward the own vehicle and meets the conditions, the hard lateral planning constraints on that side are reduced. Dynamically expand the lateral constraints for the obstacles occupying the lane, screen out the obstacles of interest, and perform sorting and clustering. Select the obstacle group closest to the ego vehicle after clustering, and expand the corresponding lateral hard lateral planning constraints when it is safe.
  • the embodiments of the present application provide a method for determining lateral planning constraints.
  • the execution body of the method may be the self-vehicle, that is, the vehicle 100 . More specifically, the method may be implemented by any component of the vehicle 100 having a data processing function, such as the reference trajectory planning module and/or the obstacle avoidance module described above.
  • the method includes the following steps.
  • Step 1301 Obtain a driving reference trajectory of the autonomous vehicle on the first road, where the driving reference trajectory includes a plurality of waypoints.
  • Step 1302 Determine a static lateral planning constraint on the first side of the first waypoint according to the lane line on the first side of the first waypoint in the plurality of waypoints; the first side is the left side or right.
  • Step 1303 when the actual driving trajectory of the self-driving vehicle does not deviate from the driving reference trajectory, according to the driving speed of the self-driving vehicle at the first waypoint, the static lateral planning constraints, the automatic driving The vehicle body width of the driving vehicle is determined, and an initial lateral planning constraint of the first side of the autonomous driving vehicle at the first waypoint is determined.
  • Step 1304 at least according to the first distance between the first object and the first waypoint and the initial lateral planning constraints, determine the position of the first side of the autonomous vehicle when the autonomous vehicle is at the first waypoint.
  • the first object is on the first road and is the object that the autonomous driving vehicle pays attention to.
  • the lane line on the first side is a changeable lane line;
  • the static lateral planning constraint includes a first soft static lateral planning constraint and a first hard static lateral planning constraint; wherein the first soft lateral planning constraint The static lateral planning constraint is less than half the width of the lane where the first waypoint is located, and the first hard static lateral planning constraint is greater than half the width of the lane where the first waypoint is located;
  • the initial lateral planning The constraints include a first soft initial lateral planning constraint and a first hard initial lateral planning constraint; wherein the first soft initial lateral planning constraint is determined by the driving speed of the autonomous vehicle at the first waypoint, the first A soft static lateral planning constraint, the body width of the autonomous vehicle is determined; the first hard initial lateral planning constraint is determined by the driving speed of the autonomous vehicle at the first waypoint, the first hard static The lateral planning constraints and the body width of the self-driving vehicle are determined.
  • the first soft static lateral planning constraint is obtained by subtracting a first preset value from a half width of the lane where the first waypoint is located, and the first hard static lateral planning constraint is determined by the The first soft static lateral planning constraint is obtained by adding the second preset value.
  • the first waypoint is in a first lane on the first road, the first object is in a second lane adjacent to the first lane, and the second lane is in the first side of the first lane;
  • the actual lateral planning constraints include hard actual lateral planning constraints;
  • the initial lateral planning is based on at least a first distance between a first object and the autonomous vehicle Constraints, determining the actual lateral planning constraints of the first side of the autonomous driving vehicle at the first waypoint includes: according to the first distance, the movement speed of the first object, the first object determine the collision risk between the first object and the autonomous vehicle; when the collision risk is less than a preset When the safety threshold is reached, reduce the first hard initial lateral planning constraint so that the first hard initial lateral planning constraint is less than or equal to half the width of the lane where the first waypoint is located;
  • the first hard initial lateral planning constraint is the hard actual lateral planning constraint.
  • the collision risk includes time to collision TTC and/or headway HWT.
  • the first object is in the same lane as the first waypoint, and the autonomous vehicle gradually approaches the first object during driving of the autonomous vehicle;
  • the lane line is a lane change line;
  • the first distance includes a first lateral offset between the first object and the first waypoint, and the first lateral offset is the first object and all distance between the first waypoints in a first direction;
  • the actual lateral planning constraints include hard actual lateral planning constraints, the first direction is perpendicular to the driving of the autonomous vehicle at the first waypoint direction; determining the first side of the autonomous vehicle at the first waypoint according to at least a first distance between the first object and the first waypoint and the initial lateral planning constraint
  • the actual lateral planning constraints include: adding the first difference and the width of the autonomous vehicle to obtain a first sum; the first difference is subtracted from the first lateral offset by the first soft
  • the initial lateral planning constraints are obtained; when the first sum ⁇ the lateral expansion width of the first side, or, the sum of the first sum and the preset first safety
  • the first object is in the same lane as the first waypoint, and the autonomous vehicle gradually approaches the first object during driving of the autonomous vehicle;
  • the lane line is a changeable lane line, and the lane line on the opposite side of the first side is a non-change lane line;
  • the first distance includes a first lateral offset between the first object and the first waypoint.
  • the first lateral offset is the distance between the first object and the first waypoint in the first direction;
  • the actual lateral planning constraints include hard actual lateral planning constraints, and the first direction perpendicular to the driving direction of the autonomous driving vehicle at the first waypoint;
  • determining the The actual lateral planning constraints on the first side when the autonomous vehicle is at the first waypoint includes: adding the first difference and the width of the autonomous vehicle to obtain a first sum; the first The difference is obtained by subtracting the first soft initial lateral planning constraint from the first lateral offset; when the first sum > the lateral expansion width of the first side, or, the first sum and When the sum of the preset first safety distances > the lateral expansion width, the first hard initial lateral planning constraint is determined as the hard actual lateral planning constraint.
  • the method further includes determining a second distance between the first waypoint and the first object in a direction of travel of the autonomous vehicle when the autonomous vehicle is at the first waypoint, and determining a first length that the autonomous driving vehicle travels within a first period of time; when the third distance is less than the first length, determining that the third distance is the longitudinal direction of the autonomous driving vehicle at the first waypoint planning constraints; or, when the third distance is greater than the first length, determine that the first length is a longitudinal planning constraint of the autonomous vehicle at the first waypoint; wherein the third The distance is equal to the second distance, or the third distance is obtained by subtracting a preset second safety distance from the second distance; when the autonomous driving vehicle is located at the first waypoint, according to the longitudinal direction
  • the road condition information within the planning constraints is determined, and the driving strategy of the autonomous driving vehicle is determined.
  • the automatic driving vehicle is determined at the location according to the driving speed of the automatic driving vehicle at the first waypoint, the static lateral planning constraints, and the body width of the automatic driving vehicle.
  • the initial lateral planning constraints on the first side at the first waypoint include: when the traveling speed is less than or equal to a preset first speed threshold, determining that the initial lateral planning constraints are equal to the static lateral planning constraints; When the traveling speed ⁇ the preset second speed threshold, it is determined that the initial lateral planning constraint is equal to one half of the vehicle body width; when the first speed threshold ⁇ the traveling speed ⁇ the second speed When the threshold value, the initial lateral planning constraint is determined according to the speed-lateral planning constraint curve according to the driving speed; wherein, on the lateral planning constraint-speed curve, the magnitude of the speed and the magnitude of the lateral planning constraint are negatively correlated .
  • the method further includes: when the autonomous vehicle performs an avoidance action, determining the an actual lateral planning constraint on the first side of the current position of the autonomous vehicle; wherein the autonomous vehicle performs the avoidance action from the second waypoint.
  • the determining the actual lateral direction of the first side of the current position of the autonomous vehicle according to the initial lateral planning constraints of the first side of the second waypoint of the plurality of waypoints The planning constraints include: when the current speed of the autonomous vehicle is greater than or equal to the speed of the autonomous vehicle at the second waypoint, determining that the actual lateral planning constraint on the first side of the second waypoint is The actual lateral planning constraints on the first side of the current position; when the current speed of the autonomous vehicle ⁇ the speed of the autonomous vehicle at the second waypoint, according to the current speed, the The static lateral planning constraints of the first side of the second waypoint and the body width of the autonomous vehicle are used to determine the actual lateral planning constraints of the first side of the current position.
  • the method further includes constraining the static lateral planning of the first side of the autonomous vehicle when the autonomous vehicle performs a lane change maneuver from a third lane to a fourth lane Gradually changing from a first static lateral planning constraint to a second static lateral planning constraint; wherein the first static lateral planning constraint is determined by the lane lines on the first side of the third lane, the second static lateral planning constraint Planning constraints are determined by lane lines on the first side of the fourth lane.
  • the method further comprises: controlling a lateral displacement of the first side of the autonomous vehicle when the autonomous vehicle is at the first waypoint, within the actual lateral planning constraints, the lateral The displacement is the displacement in a first direction, the first direction being perpendicular to the driving direction of the autonomous vehicle at the first waypoint.
  • the method for determining lateral planning constraints provided by the embodiments of the present application can dynamically release or reduce lateral planning constraints according to the traffic environment, thereby enhancing the passability of the own vehicle while ensuring the driving safety of the own vehicle.
  • an embodiment of the present application provides an apparatus 1400 for determining lateral planning constraints, which may be configured in an autonomous vehicle, such as the vehicle 100 .
  • the apparatus 1400 includes an acquisition unit 1410 , a first determination unit 1420 , a second determination unit 1430 , and a third determination unit 1440 .
  • an obtaining unit 1410 configured to obtain a driving reference trajectory of the autonomous driving vehicle on the first road, where the driving reference trajectory includes a plurality of waypoints;
  • a first determining unit 1420 configured to determine a static lateral planning constraint on the first side of the first waypoint according to the lane line on the first side of the first waypoint in the plurality of waypoints; the The first side is left or right;
  • the second determining unit 140 is configured to, when the actual driving trajectory of the autonomous driving vehicle does not deviate from the driving reference trajectory, according to the driving speed of the autonomous driving vehicle at the first waypoint, the static lateral planning Constraints, the body width of the self-driving vehicle, and determining the initial lateral planning constraints of the first side of the self-driving vehicle when the self-driving vehicle is at the first waypoint;
  • the third determining unit 1440 is configured to determine, at least according to the first distance between the first object and the first waypoint and the initial lateral planning constraint, all the automatic driving vehicles at the first waypoint. the actual lateral planning constraint of the first side; the first object is on the first road and is the object of interest of the autonomous driving vehicle.
  • Each functional unit of the apparatus 1400 may be implemented by the method embodiments described above, for example, the method embodiment shown in FIG. 13 , which will not be repeated here.
  • each terminal includes corresponding hardware structures and/or software modules for executing each function.
  • the present application can be implemented in hardware or a combination of hardware and computer software with the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein. Whether a function is performed by hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.
  • functional modules may be divided into electronic devices and the like according to the method embodiments shown in FIG. 13 or FIG. 3 or FIG. 12 .
  • each functional module may be divided corresponding to each function, or two or more
  • the functions are integrated in a processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. It should be noted that, the division of modules in the embodiments of the present application is schematic, and is only a logical function division, and there may be other division manners in actual implementation.
  • the device for determining lateral planning constraints provided by the embodiments of the present application can dynamically release or reduce lateral planning constraints according to the traffic environment, thereby enhancing the passability of the own vehicle while ensuring the driving safety of the own vehicle.
  • an embodiment of the present application provides an apparatus 1500 for determining lateral planning constraints.
  • the apparatus 1500 may perform the operations performed by the autonomous driving vehicle in the foregoing method embodiments, for example, as shown in FIG. 13 .
  • the apparatus 1500 may include a processor 1510 and a memory 1520 .
  • the memory 1520 stores instructions that are executable by the processor 1510 .
  • the apparatus 1500 may perform the operations performed by the autonomous driving vehicle in the foregoing method embodiments, for example, as shown in FIG. 13 .
  • the processor 1510 may perform data processing operations
  • the transceiver 1530 may perform data transmission and/or reception operations.
  • processor in the embodiments of the present application may be a central processing unit (central processing unit, CPU), and may also be other general-purpose processors, digital signal processors (digital signal processors, DSP), application-specific integrated circuits (application specific integrated circuit, ASIC), field programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof.
  • CPU central processing unit
  • DSP digital signal processors
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • a general-purpose processor may be a microprocessor or any conventional processor.
  • the method steps in the embodiments of the present application may be implemented in a hardware manner, or may be implemented in a manner in which a processor executes software instructions.
  • Software instructions can be composed of corresponding software modules, and software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (programmable rom) , PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically erasable programmable read-only memory (electrically EPROM, EEPROM), registers, hard disks, removable hard disks, CD-ROMs or known in the art in any other form of storage medium.
  • An exemplary storage medium is coupled to the processor, such that the processor can read information from, and write information to, the storage medium.
  • the storage medium can also be an integral part of the processor.
  • the processor and storage medium may reside in an ASIC.
  • the above-mentioned embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • software it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated.
  • the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted over a computer-readable storage medium.
  • the computer instructions can be sent from one website site, computer, server, or data center to another website site by wire (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.) , computer, server or data center.
  • the computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media.
  • the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)), and the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

自动驾驶领域的一种横向规划约束确定方法及装置。该方法包括：获取自动驾驶车辆在第一道路上的行驶参考轨迹，行驶参考轨迹包括多个路点（1301）；根据多个路点中的第一路点的第一侧的车道线，确定第一路点的第一侧的静态横向规划约束（1302）；当自动驾驶车辆的实际行驶轨迹没有偏离行驶参考轨迹时，根据自动驾驶车辆在第一路点时的行驶速度、静态横向规划约束、自动驾驶车辆的车身宽度，确定自动驾驶车辆在第一路点时的第一侧的初始横向规划约束（1303）；至少根据第一对象和第一路点之间的第一距离和初始横向规划约束，确定自动驾驶车辆在第一路点时的第一侧的实际横向规划约束（1304）。该方法可以实现横向规划约束随着交通环境的变化而变化。

Description

一种横向规划约束确定方法及装置
本申请要求于2020年11月24日提交中国专利局、申请号为202011331404.0、申请名称为“一种横向规划约束确定方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及自动驾驶领域，具体涉及一种横向规划约束确定方法及装置。
背景技术
人工智能的发展，使得自动驾驶成为可能。自动驾驶实现的关键技术包括地图及定位、环境感知、融合预测、决策、规划以及底层控制。其中，规划主要是对自动驾驶车辆行驶过程中的纵向速度规划以及横向路径规划。在横向路径规划中，自动驾驶车辆在真实道路上行驶时，横向路径规划模块可根据自动驾驶车辆周围的交通环境以及自身的状态，规划出使自动驾驶车辆进行小幅车道内避让或大幅度跨车道避让的避让轨迹。避让轨迹的规划需要在一定横向避让范围内进行，以便使自动驾驶车辆的避让过程合理、可控。由此，在自动驾驶车辆行驶过程中，需要合理的横向避让范围，以便进行横向路径规划。
发明内容
本申请实施例提供了一种横向规划约束确定方法及装置，可以根据自动驾驶车辆在行驶过程中得自身状态以及路况，动态确定横向规划约束。
第一方面，本申请实施例提供了一种横向规划约束确定方法，应用于自动驾驶车辆；所述方法包括：获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹包括多个路点；根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧；当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束；至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
也就是说，本申请实施例提供的横向规划约束确定方法，可以根据行驶参考上的路点先初步确定该路点的横向规划约束；然后，再根据车辆在该路点时的行驶速度、车辆的车身宽度以及障碍物，收缩或放开该路点的横向规划约束。
在一种可能的实现方式中，所述第一侧的车道线为可换道线；所述静态横向规划约束包括第一软静态横向规划约束和第一硬静态横向规划约束；其中，所述第一软静态横向规 划约束小于所述第一路点所在车道的二分之一宽度，所述第一硬静态横向规划约束大于所述第一路点所在车道的二分之一宽度；所述初始横向规划约束包括第一软初始横向规划约束和第一硬初始横向规划约束；其中，所述第一软初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一软静态横向规划约束、所述自动驾驶车辆的车身宽度确定；所述第一硬初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一硬静态横向规划约束、所述自动驾驶车辆的车身宽度确定。
也就是说，在该实现方式中，可以确定两种横向规划约束，其中，自车在软横向规划约束范围内进行横向偏移的代价小于在硬横向规划约束范围内进行横向偏移的代价。由此，自车可以灵活根据软横向规划约束和/或硬横向规划约束进行横向偏移。例如，自车可以优先在软横向规划约束范围内进行横向偏移；若在软横向规划约束的范围内进行横向偏移不满足实际行驶要求(例如无法避让障碍物时，或者无法应对突发路况时)，可以再在硬横向规划约束的范围进行横向偏移。
在一种可能的实现方式中，所述第一软静态横向规划约束由所述第一路点所在车道的二分之一宽度减去第一预设值得到，所述第一硬静态横向规划约束由所述第一软静态横向规划约束加上第二预设值得到。
也就是说，在该实现方式中，可以设置软静态横向规划约束小于车道的一半，由此，方便自车可以以较小的代价进行横向偏移。然后，在软静态横向规划约束的基础上再加宽，得到硬静态横向规划约束，由此，使得自车在进行避让时可以有较大的避让空间。
在一种可能的实现方式中，所述第一路点位于所述第一道路上的第一车道，所述第一对象位于与所述第一车道相邻的第二车道，且所述第二车道位于所述第一车道的所述第一侧；所述实际横向规划约束包括硬实际横向规划约束；所述至少根据第一对象和所述自动驾驶车辆之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：根据所述第一距离、所述第一对象的运动速度、所述第一对象的运动方向、所述自动驾驶车辆的运动速度、所述自动驾驶车辆的运动方向，确定所述第一对象和所述自动驾驶车辆之间的碰撞风险度；当所述碰撞风险度小于预设的安全阈值时，缩小所述第一硬初始横向规划约束，以使所述第一硬初始横向规划约束小于或等于所述第一路点所在车道的二分之一宽度；确定缩小后的所述第一硬初始横向规划约束，为所述硬实际横向规划约束。
也就是说，在该实现方式中，可以根据自车与相邻车道上的车辆之间的碰撞风险度来确定横向规划约束的宽度，以在保障自车的行驶安全性。
在一种可能的实现方式中，所述碰撞风险度包括碰撞时间TTC和/或头车时距HWT。
在一种可能的实现方式中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向 偏移减去所述第一软初始横向规划约束得到；当所述第一加和≤所述第一侧的横向扩张宽度，或者，所述第一加与和预设的第一安全距离的相加和≤所述横向扩张宽度时，将所述第一软初始横向规划约束和所述横向扩张宽度进行相加，得到第二加和；确定所述第二加和为所述硬实际横向规划约束。
也就是说，在该实现方式中，可以先确定自车是否可以绕过前方障碍物。若可以绕过，则扩张硬横向规划约束，以便自车可以通过横向偏移绕过前方障碍物。
在一种可能的实现方式中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线，所述第一侧的对侧的车道线为不可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；当所述第一加和＞所述第一侧的横向扩张宽度，或者，所述第一加和和预设的第一安全距离的相加和＞所述横向扩张宽度时，确定所述第一硬初始横向规划约束为所述硬实际横向规划约束。
也就是说，在该实现方式中，可以先确定自车是否可以绕过前方障碍物。若不可以绕过，则不扩张硬横向规划约束，使得自车不再试图绕过前方障碍物。
在一种可能的实现方式中，所述方法还包括：确定所述第一路点和所述第一对象在所述自动驾驶车辆位于所述第一路点时的行驶方向上的第二距离，以及确定所述自动驾驶车辆在第一时长内行驶的第一长度；当第三距离小于所述第一长度时，确定所述第三距离为所述自动驾驶车辆在所述第一路点时的纵向规划约束；或者，当所述第三距离大于所述第一长度时，确定所述第一长度为所述自动驾驶车辆在所述第一路点时的纵向规划约束；其中，所述第三距离等于所述第二距离，或者所述第三距离由所述第二距离减去预设的第二安全距离得到；当所述自动驾驶车辆位于所述第一路点时，根据所述纵向规划约束范围内的路况信息，确定所述自动驾驶车辆的行驶策略。
也就是说，在该实现方式中，在自车不可以绕过前方障碍物的情况下，可以确定纵向规划约束范围。自车在规划实际行驶轨迹时，可以仅考虑纵向规划约束范围内的物体，可以避免误避让或避让后自车仍然无法通过前方障碍物的现象。
在一种可能的实现方式中，所述根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束包括：当所述行驶速度≤预设的第一速度阈值时，确定所述初始横向规划约束等于所述静态横向规划约束；当所述行驶速度≥预设的第二速度阈值时，确定所述初始横向规划约束等于二分之一的所述车身宽度；当所述第一速度阈值＜所述行驶速度＜所述第二速度阈值时，根据所述行驶速度，按照速度-横向规划约束曲线，确定所述初始横向规划约束；其中，在所述横向规划约束-速度曲线上，速度的大小和横向规划约束的大小呈负相关。
也就是说，在该实现方式中，可以在车速较高的情况下，缩小横向规划约束的宽度，由此可提高车辆行驶的安全性；还可以在车速较低的情况下，扩大横向规划约束的宽度，由此可增加了自车的横向移动范围。
在一种可能的实现方式中，所述方法还包括：当所述自动驾驶车辆执行避让动作时，根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束；其中，所述自动驾驶车辆从所述第二路点开始执行所述避让动作。
在一种可能的实现方式中，所述根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束包括：当所述自动驾驶车辆的当前速度≥所述自动驾驶车辆在所述第二路点时的速度时，确定所述第二路点的所述第一侧的实际横向规划约束为所述当前位置的所述第一侧的实际横向规划约束；当所述自动驾驶车辆的当前速度＜所述自动驾驶车辆在所述第二路点时的速度时，根据所述当前速度、所述第二路点的所述第一侧的静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述当前位置的所述第一侧的实际横向规划约束。
也就是说，在该实现方式中，若自车在避让时加速，则锁定自车的横向规划约束的宽度，使之不随着车速增加而增加，由此提高自车的行驶安全性；若自车在避让时减速，则可以增加自车的横向规划约束的宽度，由此增加自车的横向移动空间。
在一种可能的实现方式中，所述方法还包括：当所述自动驾驶车辆执行从第三车道到第四车道的换道动作时，将所述自动驾驶车辆的所述第一侧的静态横向规划约束逐渐从第一静态横向规划约束变化到第二静态横向规划约束；其中，所述第一静态横向规划约束由所述第三车道的所述第一侧的车道线确定，所述第二静态横向规划约束由所述第四车道的所述第一侧的车道线确定。
也就是说，在该实现方式中，在自车换道期间时，可以将自车的横向规划约束逐渐从由原车道确定的横向规划约束过渡到由目标车道确定的横向规划约束。
在一种可能的实现方式中，所述方法还包括：在所述实际横向规划约束的范围内，控制所述自动驾驶车辆在所述第一路点时的所述第一侧的横向位移，所述横向位移为第一方向上的位移，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向。
也就是说，在该实现方式中，自车可以在确定出的横向规划约束范围内进行横向位移，可以在保证自车行驶安全性的同时，增加自车的通过性。
第二方面，提供了一种横向规划约束确定装置，配置于自动驾驶车辆；所述装置包括获取单元，用于获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹包括多个路点；第一确定单元，用于根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧；第二确定单元，用于当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束；第三确定单元，用于至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
在一种可能的实现方式中，所述第一侧的车道线为可换道线；所述静态横向规划约束包括第一软静态横向规划约束和第一硬静态横向规划约束；其中，所述第一软静态横向规划约束小于所述第一路点所在车道的二分之一宽度，所述第一硬静态横向规划约束大于所述第一路点所在车道的二分之一宽度；所述初始横向规划约束包括第一软初始横向规划约束和第一硬初始横向规划约束；其中，所述第一软初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一软静态横向规划约束、所述自动驾驶车辆的车身宽度确定；所述第一硬初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一硬静态横向规划约束、所述自动驾驶车辆的车身宽度确定。
在一种可能的实现方式中，所述第一软静态横向规划约束由所述第一路点所在车道的二分之一宽度减去第一预设值得到，所述第一硬静态横向规划约束由所述第一软静态横向规划约束加上第二预设值得到。
在一种可能的实现方式中，所述第一路点位于所述第一道路上的第一车道，所述第一对象位于与所述第一车道相邻的第二车道，且所述第二车道位于所述第一车道的所述第一侧；所述实际横向规划约束包括硬实际横向规划约束；所述第三确定单元还用于：根据所述第一距离、所述第一对象的运动速度、所述第一对象的运动方向、所述自动驾驶车辆的运动速度、所述自动驾驶车辆的运动方向，确定所述第一对象和所述自动驾驶车辆之间的碰撞风险度；当所述碰撞风险度小于预设的安全阈值时，缩小所述第一硬初始横向规划约束，以使所述第一硬初始横向规划约束小于或等于所述第一路点所在车道的二分之一宽度；确定缩小后的所述第一硬初始横向规划约束，为所述硬实际横向规划约束。
在一种可能的实现方式中，所述碰撞风险度包括碰撞时间TTC和/或头车时距HWT。
在一种可能的实现方式中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述第三确定单元还用于：将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；当所述第一加和≤所述第一侧的横向扩张宽度，或者，所述第一加与和预设的第一安全距离的相加和≤所述横向扩张宽度时，将所述第一软初始横向规划约束和所述横向扩张宽度进行相加，得到第二加和；确定所述第二加和为所述硬实际横向规划约束。
在一种可能的实现方式中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线，所述第一侧的对侧的车道线为不可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在所述第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述第三确定单元还用于：将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；当所述第一加和＞所述第一侧的横向扩张宽度，或者，所述第一加和和预设的第一安全距离的相加和＞所述横向扩张宽度时，确定所 述第一硬初始横向规划约束为所述硬实际横向规划约束。
在一种可能的实现方式中，所述装置还包括：第四确定单元，用于确定所述第一路点和所述第一对象在所述自动驾驶车辆位于所述第一路点时的行驶方向上的第二距离，以及确定所述自动驾驶车辆在第一时长内行驶的第一长度；第五确定单元，用于当第三距离小于所述第一长度时，确定所述第三距离为所述自动驾驶车辆在所述第一路点时的纵向规划约束；或者，当所述第三距离大于所述第一长度时，确定所述第一长度为所述自动驾驶车辆在所述第一路点时的纵向规划约束；其中，所述第三距离等于所述第二距离，或者所述第三距离由所述第二距离减去预设的第二安全距离得到；第六确定单元，用于当所述自动驾驶车辆位于所述第一路点时，根据所述纵向规划约束范围内的路况信息，确定所述自动驾驶车辆的行驶策略。
在一种可能的实现方式中，所述第二确定单元还用于：当所述行驶速度≤预设的第一速度阈值时，确定所述初始横向规划约束等于所述静态横向规划约束；当所述行驶速度≥预设的第二速度阈值时，确定所述初始横向规划约束等于二分之一的所述车身宽度；当所述第一速度阈值＜所述行驶速度＜所述第二速度阈值时，根据所述行驶速度，按照速度-横向规划约束曲线，确定所述初始横向规划约束；其中，在所述横向规划约束-速度曲线上，速度的大小和横向规划约束的大小呈负相关。
在一种可能的实现方式中，所述装置还包括：第七确定单元，用于当所述自动驾驶车辆执行避让动作时，根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束；其中，所述自动驾驶车辆从所述第二路点开始执行所述避让动作。
在一种可能的实现方式中，所述第七确定还用于：当所述自动驾驶车辆的当前速度≥所述自动驾驶车辆在所述第二路点时的速度时，确定所述第二路点的所述第一侧的实际横向规划约束为所述当前位置的所述第一侧的实际横向规划约束；当所述自动驾驶车辆的当前速度＜所述自动驾驶车辆在所述第二路点时的速度时，根据所述当前速度、所述第二路点的所述第一侧的静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述当前位置的所述第一侧的实际横向规划约束。
在一种可能的实现方式中，所述装置还包括第八确定单元，用于当所述自动驾驶车辆执行从第三车道到第四车道的换道动作时，将所述自动驾驶车辆的所述第一侧的静态横向规划约束逐渐从第一静态横向规划约束变化到第二静态横向规划约束；其中，所述第一静态横向规划约束由所述第三车道的所述第一侧的车道线确定，所述第二静态横向规划约束由所述第四车道的所述第一侧的车道线确定。
在一种可能的实现方式中，所述装置还包括控制单元，用于在所述实际横向规划约束的范围内，控制所述自动驾驶车辆在所述第一路点时的所述第一侧的横向位移，所述横向位移为第一方向上的位移，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向。
第三方面，本申请实施例提供了一种自动驾驶车辆的横向规划约束确定装置，包括：存储器，用于存储程序；处理器，用于执行所述存储器存储的程序；当所述存储器存储的程序被执行时，所述处理器执行第一方面或第一方面任一可能实现方式所述的方法。
第四方面，本申请实施例提供了一种自动驾驶车辆，包括第二方面所述的横向规划约 束确定装置。
第五方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读介质存储用于计算设备执行的指令，所述计算设备执行所述指令时，实现如第一方面或第一方面任一可能实现方式所述的方法。
第六方面，本申请实施例提供一种包括指令的计算机程序产品，当所述计算机程序产品在计算设备上运行时，使得所述计算设备执行第一方面或第一方面任一可能实现方式所述的方法。
第七方面，本申请实施例提供了提供一种芯片，所述芯片包括处理器与数据接口，所述处理器通过所述数据接口读取存储器上存储的指令，执行第一方面或第一方面的任一可能的实现方式中的方法。
本申请实施例提供的横向规划约束确定方法及装置，可以根据交通环境动态的放开或者缩小横向规划约束，从而可以在保证自车行驶的安全性的同时，增强自车的通过性。
附图说明
图1为本申请实施例提供的一种自动驾驶车辆的功能框图；
图2为本申请实施例提供的一种自动驾驶系统架构图；
图3为本申请实施例提供的一种横向规划约束确定方法流程图；
图4为本申请实施例提供的一种横向规划约束-速度曲线图；
图5为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图6为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图7为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图8为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图9为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图10为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图11为本申请实施例提供的横向规划约束确定方法可应用的一种场景示意图；
图12为本申请实施例提供的一种横向规划约束确定方法流程图；
图13为本申请实施例提供的一种横向规划约束确定方法流程图；
图14为本申请实施例提供的一种横向规划约束装置结构示意图；
图15为本申请实施例提供的一种横向规划约束装置结构示意图。
具体实施方式
下面将结合附图，对本发明实施例中的技术方案进行描述。显然，所描述的实施例仅是本说明书一部分实施例，而不是全部的实施例。
在本说明书的描述中“一个实施例”或“一些实施例”等意味着在本说明书的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此，在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其他一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例，而是意味着“一个或多个但不是所有的实施例”，除非是以其他方式另外特别强调。
其中，在本说明书的描述中，除非另有说明，“/”表示或的意思，例如，A/B可以表 示A或B；本文中的“和/或”仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，在本说明书实施例的描述中，“多个”是指两个或多于两个。
在本说明书的描述中，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个该特征。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”，除非是以其他方式另外特别强调。
图1示出了本申请实施例提供的车辆100的功能框图。
参阅图2，车辆100可以包括计算系统102、交互系统104、推进系统106、传感器系统108、控制系统110、电源112等。
可以理解的是，本申请实施例示意的结构并不构成对车辆100的具体限定。在本申请另一些实施例中，车辆100可以包括比图示更多或更少的部件，或者组合某些部件，或者拆分某些部件，或者不同的部件布置。图示的部件可以以硬件，软件或软件和硬件的组合实现。
可将车辆100的各部件通过系统总线(例如控制器局域网络总线(controller area network bus)，CAN总线)、网络和/或其他连接机构连接在一起，以使各部件可按照互连方式工作。
计算系统102可以包括处理器1021、存储器1022等。
处理器1021可以包括一个或多个处理单元，例如：处理器110可以包括应用处理器(application processor，AP)，调制解调处理器，图形处理器(graphics processing unit，GPU)，图像信号处理器(image signal processor，ISP)，控制器，视频编解码器，数字信号处理器(digital signal processor，DSP)，基带处理器，和/或神经网络处理器(neural-network processing unit，NPU)等。其中，不同的处理单元可以是独立的器件，也可以集成在一个或多个处理器中。
存储器1022可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件，闪存器件，通用闪存存储器(universal flash storage，UFS)等。存储器1022通常用于存储可以在处理器1021上运行以实现各种操作的指令。该指令包括一个或多个软件应用(如图1所示的行为规划模块(behavior planner)10221、横向规划模块(lateral planner)10222、纵向规划模块(speed planner)10223、控制模块(controller)10224、高精地图10225等，具体将在下文进行介绍)的指令。存储器1022还可以用于存储该一个或多个软件应用使用的和/或生成的数据。
处理器1021可通过运行存储在存储器1022的指令，执行下文所述的各种功能以及数据处理。
示例性的，该计算系统102可以实现为车载智能系统或自动驾驶系统，可以实现车辆100的自动驾驶(在车辆100行驶时，车辆100完全自主驾驶，无需驾驶员的控制或仅需驾驶员的少量控制)。也可以实现车辆100的半自动驾驶(在车辆行驶时，车辆非完全自主驾驶，需要驾驶员适度控制)。驾驶员也可以手动驾驶车辆100行驶(驾驶员高度控制车辆100)。可以设定自动驾驶、半自动驾驶、手动控制车辆分别对应不同的自动化级别。
交互系统104可以包括无线通信系统1041、显示屏1042、麦克风1043、扬声器1044 等。
无线通信系统1041可以包括一个或多个天线、调制解调器、基带处理器等，可与其他汽车以及其他通信实体进行通信。一般而言，无线通信系统可以被配置为根据一种或多种通信技术进行通信，例如2G/3G/4G/5G等移动通信技术，以及无线局域网(wireless local area networks，WLAN)(如无线保真(wireless fidelity，Wi-Fi)网络)，蓝牙(bluetooth，BT)，全球导航卫星系统(global navigation satellite system，GNSS)，调频(frequency modulation，FM)，近距离无线通信技术(near field communication，NFC)，红外技术(infrared，IR)等无线通信技术，以及其他通信技术，此处不再一一列举。
显示屏1042用于显示图像，视频等。显示屏1042包括显示面板。显示面板可以采用液晶显示屏(liquid crystal display，LCD)，有机发光二极管(organic light-emitting diode，OLED)，有源矩阵有机发光二极体或主动矩阵有机发光二极体(active-matrix organic light emitting diode的，AMOLED)，柔性发光二极管(flex light-emitting diode，FLED)，Miniled，MicroLed，Micro-oLed，量子点发光二极管(quantum dot light emitting diodes，QLED)等。
在一些实施例中，显示面板上可覆盖有触控面板，当触控面板检测到其上或其附近的触摸操作后，可将触摸操作传递给处理器1021，以确定触摸事件类型。可以通过显示屏1042提供与触摸操作相关的视觉输出。在另一些实施例中，触控面板可以与显示屏1042所处的位置不同。
麦克风1043，也称“话筒”，“传声器”，用于将声音信号转换为电信号。当用户想通过语音控制车辆100时，用户可以通过人嘴靠近麦克风1043发声，将语音命令输入到麦克风1043。车辆100可以设置至少一个麦克风1043。在一些实施例中，车辆100可以设置两个麦克风1043，除了采集声音信号，还可以实现降噪功能。在另一些实施例中，电子设备100还可以设置三个，四个或更多个麦克风1043，实现采集声音信号，降噪，还可以识别声音来源，实现定向录音功能等。
扬声器1044，也称“喇叭”，用于将音频电信号转换为声音信号。车辆100可以通过扬声器1044收听音乐，或收听提示信息。
推进系统106可以包括动力部件1061、能源部件1062、传动部件1063、施动部件1064等。
动力部件1061可以为发动机，可以为汽油发动机、电动汽车的电动机、柴油发动机、混合动力发动机等发动机中的任一种或多种的组合，也可以为其他形式的发动机。
能源部件1062可以是能量的来源，全部或部分地为动力部件1061提供动力。也就是说，可将动力部件1061配置为将能源部件1062提供的能源转换为机械能。能源部件1062可提供的能源包括汽油、柴油、其他基于石油的燃料、丙烷、其他基于压缩气体的燃料、乙醇、太阳能电池板、电池、以及其他电功率来源。能源部件1062还可包括燃料箱、电池、电容器、和/或飞轮的任意组合。在一些实施例中，能源部件1062也可以为车辆100的其他系统提供能量。
传动部件1063可包括变速箱、离合器、差动器、传动轴以及其他组件。经配置，传动部件1063可将机械能从动力部件1061传输给施动部件1064。
施动部件1064可以包括车轮、轮胎等。车轮可配置为各种款式，包括单轮车、双轮车 /摩托车、三轮车、或者轿车/卡车四轮款式等。轮胎可以附接车轮，车轮可以附接到传动部件1063，可响应传动部件1063传动的机械功率而进行转动，以驱动车辆100行动。
传感器系统108可以包括定位部件1081、相机1082、惯性测量单元1083、雷达1084等。
定位部件1081可被配置为用于估计车辆100的位置。定位部件1081可包括被配置为基于卫星定位数据估计车辆100相对于地球的位置的收发器。在一些实施例中，可将计算系统102配置为结合地图数据使用定位部件1081来估计车辆100可能行驶的道路以及车辆100在道路上的位置。具体的，定位部件1081可以包括全球定位系统(global positioning system，GPS)模块，也可以包括北斗卫星导航系统(beidou navigation satellite system，BDS)，也可以包括伽利略卫星导航系统(galileo satellite navigation system)，等等。
相机1082可以包括被配置为捕获车辆100外部环境的车外相机，也可以包括被配置为捕获车辆100内部环境的车内相机。相机1082可以为检测可见光的相机，或者为检测来自光谱其他部分的光线(红外线或紫外线等)。相机1082用于捕获二维图像，也可以用于捕获深度图像。
惯性测量单元(inertial measurement unit，IMU)1083被配置为基于惯性加速度来感测车辆100的位置和方位变化的传感器的任意组合。在一些实施例，惯性测量单元1083可包括一个或多个加速计和陀螺仪。
雷达1084可包括被配置为使用光波或声波来感测或检测车辆100所在环境中的对象的传感器。具体的，雷达1084可以为包括激光雷达、毫米波雷达或超声波雷达等。
计算系统102可以根据传感器系统108中各部件所采集的数据，确定车辆100的感知信息。例如，计算系统102可以利用计算机视觉计算以及传感器融合技术，根据传感器系统108中一种或多种部件所采集的数据，确定车辆100的感知信息。
控制系统110可以包括转向部件1101、制动部件1102、油门1103等。
转向部件1101可以是被配置为响应驾驶员操作或计算机指令，调整车辆100的行驶方向的部件。示例性的，转向部件1101可以包括方向盘。
节流阀1102可以是被配置为控制动力部件1061的运行速度和加速度并进而控制车辆100的速度和加速度的部件。
制动部件1103可以是被配置为降低车辆100的行驶速度的部件。例如，制动部件1103可以使用摩擦力来减慢施动部件1064中车轮的转动速度。
油门1104可以是被配置为增加车辆100的行驶速度的部件。
控制系统110中的各部件可以响应控制模块1024下发的控制指令，执行相应操作。例如，转向部件1101中的方向盘可以响应控制模块1024下发的控制指令，旋转特定的角度。再例如，制动部件1103可以响应控制模块1024下发的控制指令，进行刹车。等等，此处不再一一列举。
电源112可配置为向车辆100的一部分或全部部件提供电力。在一些实施例中，电源112可包括可重复冲放电的锂离子电池或者铅蓄电池。在一些实施例中，电源112可包括一个或多个电池组。在一些实施例中，可以将电源112和能源部件1062共同实现，通过动力部件1061可将电源112提供的化学能转变为电机的机械能，并通过传动部件1063传递 给施动部件1064，实现车辆100的行驶。
基于车辆100的上述功能，本申请实施例提供了一种横向规划约束确定方法，车辆100可以根据其行驶过程中的车辆自身信息、道路的路况信息，动态的放开或者缩小横向规划约束。在本申请实施例中，车辆100也可以称为自车。横向规划约束也可以称为横向约束，其是指自车的横向避让范围。也就是说，自车需要在横向规划约束的范围或者说约束下进行横向位移。其中，此处的横向，是指与车辆行驶方向垂直的方向或近似垂直方向。
车辆100可以的处理器1021可以通过运行存储器1022中存储的一个或多个应用的指令，实现本申请实施例所提供的横向规划约束确定方法。为方便描述，在下文中，以该一个或多个应用中的应用为主体进行描述。可以理解，当提到某一应用执行某步骤或具有某功能时，其是指处理器1021通过运行该应用的指令而执行该步骤或实现该功能。
在驾驶车辆时，通常需要高层决策、路径规划(也称轨迹规划)和底层控制。以换道为例，人类驾驶员首先需要考虑安全因素、交规因素等，决定换道时机；其次，为车辆规划出一条行驶轨迹；然后，控制方向盘、油门或刹车等，使得车辆可以按照规划的行驶轨迹行驶。自动驾驶的决策过程和人类驾驶员的决策过程类似。参阅图2，行为规划模块10221负责高层决策，例如变换车道或保持车道。横向规划模块10222负责规划行驶轨迹，纵向规划模块10223负责规划行驶速度。控制模块10224根据规划的行驶轨迹和速度，操作转向部件1101、节流阀1102、制动部件1103或油门1104，以使自车按照规划的行驶轨迹和速度行驶。具体而言，横向规划模块10222在获取行为规划模块10221下放的决策指令(例如，保持车道)时，可以进行横向路径规划，并把规划好的横向路径传递给纵向规划模块10223。纵向规划模块10223可以根据该横向路径进行速度规划。控制模块10224可以根据横向规划模块10222的规划结果和纵向规划模块10223的规划结果，向转向部件1101、节流阀1102、制动部件1103或油门1104发出控制指令。
更具体地，如图2所示，横向规划模块10222还可以包括参考轨迹规划模块(reference path planner)、障碍避让模块(obstacle avoidance planner)、轨迹平滑模块(trajectory smoother)以及碰撞检测模块。示例性地，碰撞检测模块具体可以为基于模型的碰撞检测模块(model based collision checker)。其中，参考轨迹规划模块可以根据行为规划模块10221下发的决策指令、自车感知信息、地图信息等，确定自车在时间段T1内的行驶参考轨迹。时间段T1可以包括当前时刻。示例性的，时间段T1为当前时刻前的N秒时长和当前时刻后的M秒时长组成。时间段T1、N的值、M的值可以预先设定。其中，可以理解，当前时刻前的行驶参考轨迹为自车的实际已行驶的轨迹或者说历史行驶轨迹。当前时刻后的行驶参考轨迹为规划的或者说预测的轨迹。行驶参考轨迹可以包括多个路点，该多个路点位于行驶参考轨迹上。或者说，行驶参考轨迹由该多个路点中相邻路点之间的连线组成。示例性的，行驶参考轨迹可以和自车所在车道的中间线重合。
参考轨迹规划模块可以执行本申请实施例提供的横向规划约束确定方法，并确定的横向规划约束输出给障碍避让模块，以便障碍避让模块、轨迹平滑模块以及碰撞检测模块在横向规划约束内进行横向路径规划。
其中，本申请实施例所提供是确定横向规划约束的方案，因此，下文将对确定横向规划约束的过程进行具体介绍。横向规划约束内进行横向路径规划的过程可以参考现有技术的介绍。
接下来，结合附图，对本申请实施例提供的横向规划约束确定方法进行示例说明。其中，在本申请实施例中，“横向”是指垂直于车辆行驶方向的方向。“纵向”是指车辆行驶方向。
在一些实施例中，横向规划约束确定方法可以包括图3所示的步骤。具体如下。
参考轨迹规划模块可以执行步骤301，获取自车在道路A1上的行驶参考轨迹B1。示例性的，参考轨迹规划模块可以根据地图信息(例如，可以从高精地图10225获取地图信息)，确定行驶参考轨迹B1。示例性的，行驶参考轨迹B1可以由其他应用或部件确定，然后，参考轨迹规划模块可以从该其他应用或部件获取行驶参考轨迹B1。具体过程可以参考现有技术介绍，在此不再赘述。可以理解，行驶参考轨迹B1上具有多个路点，或者说，行驶参考轨迹是多个路点中相邻路点进行连线而得到的。
参考轨迹规划模块可以确定行驶轨迹B1上的多个路点中的路点的横向规划约束。以该多个路点中的路点B11为例，如图3所示，参考轨迹规划模块可以执行步骤302，根据行驶参考轨迹B1上的路点B11的C1侧的车道线，确定路点B11的C1侧的静态横向规划约束D1。C1侧可以为左侧，或者右侧。可以理解，路点B11位于车道内，或者位于虚拟车道内(例如，路点B11位于路口，参考轨迹规划模块可以根据地图信息确定虚拟车道)。由此，路点B11左右两侧具有车道线。其中，路点B11的C1侧车道线是指从路点B11向路点B11所在车道的C1侧的车道线做垂线，该垂线与车道的C1侧的车道线的交点处为路点B11的C1侧的车道线。其中，路点B11位于虚拟车道时，可以默认路点B11的两侧的车道线都为虚线。
虚线也可称为可换道线。根据交通法规，自车可以在虚线处换道。与虚线相对的是实线，实线也可称为不可换道线。根据交通法规，自车不可以在实线处换道。
在一个说明性示例中，静态横向规划约束D1可以包括软静态横向规划约束(soft margin)D11和硬静态横向规划约束(hard margin)D12。其中，硬静态横向规划约束的宽度大于或等于软静态横向规划约束。
示例性的，若路点B11的C1侧的车道线为实线，软静态横向规划约束D11小于路点B11所在车道的二分之一宽度。其中，车道的中心线可以将车道分为两个单侧车道，每个单侧车道的宽度为车道的二分之一宽度。因此，在本申请实施例中，车道的二分之一宽度等于单侧车道的宽度，车道的二分之一也可以称为单侧车道。在一个例子中，可以将路点B11所在车道的单侧车道宽度(或者说车道的二分之一宽度)减去宽度W1，得到软静态横向规划约束D11。宽度W1可以为预设，例如可以为0.3米，或者为0.4米，或者为0.5米，等等，此处不再一一列举。
若路点B11的C1侧的车道线为实线，硬静态横向规划约束D12的宽度等于软静态横向规划约束D12的宽度。
示例性的，若路点B11的C1侧的车道线为虚线，软静态横向规划约束D11小于路点B11所在车道的二分之一宽度。在一个例子中，可以将路点B11所在车道的单侧车道宽度(或者说车道的二分之一宽度)减去宽度W1，得到软静态横向规划约束D11。宽度W1可以参考上文介绍，在此不再赘述。
若路点B11的C1侧的车道线为虚线，硬静态横向规划约束D12的宽度大于软静态横向规划约束D12的宽度。示例性的，可以将软静态横向规划约束D11加上宽度W2，得到硬 静态横向规划约束D12。其中，宽度W1可以为预设，例如可以为0.4米，或者为0.8米，或者为1.2米，等等，此处不再一一列举。示例性的，宽度W2可以大于宽度W1。也就是说，在路点B11的C1侧的车道线为虚线的情况下，即自车可以越过或跨过路点B11的C1侧的车道线的情况下，可以设置路点B11的C1侧的硬静态横向规划约束的宽度较大，以便可以自车可以借用路点B11的C1侧的车道进行横向位移，增加了横向位移的可调节范围。
在一个具体实例中，可以设定自车在最右侧车道正常行驶，当前车道宽度为3.6米。车道左侧为实线，不可进行换道。自车当前所在位置车道右侧车道边线为虚线，可进行换道。自车前方80m–120m处，虚线变为实线，转为不可换道区域。自车前方120m以后为路口区域。行驶参考轨迹的长度截止到150米。则此时，行驶参考轨迹上距自车120米以内所有路点的左侧软静态横向规划约束和左侧硬静态横向规划约束都为3.6/2–0.4＝1.4米。行驶参考轨迹上的，且位于处于右可换道区域的路点(距自车80m内的路点)的右侧软静态横向规划约束都为3.6/2–0.4＝1.4米，右侧硬静态横向规划约束为1.4+1.2＝2.6米。距离自车80m-120m的路点的右侧软静态横向规划约束、右侧硬静态横向规划约束都为3.6/2–0.4＝1.4米。距自车120m后的路点的左软静态横向规划约束、右软静态横向规划约束都为3.6/2–0.4＝1.4米，左硬静态横向规划约束和右硬静态横向规划约束为1.4+1.2＝2.6米。
需要说明的是，“前”是指自车的车头指向的方向，“后”是指自车的车尾指向的方向。
继续参阅图3，参考轨迹规划模块可以执行步骤303，判断自车是否处于换道状态。示例性的，参考轨迹规划模块可以根据其最近从行为规划模块10221接收的决策指令，判断自车是否处于换道状态。若最近从行为规划模块10221接收的决策指令为变换车道，则可以确定自车处于换道状态。若最近从行为规划模块10221接收的决策指令为保持车道，则可以确定自车不处于换道状态。
可以设定步骤303的判断结果为否，即自车不处于换道状态。参考轨迹规划模块还可以执行步骤304，判断自车是否处于避让状态。
示例性的，可以理解，行驶参考轨迹B1并非自车的实际行驶轨迹。自车在行驶过程中，可能因为避让障碍物等原因，而偏离行驶参考轨迹B1。可以判断自车当前的实际位置和行驶参考轨迹B1之间的最近距离是否大于预设阈值Y1，由此可以判断自车的实际行驶路线是否偏离了行驶参考轨迹B1。自车当前的实际位置和行驶参考轨迹B1之间的最近距离大于预设阈值Y1，可以认为自车的实际行驶路线偏离了行驶参考轨迹B1，自车正在执行避让动作，即自车处于避让状态。若自车当前的实际位置和行驶参考轨迹B1之间的最近距离小于或等于预设阈值Y1，则可以认为自车的实际行驶路线没有偏离行驶参考轨迹B1，自车没有处于避让状态。在一个例子中，可以从自车的实际位置向行驶参考轨迹B1作垂线，得到垂线和行驶参考轨迹B1的交点，然后确定自车的实际位置和该交点之间的距离，可以得到自车的实际位置和行驶参考轨迹B1之间的最近距离。阈值Y1可以为预设值，例如可以为0.2米，或者0.3米，等等，此处不再一一列举。
示例性的，可以确定横向规划模块10222最近输出(例如向纵向规划模块10223输出)的最终行驶轨迹和行驶参考轨迹B1之间的横向偏差。可以理解，横向规划模块10222可以按照预设的周期输出最终行驶轨迹。对于一个周期而言，横向规划模块10222输出的最 终行驶轨迹为障碍避让模块、轨迹平滑模块以及碰撞检测模块在参考轨迹规划模块确定行驶参考轨迹以及横向规划约束的基础上进行进一步处理得到的最终行驶轨迹。参考轨迹规划模块可以获取横向规划模块10222最近输出的最终行驶轨迹，并比较该最终行驶轨迹和行驶参考轨迹B1之间的横向偏差。横向偏差大于预设阈值Y2，则可以认为自车的实际行驶路线偏离了行驶参考轨迹B1，自车正在执行避让动作，即自车处于避让状态。若横向偏差小于或等于预设阈值Y2，则可以认为自车的实际行驶路线没有偏离行驶参考轨迹B1，自车没有处于避让状态。在一个例子中，可以从最终行驶轨迹上的任一点b1向行驶参考轨迹B1作垂线，得到垂线和行驶参考轨迹B1的交点，然后计算该交点和点b1之间的距离。重复前述方案，可以得到最终行驶轨迹上的每个点到行驶参考轨迹B1的距离。从最终行驶轨迹上的各个点到行驶参考轨迹B1的距离中，确定出最大距离，并将该最大距离作为最终行驶轨迹和行驶参考轨迹B1之间的横向偏差。阈值Y2可以为预设值，例如可以为0.2米，或者0.3米，等等，此处不再一一列举。
由此，通过上述方案，在步骤304中，可以确定出自车是否处于避让状态。
可以设定步骤304的判断结果为否，即自车不处于避让状态。参考轨迹规划模块还可以执行步骤305，根据自车在路点B11时的行驶速度V1、静态横向规划约束D1、自车的车身宽度，确定自车在路点B11时的C1侧的初始横向规划约束E1。
示例性的，可以理解，在执行步骤305时，自车可能未实际行驶到路点B11。步骤305中的自车在路点B11时的行驶速度V1可以为规划或者说预测的速度。示例性的，纵向规划模块10223可以根据车辆的感知信息、横向规划模块10222最近输出的最终行驶轨迹、以及道路A1上的交通路况信息等，预测自车在行驶到路点B11时的速度，由此可以得到行驶速度V1。其中，预测自车在某一路点时的速度的具体方案可以参考现有技术的介绍，在此不再赘述。
在一个说明性示例中，可以对行驶速度V1进行惯性滤波，得到滤波后的行驶速度V1。该惯性滤波可以为一级惯性滤波。在一个例子中，可以获取行驶参考轨迹B1上的，与路点B11相邻，且位于路点B11之前的路点B12的行驶速度V2。行驶速度V2可以为纵向规划模块10223预测的自车在行驶到路点B12时的速度，也可以为自车在路点B12的实际行驶速度。然后，可以根据行驶速度V2、惯性滤波系数，对行驶速度V1进行惯性滤波。在一个例子中，可以设定自车加速时的惯性滤波系数为0.97，自车减速时的惯性滤波系数为0.90。
在一个说明性示例中，可以比较行驶速度V1(或者滤波后的行驶速度V1)和预设的速度阈值v1。若行驶速度V1(或者滤波后的行驶速度V1)小于或等于速度阈值v1，则可以确定初始横向规划约束E1等于静态横向规划约束D1。示例性的，如上文所述，静态横向规划约束D1包括软静态横向规划约束D11和硬静态横向规划约束D12。相应的，初始横向规划约束E1包括软初始横向规划约束E11和硬初始横向规划约束E12。初始横向规划约束E1等于静态横向规划约束D1具体为软初始横向规划约束E11等于软静态横向规划约束D11，硬初始横向规划约束E12等于硬静态横向规划约束D12。速度阈值v1可以为20km/h，或者为25km/h，或者为30km/h，等等。在具体实现时，可以根据经验或实验设置速度阈值v1的大小。
在一个说明性示例中，可以比较行驶速度V1(或者滤波后的行驶速度V1)和预设的速 度阈值v2。若行驶速度V1(或者滤波后的行驶速度V1)大于或等于速度阈值v2，则可以确定初始横向规划约束E1等于自车车身宽度的一半。示例性的，初始横向规划约束E1包括软初始横向规划约束E11和硬初始横向规划约束E12。初始横向规划约束E1等于自车车身宽度的一半具体为软初始横向规划约束E11和硬初始横向规划约束E12均等于自车车身宽度的一半。速度阈值v2大于速度阈值v1。在一个例子中，速度阈值v2可以为70km/h，或者为75km/h，80km/h，等等。在具体实现时，可以根据经验或实验设置速度阈值v2的大小。
在一个说明性示例中，若行驶速度V1(或者滤波后的行驶速度V1)大于速度阈值v1，且小于速度阈值v2，则初始横向规划约束E1介于静态横向规划约束D1和自车车身宽度一半之间，且与行驶速度V1(或者滤波后的行驶速度V1)负相关。示例性的，该负相关具体为一级线性变化的负相关。可以利用图4所示的横向规划约束-速度曲线确定初始横向规划约束E1。具体而言，可以确定横向规划约束-速度曲线上纵坐标为行驶速度V1(或者滤波后的行驶速度V1)的坐标点，然后，将该坐标点的横坐标用作初始横向规划约束E1。
在该示例的一个例子中，初始横向规划约束E1包括软初始横向规划约束E11和硬初始横向规划约束E12。若行驶速度V1(或者滤波后的行驶速度V1)大于速度阈值v1，且小于速度阈值v2，则软初始横向规划约束E11介于软静态横向规划约束D11和自车车身宽度一半之间，且与行驶速度V1(或者滤波后的行驶速度V1)负相关；以及，硬初始横向规划约束E12介于硬静态横向规划约束D12和自车车身宽度一半之间，且与行驶速度V1(或者滤波后的行驶速度V1)负相关。
在一个具体例子中，在图5所示的场景中，可以设定道路A1的宽度为3.6米，宽度W1为0.4米，宽度W2为1.2米，自车车身宽度为2米。路点B11左侧的软静态横向规划约束501为3.6/2-0.4＝1.4米。路点B11的左侧为实线，路点B11的左侧的硬静态横向规划约束502也为1.4米。路点B11右侧的软静态横向规划约束503为3.6/2-0.4＝1.4米。右侧的硬静态横向规划约束504为1.4+1.2＝2.6米。行驶速度V2为26.0km/h，经过滤波后的行驶速度V1为25.2km/h，可知，自车在路点B1时处于加速状态。可以设定自车加速时的惯性滤波系数为0.97。则对行驶速度V2进行惯性滤波后的速度为26.0km/h×(1-0.97)+25.2km/h×0.97＝25.224km/h。可以设定速度阈值v1为20km/h，速度阈值v2为70km/h。根据惯性滤波后的行驶速度V1(即25.224km/h)，进行一阶线性变化，可得到路点B11的左侧的软初始横向规划约束和硬初始横向规划约束均为1.36米，以及可以得到路点B11的右侧的软初始横向规划约束为1.36米，右侧的硬初始横向规划约束为2.36米。
当道路A1存在需自车关注的对象F1时，参考轨迹规划模块还可以执行步骤306，至少根据对象F1和路点之间的距离L1和初始横向规划约束E1，确定自车在路点B11时的C1侧的实际横向规划约束H1。示例性的，初始横向规划约束E1可以包括软初始横向规划约束E11和硬初始横向规划约束E12。相应的，实际横向规划约束H1可以包括软实际横向规划约束H11和硬实际横向规划约束H12。
对象F1可以为在自车行驶到路点B11时，对自车的行驶过程具有影响的物体，因此，自车需要关注对象F1，并根据对象F1调整行驶策略，例如行驶速度、行驶路线等。对象F1也可以理解为障碍物，或者说，对象F1与自车之间具有碰撞的风险。
示例性的，对象F1具体可以为当自车行驶到路点B11时，需要关注的对象。也就是说，对象F1可以为自车行驶到路点B11时，需要考虑的对象，以避免与对象F1发生碰撞。
接下来，结合不同的应用场景对步骤305进行举例介绍。
图6示出了一种应用场景，在该场景中，自车，即车辆100，行驶在道路A1的车道A11上。道路A1还包括位于车道A11右侧，且与车道A11相邻的车道A12。车道A12上行驶有车辆200。可以理解，若车辆100越过或跨过车道，则车辆100和车辆200有碰撞的风险，由此，需要收缩车辆100的右侧的横向规划约束。
在该场景中，可以确定路点B11和车辆200之间的距离L1、车辆200的运动速度V3、车辆200的运动方向M1、车辆100的运动速度V4、车辆100的运动方向M2。然后，根据距离L1、车辆200的运动速度V3、车辆200的运动方向M1、车辆100的运动速度V4、车辆100的运动方向M2，确定车辆100和车辆200之间的碰撞风险度。若确定出的碰撞风险度小于预设的安全阈值时，可以缩小车辆100的右侧的硬初始横向规划约束E12，使得缩小后的硬初始横向规划约束E12小于或等于路点B11所在车道(即车道A11)的二分之一宽度。例如，可以将硬初始横向规划约束E12的宽度缩小到软初始横向规划约束E11。之后，将缩小后的硬初始横向规划约束E12用作硬实际横向规划约束H12。
其中，车辆200的运动速度V3可以是，预测的在车辆100行驶到路点B11时车辆200的运动速度。车辆200的运动方向M1可以是，预测的在车辆100行驶到路点B11时车辆200的运动方向。车辆100的运动速度V4可以是，预测的在车辆100行驶到路点B11时车辆100的运动速度。运动方向M2可以是，预测的车辆100行驶到路点B11时车辆100的运动方向。
在一个示例中，确定出的碰撞风险度可以为碰撞时间(time to collision，TTC)。在一个例子中，可以根据运动速度V3和运动速度V4之间的差值，以及车辆200的运动方向、车辆100的运动方向，可以确定车辆100和车辆200在逐渐接近。可以根据距离L1、运动速度V3和运动速度V4之间的差值，确定在保持车辆100、车辆200的运动方向和运动速度不变的情况下，车辆100和车辆200多久会发生碰撞，由此，得到碰撞时间。在其他例子中，还可以根据现有技术的方案确定车辆100和车辆200之间的碰撞时间，在此不再赘述。
安全阈值可以包括时间阈值T1。时间阈值T1可以称为碰撞时间安全阈值。时间阈值T1可以为4.5秒，或者为5秒，或者为6秒，等等。开发人员可以设置时间阈值T1。当确定出的碰撞时间小于时间阈值T1时，可以缩小车辆100的右侧的硬初始横向规划约束E12。当碰撞时间大于或等于时间阈值T1时，可以不缩小车辆100的右侧的硬初始横向规划约束E12。
在另一个示例中，确定出的碰撞风险度可以为头车时距(head way time，HWT)。在一个例子中，可以根据距离L1、运动速度V3、运动速度V4确定车辆100和车辆200的头车时距。在其他例子中，可以根据现有技术的方案确定车辆100和车辆200的头车时距，在此不再赘述。
安全阈值可以包括时间阈值T2。时间阈值T2可以称为头车时距安全阈值。时间阈值T2可以为0.4秒，或者为0.5秒，或者为0.6秒，等等。开发人员可以设置时间阈值T2。当确定出的头车时距小于时间阈值T2时，可以缩小车辆100的右侧的硬初始横向规划约 束E12。当头车时距大于或等于时间阈值T2时，可以不缩小车辆100的右侧的硬初始横向规划约束E12。
在又一个示例中，确定出的碰撞风险度可以包括碰撞时间和头车时距。相应的，安全阈值可以包括时间阈值T1和时间阈值T1。当碰撞时间小于时间阈值T1和/或头车时距小于时间阈值T2时，可以缩小车辆100的右侧的硬初始横向规划约束E12。当碰撞时间大于或等于时间阈值T1，且头车时距大于或等于时间阈值T2时，可以不缩小车辆100的右侧的硬初始横向规划约束E12。
在一个例子中，可以设定，预测车辆100行驶到路点B11时，车辆200位于车辆100的后方10米，车辆200的运动速度为45km/h。确定出车辆200距离车辆100的碰撞时间为1.9秒、头车时距为0.8秒。其中，虽然车辆200距离车辆100的头车时距大于头车时间安全阈值0.5秒，但车辆200距离车辆100的碰撞时间小于碰撞时间安全阈值4.5秒。则确定路点B11的右侧不安全，具有较高的碰撞风险，将路点B11的右侧初始硬横向规划约束602回收至右侧的软初始横向规划约束601的宽度位置。例如，若右侧的软初始横向规划约束的宽度为1.36米，则将路点B11的右侧硬初始横向规划约束的宽度缩小为1.36米。
图7示出了另一种应用场景。对象F1与路点B11处于同一车道，或者，对象F1阻塞了路点B11所在车道。车辆100在行驶期间逐渐靠近对象F1。为此，车辆100需要规划如何避让对象F1。
示例性的，对象F1可以为单一障碍物。
示例性的，如图7所示，对象F1可以为多个障碍物组成的障碍物群。该多个障碍物无法满足车辆100的运动学约束，车辆100无法找到一条能够从该多个障碍物之间间隙穿过的轨迹。由此，可以将该多个障碍物看作单一障碍物进行处理。
可以设定路点B11的C1侧的车道线为虚线，即可换道线。可以确定对象F1和路点B11之间的横向偏移P1。横向偏移P1为对象F1的C1侧边界相对于路点B11在方向J1上的偏移或者说距离。方向J1垂直于车辆100位于路点B11时的行驶方向。示例性的，横向偏移P1可以为对象F1的C1边界相对于路点B11在方向J1上的偏移或者说距离。C1侧的边界可以为C1侧的最远点，或者说最C1侧(C1侧为右侧或左侧)。
实际横向规划约束H1包括软实际横向规划约束H11和硬实际横向规划约束H12。其中，为了避让对象F1，车辆100需要确定出硬实际横向规划约束H12。方案具体如下。
可以将横向偏移P1减去路点B11的C1侧的软初始横向规划约束，即将横向偏移P1减去软初始横向规划约束E11，得到差值Q1。将差值Q1加上车辆100的车身宽度，得到加和S1。
在一个示例中，可以判断加和S1是否小于或等于横向扩张宽度。横向扩张阔度可以是指在交通法规允许的范围内，车辆可以越过车道线，侵入到相邻车道的横向距离。横向扩张宽度可以预设，例如，可以为2.8米，或者为3米，或者为3.2米，等等，此处不再一一列举。若加和S1小于或等于横向扩张宽度，则可以将路点B11的C1侧的软初始横向规划约束(即软初始横向规划约束E11)加上横向扩张宽度，得到的相加和可以作为硬实际横向规划约束H12。
在一个示例中，可以预先设置横向安全距离。示例性的，该横向安全距离表示车辆经 过障碍物时，可允许的横向误差，从而使得车辆发生了横向安全距离范围内的横向误差时也不碰撞上障碍物。示例性的，该横向安全距离表示车辆安全通过障碍物时，与障碍物应保持的距离。在一个例子中，安全距离可以为0.5米，或者为0.6米，等等，此处不再一一列举。在该示例中，可以判断加和S1和安全距离的相加和是否小于或等于横向扩张宽度。若加和S1和横向安全距离的相加和小于或等于横向扩张宽度，则可以将路点B11的C1侧的软初始横向规划约束(即软初始横向规划约束E11)加上横向扩张宽度，得到的相加和可以作为硬实际横向规划约束H12。
在一个示例中，可以将B11的C1侧的软初始横向规划约束(即软初始横向规划约束E11)直接用作软实际横向规划约束。
接下来，在一个具体实例中，介绍图7所示的场景。
在图7所示的场景中，车辆100周围没有需要收缩硬横向规划约束的其他车辆时。即车辆100所在车道的相邻车道没有向车辆100驶来的其他车辆或没有超越车辆100的其他车辆。或者，车辆100所在的车道有向自车驶来或超越的其他车辆，该其他车辆和车辆100直接的碰撞时间、头车时距都满足安全要求。因此，不需要收缩硬横向规划约束。
筛选感兴趣的障碍物或者说需关注的对象。具体而言，可以将为于车辆100前方，车速比车辆100快的障碍物剔除；与位于路点B11时的车辆100直接的碰撞时间大于4.5秒的障碍物剔除；行驶轨迹与车辆100所在车道没有干涉的障碍物剔除；有横向速度的其他车辆剔除。将剩余的障碍物作为感兴趣障碍物或者说需关注对象。计算需关注对象在frenet坐标系上的坐标，并按照与路点B11的纵向距离由近到远，对需关注对象进行排序。然后，可根据车辆100的运动学约束对障碍物进行聚类。如图7所示，可以设定障碍物群中的所有障碍物与路点B11的纵向距离20–30米内，且按照车辆100的运动学约束，车辆100无法找到一条能够穿过从障碍物群中穿过的轨迹，则对图7所示的障碍物群的所有障碍物进行聚类。聚类后的障碍物作为单一障碍物处理。可以设定，聚类后的障碍物的相对于路点B11的纵向坐标的起点在纵向方向上距离路点B11为23米，终点在纵向方向上距离路点B11为28米。聚类后的障碍物的相对于路点B11的横向坐标起点在横向方向上距离路点B11为-1.5米，超过此时的右侧软初始横向规划约束为0.14米(右侧)；横向坐标的终点在横向方向上距离路点B11为1.6米(左侧)。路点B11所在车道左侧为实线，右侧为虚线。因为左侧为实线区域，车辆100无法从聚类后的障碍物左侧通过。可以设定车辆100可以安全通过障碍物时，与障碍物横向的安全距离需保持0.5米，车辆100的车身宽度为2米。则当车辆100规划通过前方障碍物时，需要超出右侧的软初始横向规划约束701的宽度为0.14+0.5+2.0＝2.64米。可以设定车辆100的扩张硬横向规划约束的最大值为2.8米。即横向扩张宽度为2.8米。则当右侧硬横向规划约束扩张后，车辆100可以规划通过前方聚类后的障碍物。此时会将右侧硬横向规划约束扩大为1.36+2.8＝4.16米，即确定右侧硬实际横向规划702约束为4.16米。
图8示出了又一种应用场景。在该场景中，对象F1与路点B11处于同一车道，或者，对象F1阻塞了路点B11所在车道。车辆100在行驶期间逐渐靠近对象F1。为此，车辆100需要规划如何避让对象F1。对象F1可以参考上文对图7所示实施例的介绍，在此不再赘述。
可以设定路点B11的C1侧的车道线为虚线，即可换道线。可以设定路点B11的另一 侧为实线，即不可换道线。该另一侧是指C1的对侧。具体而言，C1侧为右侧，则该另一侧为左侧。C1侧为左侧，则该另一侧为右侧。可以确定对象F1和路点B11之间的横向偏移P1。具体可以参考图7所示实施例的介绍，在此不再赘述。
实际横向规划约束H1包括软实际横向规划约束H11和硬实际横向规划约束H12。其中，为了避让对象F1，车辆100需要确定出硬实际横向规划约束H12。方案具体如下。
可以将横向偏移P1减去路点B11的C1侧的软初始横向规划约束，即将横向偏移P1减去软初始横向规划约束E11，得到差值Q1。将差值Q1加上车辆100的车身宽度，得到加和S1。
在一个示例中，可以判断加和S1是否小于或等于横向扩张宽度。若加和S1大于横向扩张宽度，则将路点B11的C1侧硬初始横向规划约束(即硬初始横向规划约束E12)作为硬实际横向规划约束H12。
在一个示例中，可以预先设置安全距离。安全距离可以参考上文对图7所示实施例的介绍，在此不再赘述。在该示例中，可以判断加和S1和安全距离的相加和是否小于或等于横向扩张宽度。若加和S1和安全距离的相加和大于横向扩张宽度，则将路点B11的C1侧硬初始横向规划约束(即硬初始横向规划约束E12)作为硬实际横向规划约束H12。
在一个示例中，在一个示例中，可以将B11的C1侧的软初始横向规划约束(即软初始横向规划约束E11)直接用作软实际横向规划约束。
也就是说，在图8所示的应用场景中，车辆100在位于路点B11时不能规划处可以避让对象F1的轨迹。换言之，对象F1直接封死了车辆100在位于路点B11时的行驶方向。为此，可以直接输出路点B11的初始横向规划约束，使得后续规划模块不再考虑当车辆100位于路点B11时应如何避让对象F1。
在一些实施例中，在对象F1直接封死车辆100在位于路点B11时的行驶方向的情况下，车辆100还可以规划或确定路点B11的纵向规划约束。纵向规划约束也可以称为纵向可避让长度约束。车辆100在行驶到路点B11时，可以仅考虑路点B11的纵向规划约束范围的路况信息，规划行驶策略。其中，路况信息可以包括交通信号灯信息、道路限速信息、变道空挡信息、障碍物信息中的一种或多种。行驶策略可以包括行驶轨迹和/或行驶速度。
确定路点B11的纵向规划约束的方案具体如下。
可以确定路点B11和对象F1之间的纵向距离L2。纵向距离L2具体为在车辆行驶到路点B11时的行驶方向上的距离。在一个例子中，纵向距离L2可以通过frenet坐标系得到。纵向距离L2可以为对象F1在frenet坐标系的纵坐标起点在纵向方向上距离路点B11的距离。纵向方向为车辆行驶到路点B11时的行驶方向。在一个例子中，纵向距离L2可以通过漫水填充法得到。漫水填充法将在下文进行具体介绍，在此不再赘述。
可以确定按照车辆100行驶到路点B11时的行驶速度，在时长T3内可以行驶的长度L3。
在一个说明性示例中，可以比较纵向距离L2和长度L3。若纵向距离L2小于长度L3，可以将纵向距离L2确定为路点B11的纵向规划约束。若纵向距离L2大于长度L3，可以将长度L3确定为路点B11的纵向规划约束。
在一个说明性示例中，可以设置纵向安全距离。纵向安全距离可以预先设置，例如可以为2米、2.5米等。可以将纵向距离L2减去纵向安全距离，得到纵向距离L4。可以比较 纵向距离L4和长度L3。若纵向距离L4小于长度L3，可以将纵向距离L4确定为路点B11的纵向规划约束。若纵向距离L4大于长度L3，可以将长度L3确定为路点B11的纵向规划约束。
接下来，在一个具体实例中，介绍纵向规划约束的确定方案。
在图8所示的场景中，车辆100周围没有需要收缩硬横向规划约束的其他车辆时。例如，车辆100所在车道的相邻车道没有向车辆100驶来的其他车辆或没有超越车辆100的其他车辆。或者，车辆100所在的车道有向自车驶来或超越的其他车辆，该其他车辆和车辆100直接的碰撞时间、头车时距都满足安全要求。因此，不需要收缩硬横向规划约束。
筛选感兴趣的障碍物或者说需关注的对象。具体而言，可以将为于车辆100前方，车速比车辆100快的障碍物剔除；与位于路点B11时的车辆100直接的碰撞时间大于4.5秒的障碍物剔除；行驶轨迹与车辆100所在车道没有干涉的障碍物剔除；有横向速度的其他车辆剔除。将剩余的障碍物作为感兴趣障碍物或者说需关注对象。计算需关注对象在frenet坐标系上的坐标，并按照与路点B11的纵向距离由近到远，对需关注对象进行排序。然后，可根据车辆100的运动学约束对障碍物进行聚类。如图7所示，可以设定障碍物群中的所有障碍物与路点B11的纵向距离20–30米内，且按照车辆100的运动学约束，车辆100无法找到一条能够穿过从障碍物群中穿过的轨迹，则对图7所示的障碍物群的所有障碍物进行聚类。聚类后的障碍物作为单一障碍物处理。可以设定，聚类后的障碍物的相对于路点B11的纵向坐标的起点在纵向方向上距离路点B11为23米，终点在纵向方向上距离路点B11为28米。聚类后的障碍物的相对于路点B11的横向坐标起点在横向方向上距离路点B11为-2.5米，超过此时的右侧软初始横向规划约束为1.14米(右侧)；聚类后的障碍物的相对于路点B11的横向坐标终点点在横向方向上距离路点B11为1.6米(左侧)。路点B11所在车道左侧为实线线型，右侧为虚线线型。因为左侧为实线区域，车辆100无法从聚类后的障碍物左侧通过。可以设定车辆100可以安全通过障碍物时，与障碍物横向的安全距离需保持0.5米，自车宽度为2米。则当车辆100规划通过前方障碍物时，需要超出右侧软初始横向规划约束的宽度1.14+0.5+2.0＝3.64米。可以设定扩张硬横向规划约束的最大值为2.8米，即横向扩张宽度为2.8米。则当右侧硬横向规划约束扩张后，车辆100也无法规划通过前方聚类后的障碍物。在这种情况下，将右侧的硬初始横向规划约束作为右侧的硬实际横向规划约束。即右侧的硬实际横向规划约束的宽度为2.36米。
可以设定对象F1为静止障碍物，或者对象F1所包括障碍物为静止障碍物。
可以将路点B11右侧的硬实际横向规划约束和静止障碍物投影到占用栅格地图(occupancy grid map，OGM)中。可以设定OGM的长为以路点B11为中心向前60米，向后10米，向左、向右各20米，使用漫水填充法以B11为起始点开始填充，找到漫水填充法可以向前填充的最远距离。
在图7所示的场景中，路点B11的右侧硬横向规划约束扩出后，使用漫水填充法，可以填充到前方60米的地图边缘，则可以向前填充的最远距离为OGM的极大值(未示出)。
参阅图9，在图8所示的场景中，路点B11的右侧硬横向规划约束没有扩出，使用漫水填充法，可以填充到路点B11前方聚类后障碍物起点的位置，即为23米。即确定出的纵向距离L2为23米。可以设定纵向安全阈值为2米，则可以向前填充的最远距离为21米。 即确定出的纵向距离L4为21米。
可以根据车辆100在路点B11时的车速计算出期望避让静止障碍物的最远距离。可以设定预测或规划得到的车辆100在路点B11时的车速为26.0km/h，3秒中时长可行驶的距离为期望避让静止障碍物的最远距离。则可以计算出车辆100在路点B11时期望避让静止障碍物的最远距离为26×3/3.6＝21.67米。即长度L3为21.67米。
路点B11的纵向规划约束在纵向距离L4和长度L3之间取小，为21米。
参阅图10，路点B11的纵向规划约束在静止障碍物之前截至。车辆100在进行横向规划时，可以忽略纵向规划约束之外的障碍物，只考虑纵向规划约束之内的障碍物。也就说，当车辆100位于路点11时，根据纵向规划约束范围内的路况信息，确定车辆100的行驶策略。
示例性的，上述纵向规划约束的确定方案具体可以由车辆100的参考轨迹规划模块执行。
上文介绍了，在道路A1存在需要车辆100(即自车)关注的对象F1时，实际横向规划约束的确定方案。在道路A1不存在需要自车关注的对象F1情况下，参考轨迹规划模块可以直接将确定初始横向规划约束作为实际横向规划约束，并输出障碍避让模块。
上文介绍了在步骤303和步骤304的判断结果均为否的情况下，确定横向规划约束的方案。接下来，介绍在步骤303的判断结果为是的情况下，确定横向规划约束的方案。
在一些实施例中，在步骤303的判断结果为是的情况下，参考轨迹规划模块可以执行步骤308，根据换道起始路点的C1侧的车道线，确定静态横向规划约束D2，以及根据换道结束路点的C1侧的车道线，确定静态横向规划约束D3；其中，在自车换道期间，自车的C1侧的静态横向规划约束从静态横向规划约束D2逐渐变化到静态横向规划约束D3。
可以参阅图11，可以设定自车(即车辆100)处于从车道A1到车道A3的换道状态，即自车可以从车道A1到车道A3的换道动作。可以根据车道A1的C1侧的车道线，确定车道A1的C1侧的静态横向规划约束。还可以根据车道A3的C1侧的车道线，确定车道A3的C1的静态横向规划约束。根据车道线，确定静态横向规划约束的方式可以参考上文介绍，在此不再赘述。在换道期间，将自车的C1侧的静态横向规划约束逐渐从车道A1的C1侧的静态横向规划约束变化到车道A3的C1侧的静态横向规划约束。
可以理解，参考轨迹规划模块可以响应行为规划模块10221下发的换道指令，根据自车感知信息、地图信息等，规划出从车道A1到车道A3的换道轨迹。换道轨迹上可有多个路点，或者说，换道轨迹可以由多个路点中相邻路点之间的连续组成。可以确定换道轨迹上偏离车道A1中心线的第一个路点为换道起始路点，以及确定换道轨迹上回到车道A3中心线的第一个路点为换道结束路点。其中，偏离车道A1中心线的路点可以是指与车道A1中心线之间的距离大于预设阈值的路点。该预设阈值可以为0.2米，也可以为0.25米，等等，在具体实现时，可以进行预先设置。回到车道A3中心线的路点可以是指与车道A3中心线之间的距离小于或等于该预设阈值。其中，路点到中心线的距离可以是指路点到中心线的最近距离。例如，可以做通过路点到中心线的垂线，然后将路点到垂足之间的距离作为路点到中心线的距离。
可以根据换道起始路点的C1侧的车道线，确定换道起始路点的C1侧的静态横向规划约束。可以根据换道结束路点的C1侧的车道线，确定换道结束路点的C1的静态横向规划 约束。可以确定出换道轨迹上位于换道起始路点和换道结束路点之间的至少路点，该至少一个路点中的各个路点，按照在换道轨迹上的排列位置，依次从起始路点的C1侧的静态横向规划约束变化到换道结束路点的C1的静态横向规划约束。示例性的，从起始路点的C1侧的静态横向规划约束到换道结束路点的C1的静态横向规划约束的变化，可以为一阶线性变化。
在一个具体实例中，参阅图11，可以设定车道A1的宽度为3.6米。从换道轨迹的所有路点中，确定换道起始路点和换道结束路点。从换道轨迹的起始到换道起始路点的所有路点右侧的软静态横向规划约束和硬静态横向规划约束都为3.6/2–0.4＝1.4米。换道轨迹的起始到换道起始路点的所有路点左侧的软静态横向规划约束为1.4米，左侧的硬静态横向规划约束为1.4+1.2＝2.6米。从换道结束路点到换道轨迹的最后一个路点的所有路点的左侧的软静态横向规划约束和硬静态横向规划约束都为3.6/2–0.4＝1.4米，右侧的软静态横向规划约束为1.4米，右侧的硬静态横向规划约束为1.4+1.2＝2.6米。在换道起始路点和换道结束路点中间的路点，左侧、右侧的软静态横向规划约束都为1.4米。左侧的硬静态横向规划约束从2.6米逐渐缩小到1.4米，右侧的硬静态横向规划约束从1.4米逐渐增加到2.6米。具体可如图11所示。
在确定了换道轨迹中路点的静态横向规划约束后，可以根据路点的静态横向规划约束，确定路点的初始横向规划约束，以及确定路点的实际横向规划约束。其中，初始横向规划约束的确定方案和实际横向规划约束的确定方案可以参考上文介绍，在此不再赘述。
接下来，介绍在步骤303的判断结果为否，步骤304的判断结果为是的情况下，确定横向规划约束的方案。
在一些实施例中，在步骤303的判断结果为否，步骤304的判断结果为是的情况下，参考轨迹规划模块还可以执行步骤309，根据避让起始路点的初始横向规划约束，确定自车在避让状态下的实际横向规划约束。
示例性的，可以通过判断自车当前的实际位置和行驶参考轨迹B1之间的最近距离是否大于预设阈值Y1，而判断自车的实际行驶路线是否偏离了行驶参考轨迹B1，进而可以确定自车是否处于避让状态。示例性的，可以通过确定横向规划模块10222最近输出(例如向纵向规划模块10223输出)的最终行驶轨迹和行驶参考轨迹B1之间的横向偏差，并判断横向偏差是否大于预设阈值Y2，而判断自车是否处于避让状态。具体可以上文介绍，在此不再赘述。
可以根据自车进入避让状态之前的最后一个路点B13的C1侧的初始横向规划约束，确定自车在避让状态下的C1侧的实际横向规划约束。其中，路点B13为行驶参考轨迹B1上的一个路点。路点B13可以称为避让起始路点。也就是说，自车从路点B13开始执行避让动作。路点B13的C1侧的初始横向规划约束的确定过程可以参考上文对路点B11的C1侧的初始横向规划约束的确定过程的介绍。
具体而言，在避让期间，当自车的当前速度大于或等于自车在第二路点时的速度时，可以确定路点B13的C1侧的初始横向规划约束为自车的当前位置的C1侧的实际横向规划约束。也就是说，在避让期间，若自车加速，则保持横向规划约束不变，使得横向规划约束不随着车速加快而减少。
在避让期间，当自车的当前速度大于或等于自车在路点B13时的速度时，则根据自车 的当前速度、路点B13的C1的静态横向规划约束、自身的车身宽度，确定自车当前位置的C1侧的实际横向规划约束。其中，当前速度与实际横向规划约束负相关。具体可以参考上文对步骤305所示实施例的介绍。也就是说，在避让期间，若自车减速，可以扩大横向规划约束，从而可扩大自车的避让空间。
另外，需要说明的是，上文以步骤301-步骤306和步骤308、步骤309由参考轨迹规划模块执行，以及步骤307由障碍避让模块执行为例，对本申请实施例提供的横向规划约束确定方法进行了示例说明，但并不限定横向规划约束确定方法中各步骤的执行主体。在其他实施例中，横向规划约束确定方法中的各步骤可以其他一个或多个应用或部件执行。例如，可以单独设置横向规划约束确定模块，并由横向规划约束确定模块执行横向规划约束确定方法中的各步骤。在具体实现时，开发人员可以设置本申请实施例提供的横向规划约束确定方法中各步骤的执行主体。
由此，通过上述方案可以确定出自车在车道正常行驶状态下的横向规划约束，也可以确定出保持车道但处于避让状态下的横向规划约束，也可以确定出换道状态下的横向规划约束。
在一些实施例中，横向规划约束确定方法可以包括图12所示的步骤。具体如下。
步骤S1，作为输入获得周围的障碍物信息，车道线信息，当前自车车速以及当前自车位置信息，同时获得自车行驶参考轨迹的所有路点的位置信息。
步骤S2，根据当前不同的行驶状态(车道内行驶、换道行驶、路口中行驶)，生成参考轨迹上每个路点的车道内的软静态横向规划约束及可以跨车道的硬静态横向约束。当此时参考轨迹的路点位于可换道区域时(该测车道线为虚线)，软静态横向规划约束宽度小于当前车道宽度，可换道侧的硬静态横向约束宽度大于该测软静态横向规划约束。当此时参考轨迹的路点位于不可换道区域时(两侧车道线皆为实线)，软静态横向规划约束宽度小于当前车道宽度，硬静态横向约束宽度等于软静态横向规划约束。当此时参考轨迹的路点位于路口区域中时，软静态横向规划约束宽度小于当前车道宽度，硬静态横向约束宽度大于软静态横向规划约束。当此时参考轨迹的路点位于参考轨迹的换道阶段时，软静态横向规划约束宽度由当前车道的宽度向目标车道的宽度逐渐过渡，硬静态横向规划约束则根据当前车道的左右侧线型向目标车道的左右侧线型参考当参考轨迹路点位于可换道区域或不可换道区域的规则逐渐变化。
步骤S3，根据自车当前车速，对硬静态横向规划约束和软静态横向约束进行随速变化。通过不同的惯性滤波使车速升高时硬静态横向规划约束与软静态横向约束缓慢缩小，自车车速降低时快速增大，产生滞回的效果。
步骤S4，判断此时自车的行驶状态，如果自车正在避让，则进行S5步骤，如果自车当前按照参考轨迹行驶没有横向位移，则跳过S5步，直接进行S6步。
步骤S5，自车在离开参考轨迹开始避让或避让过程中，在自车加速的过程中保持锁定硬静态横向规划约束和软静态横向约束进的宽度不变，自车减速的过程中放大硬静态横向规划约束和软静态横向约束进的宽度，使横向约束只放宽不收窄，保证自车避让空间。
步骤S6，筛选出位于自车相邻车道正常行驶的社会车，如社会车到自车当前位置的TTC，HWT小于一定阈值，则收缩在参考轨迹每个路点上的该社会车侧硬横向规划约束至软横向规划约束的宽度，减小自车可避让范围。当自车周围没有需要收缩约束的社会车时，筛选 大幅度入侵自车当前车道的感兴趣障碍物，针对感兴趣障碍物扩大硬横向规划约束，使自车能够有足够的空间规划通过。感兴趣障碍物的筛选是通过剔除自车周围障碍物中，无关障碍物得到的。无关障碍物的类型有，位于自车后方或前方太远或正常行驶不需要避让的障碍物，远离自车车道，不相交的障碍物，大角度横穿或加塞的压线障碍物，压线行驶，但速度比自车快的或离自车较远的障碍物。将剩下的感兴趣障碍物沿SL坐标系离自车由近及远进行排序。按照自车运动学约束对障碍物进行聚类。对聚类后距离自车最近的障碍物群计算两侧侵占自车车道的宽度，当扩大硬横向规划约束能够使自车规划通过且该侧无碰撞风险时，对参考轨迹上的每个路点扩大该侧的硬横向规划约束。
步骤S7，将硬横向规划约束与完全静止的障碍物投影到OGM中，使用漫水填充法得到自车可稳定避让的最远距离。同时，根据自车车速算出期望避让完全静止障碍物的最远距离，将此距离与漫水填充法得到的距离取小，作为最终的可避让长度输出。自车在横向规划时，会忽略可避让长度之后的障碍物，只避让可避让长度之前的障碍物。
步骤S8，输出带有硬横向规划约束和软横向规划约束的参考轨迹及可避让长度。给后续避让规划模块。
本申请实施例提供的横向规划约束确定方法，可以根据环境动态的放开或收缩横向规划约束，当自车高速行驶或周围环境存在风险时，收缩横向规划约束，使自车在车道内部行驶。当需要大幅避让时，放开横向规划约束，支持自车大幅跨车道避让。从而可以在保证自车行驶的安全性的同时，增强系统的通过性。
本申请实施例提供的横向规划约束确定方法，还可以生成纵向可避让长度，来处理静态障碍物，当遇到前方道路被单个或多个静止障碍物封死，自车无法通过时，剩余宽度不够自车安全同过时，使规划模块不避让前方静止障碍物。即可使横向规划模块忽略可避让长度之后的障碍物，避免引起误避让或避让后自车仍然无法通过前方障碍物的现象，同时还可以控制自车避让静止障碍物的预瞄时间。
本申请实施例提供的横向规划约束确定方法，可将避让约束分为横向规划约束及纵向规划约束。横向约束控制避让范围，分为跨车道硬横向规划约束和车道内软横向规划约束两部分。纵向约束控制避让完全静态障碍物的预瞄时间。
本申请实施例提供的横向规划约束确定方法，可以根据道路线信息，路口信息等生成静态横向规划约束。静态横向规划约束可随自车速度调整，避让中横向规划约束。横向规划约束随自车车速升高缓慢减小，随自车车速降低迅速扩大。当自车正在避让时，锁定硬横向规划约束与软横向规划约束，使其只扩大不减小。
本申请实施例提供的横向规划约束确定方法，可以针对侧方障碍物动态缩小横向约束，当相邻车道车辆向自车行驶且满足条件时，缩小该侧的硬横向规划约束。针对侵占本车道的障碍物动态扩大横向约束，筛选出感兴趣的障碍物，并进行排序及聚类。选择聚类后离自车最近的障碍物群，在安全的情况下，扩大相应侧硬横向规划约束。
综合以上，本申请实施例提供了一种横向规划约束确定方法。该方法的执行主体可以为自车，即车辆100。更具体的，该方法可以由车辆100中任一具有数据处理功能的部件，例如上文所述的参考轨迹规划模块和/或障碍避让模块。
如图13所示，该方法包括如下步骤。
步骤1301，获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹 包括多个路点。
步骤1302，根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧。
步骤1303，当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束。
步骤1304，至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
在一些实施例中，所述第一侧的车道线为可换道线；所述静态横向规划约束包括第一软静态横向规划约束和第一硬静态横向规划约束；其中，所述第一软静态横向规划约束小于所述第一路点所在车道的二分之一宽度，所述第一硬静态横向规划约束大于所述第一路点所在车道的二分之一宽度；所述初始横向规划约束包括第一软初始横向规划约束和第一硬初始横向规划约束；其中，所述第一软初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一软静态横向规划约束、所述自动驾驶车辆的车身宽度确定；所述第一硬初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一硬静态横向规划约束、所述自动驾驶车辆的车身宽度确定。
在一些实施例中，所述第一软静态横向规划约束由所述第一路点所在车道的二分之一宽度减去第一预设值得到，所述第一硬静态横向规划约束由所述第一软静态横向规划约束加上第二预设值得到。
在一些实施例中，所述第一路点位于所述第一道路上的第一车道，所述第一对象位于与所述第一车道相邻的第二车道，且所述第二车道位于所述第一车道的所述第一侧；所述实际横向规划约束包括硬实际横向规划约束；所述至少根据第一对象和所述自动驾驶车辆之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：根据所述第一距离、所述第一对象的运动速度、所述第一对象的运动方向、所述自动驾驶车辆的运动速度、所述自动驾驶车辆的运动方向，确定所述第一对象和所述自动驾驶车辆之间的碰撞风险度；当所述碰撞风险度小于预设的安全阈值时，缩小所述第一硬初始横向规划约束，以使所述第一硬初始横向规划约束小于或等于所述第一路点所在车道的二分之一宽度；确定缩小后的所述第一硬初始横向规划约束，为所述硬实际横向规划约束。
在一些实施例中，所述碰撞风险度包括碰撞时间TTC和/或头车时距HWT。
在一些实施例中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：将第一差值和 所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；当所述第一加和≤所述第一侧的横向扩张宽度，或者，所述第一加与和预设的第一安全距离的相加和≤所述横向扩张宽度时，将所述第一软初始横向规划约束和所述横向扩张宽度进行相加，得到第二加和；确定所述第二加和为所述硬实际横向规划约束。
在一些实施例中，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线，所述第一侧的对侧的车道线为不可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；当所述第一加和＞所述第一侧的横向扩张宽度，或者，所述第一加和和预设的第一安全距离的相加和＞所述横向扩张宽度时，确定所述第一硬初始横向规划约束为所述硬实际横向规划约束。
在一些实施例中，所述方法还包括：确定所述第一路点和所述第一对象在所述自动驾驶车辆位于所述第一路点时的行驶方向上的第二距离，以及确定所述自动驾驶车辆在第一时长内行驶的第一长度；当第三距离小于所述第一长度时，确定所述第三距离为所述自动驾驶车辆在所述第一路点时的纵向规划约束；或者，当所述第三距离大于所述第一长度时，确定所述第一长度为所述自动驾驶车辆在所述第一路点时的纵向规划约束；其中，所述第三距离等于所述第二距离，或者所述第三距离由所述第二距离减去预设的第二安全距离得到；当所述自动驾驶车辆位于所述第一路点时，根据所述纵向规划约束范围内的路况信息，确定所述自动驾驶车辆的行驶策略。
在一些实施例中，所述根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束包括：当所述行驶速度≤预设的第一速度阈值时，确定所述初始横向规划约束等于所述静态横向规划约束；当所述行驶速度≥预设的第二速度阈值时，确定所述初始横向规划约束等于二分之一的所述车身宽度；当所述第一速度阈值＜所述行驶速度＜所述第二速度阈值时，根据所述行驶速度，按照速度-横向规划约束曲线，确定所述初始横向规划约束；其中，在所述横向规划约束-速度曲线上，速度的大小和横向规划约束的大小呈负相关。
在一些实施例中，所述方法还包括：当所述自动驾驶车辆执行避让动作时，根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束；其中，所述自动驾驶车辆从所述第二路点开始执行所述避让动作。
在一些实施例中，所述根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束包括：当所述自动驾驶车辆的当前速度≥所述自动驾驶车辆在所述第二路点时的速度时，确定所述第二路 点的所述第一侧的实际横向规划约束为所述当前位置的所述第一侧的实际横向规划约束；当所述自动驾驶车辆的当前速度＜所述自动驾驶车辆在所述第二路点时的速度时，根据所述当前速度、所述第二路点的所述第一侧的静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述当前位置的所述第一侧的实际横向规划约束。
在一些实施例中，所述方法还包括：当所述自动驾驶车辆执行从第三车道到第四车道的换道动作时，将所述自动驾驶车辆的所述第一侧的静态横向规划约束逐渐从第一静态横向规划约束变化到第二静态横向规划约束；其中，所述第一静态横向规划约束由所述第三车道的所述第一侧的车道线确定，所述第二静态横向规划约束由所述第四车道的所述第一侧的车道线确定。
在一些实施例中，所述方法还包括：在所述实际横向规划约束的范围内，控制所述自动驾驶车辆在所述第一路点时的所述第一侧的横向位移，所述横向位移为第一方向上的位移，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向。
本申请实施例提供的横向规划约束确定方法，可以根据交通环境动态的放开或者缩小横向规划约束，从而可以在保证自车行驶的安全性的同时，增强自车的通过性。
参阅图14，本申请实施例提供了一种横向规划约束确定装置1400，可以配置于自动驾驶车辆，例如车辆100。参阅图14，装置1400包括获取单元1410、第一确定单元1420、第二确定单元1430、第三确定单元1440。
获取单元1410，用于获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹包括多个路点；
第一确定单元1420，用于根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧；
第二确定单元140，用于当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束；
第三确定单元1440，用于至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
装置1400的各功能单元可以上文所述的方法实施例实现，例如图13所示的方法实施例，在此不再赘述。
上文主要从方法流程的角度对本申请实施例提供的装置进行了介绍。可以理解的是，各个终端为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据图13或图3或图12所示的方法实施例对电子设备等进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功 能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
本申请实施例提供的横向规划约束确定装置，可以根据交通环境动态的放开或者缩小横向规划约束，从而可以在保证自车行驶的安全性的同时，增强自车的通过性。
参阅图15，本申请实施例提供了一种横向规划约束确定装置1500。装置1500可以执行上述各方法实施例中自动驾驶车辆执行的操作，例如图13所示。其中，装置1500可以包括处理器1510、存储器1520。存储器1520中存储有指令，该指令可被处理器1510执行。当该指令在被处理器1510执行时装置1500可以执行上述各方法实施例中自动驾驶车辆执行的操作，例如图13所示。具体而言，处理器1510可以进行数据处理操作，收发器1530可以进行数据发送和/或接收的操作。
可以理解的是，本申请的实施例中的处理器可以是中央处理单元(central processing unit，CPU)，还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现场可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、晶体管逻辑器件，硬件部件或者其任意组合。通用处理器可以是微处理器，也可以是任何常规的处理器。
本申请的实施例中的方法步骤可以通过硬件的方式来实现，也可以由处理器执行软件指令的方式来实现。软件指令可以由相应的软件模块组成，软件模块可以被存放于随机存取存储器(random access memory，RAM)、闪存、只读存储器(read-only memory，ROM)、可编程只读存储器(programmable rom，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)、寄存器、硬盘、移动硬盘、CD-ROM或者本领域熟知的任何其它形式的存储介质中。一种示例性的存储介质耦合至处理器，从而使处理器能够从该存储介质读取信息，且可向该存储介质写入信息。当然，存储介质也可以是处理器的组成部分。处理器和存储介质可以位于ASIC中。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者通过所述计算机可读存储介质进行传输。所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘(solid state disk，SSD))等。
可以理解的是，在本申请的实施例中涉及的各种数字编号仅为描述方便进行的区分，并不用来限制本申请的实施例的范围。

Claims (29)

  1. 一种横向规划约束确定方法，其特征在于，应用于自动驾驶车辆；所述方法包括：
    获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹包括多个路点；
    根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧；
    当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束；
    至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
  2. 根据权利要求1所述的方法，其特征在于，所述第一侧的车道线为可换道线；所述静态横向规划约束包括第一软静态横向规划约束和第一硬静态横向规划约束；其中，所述第一软静态横向规划约束小于所述第一路点所在车道的二分之一宽度，所述第一硬静态横向规划约束大于所述第一路点所在车道的二分之一宽度；
    所述初始横向规划约束包括第一软初始横向规划约束和第一硬初始横向规划约束；其中，所述第一软初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一软静态横向规划约束、所述自动驾驶车辆的车身宽度确定；所述第一硬初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一硬静态横向规划约束、所述自动驾驶车辆的车身宽度确定。
  3. 根据权利要求2所述的方法，其特征在于，所述第一软静态横向规划约束由所述第一路点所在车道的二分之一宽度减去第一预设值得到，所述第一硬静态横向规划约束由所述第一软静态横向规划约束加上第二预设值得到。
  4. 根据权利要求2或3所述的方法，其特征在于，所述第一路点位于所述第一道路上的第一车道，所述第一对象位于与所述第一车道相邻的第二车道，且所述第二车道位于所述第一车道的所述第一侧；所述实际横向规划约束包括硬实际横向规划约束；
    所述至少根据第一对象和所述自动驾驶车辆之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：
    根据所述第一距离、所述第一对象的运动速度、所述第一对象的运动方向、所述自动驾驶车辆的运动速度、所述自动驾驶车辆的运动方向，确定所述第一对象和所述自动驾驶车辆之间的碰撞风险度；
    当所述碰撞风险度小于预设的安全阈值时，缩小所述第一硬初始横向规划约束，以使所述第一硬初始横向规划约束小于或等于所述第一路点所在车道的二分之一宽度；
    确定缩小后的所述第一硬初始横向规划约束，为所述硬实际横向规划约束。
  5. 根据权利要求4所述的方法，其特征在于，所述碰撞风险度包括碰撞时间TTC和/或头车时距HWT。
  6. 根据权利要求2或3所述的方法，其特征在于，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；
    所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：
    将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；
    当所述第一加和≤所述第一侧的横向扩张宽度，或者，所述第一加与和预设的第一安全距离的相加和≤所述横向扩张宽度时，将所述第一软初始横向规划约束和所述横向扩张宽度进行相加，得到第二加和；
    确定所述第二加和为所述硬实际横向规划约束。
  7. 根据权利要求2或3所述的方法，其特征在于，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线，所述第一侧的对侧的车道线为不可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；
    所述至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束包括：
    将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；
    当所述第一加和＞所述第一侧的横向扩张宽度，或者，所述第一加和和预设的第一安全距离的相加和＞所述横向扩张宽度时，确定所述第一硬初始横向规划约束为所述硬实际横向规划约束。
  8. 根据权利要求7所述的方法，其特征在于，所述方法还包括：
    确定所述第一路点和所述第一对象在所述自动驾驶车辆位于所述第一路点时的行驶方向上的第二距离，以及确定所述自动驾驶车辆在第一时长内行驶的第一长度；
    当第三距离小于所述第一长度时，确定所述第三距离为所述自动驾驶车辆在所述第一路点时的纵向规划约束；或者，当所述第三距离大于所述第一长度时，确定所述第一长度 为所述自动驾驶车辆在所述第一路点时的纵向规划约束；其中，所述第三距离等于所述第二距离，或者所述第三距离由所述第二距离减去预设的第二安全距离得到；
    当所述自动驾驶车辆位于所述第一路点时，根据所述纵向规划约束范围内的路况信息，确定所述自动驾驶车辆的行驶策略。
  9. 根据权利要求1所述的方法，其特征在于，所述根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束包括：
    当所述行驶速度≤预设的第一速度阈值时，确定所述初始横向规划约束等于所述静态横向规划约束；
    当所述行驶速度≥预设的第二速度阈值时，确定所述初始横向规划约束等于二分之一的所述车身宽度；
    当所述第一速度阈值＜所述行驶速度＜所述第二速度阈值时，根据所述行驶速度，按照速度-横向规划约束曲线，确定所述初始横向规划约束；其中，在所述横向规划约束-速度曲线上，速度的大小和横向规划约束的大小呈负相关。
  10. 根据权利要求1所述的方法，其特征在于；所述方法还包括：当所述自动驾驶车辆执行避让动作时，根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束；其中，所述自动驾驶车辆从所述第二路点开始执行所述避让动作。
  11. 根据权利要求10所述的方法，其特征在于，所述根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束包括：
    当所述自动驾驶车辆的当前速度≥所述自动驾驶车辆在所述第二路点时的速度时，确定所述第二路点的所述第一侧的实际横向规划约束为所述当前位置的所述第一侧的实际横向规划约束；
    当所述自动驾驶车辆的当前速度＜所述自动驾驶车辆在所述第二路点时的速度时，根据所述当前速度、所述第二路点的所述第一侧的静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述当前位置的所述第一侧的实际横向规划约束。
  12. 根据权利要求1所述的方法，其特征在于，所述方法还包括：当所述自动驾驶车辆执行从第三车道到第四车道的换道动作时，将所述自动驾驶车辆的所述第一侧的静态横向规划约束逐渐从第一静态横向规划约束变化到第二静态横向规划约束；其中，所述第一静态横向规划约束由所述第三车道的所述第一侧的车道线确定，所述第二静态横向规划约束由所述第四车道的所述第一侧的车道线确定。
  13. 根据权利要求1所述的方法，其特征在于，所述方法还包括：在所述实际横向规划约束的范围内，控制所述自动驾驶车辆在所述第一路点时的所述第一侧的横向位移，所述横向位移为第一方向上的位移，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向。
  14. 一种横向规划约束确定装置，其特征在于，配置于自动驾驶车辆；所述装置包括：
    获取单元，用于获取所述自动驾驶车辆在第一道路上的行驶参考轨迹，所述行驶参考轨迹包括多个路点；
    第一确定单元，用于根据所述多个路点中的第一路点的第一侧的车道线，确定所述第一路点的所述第一侧的静态横向规划约束；所述第一侧为左侧或右侧；
    第二确定单元，用于当所述自动驾驶车辆的实际行驶轨迹没有偏离所述行驶参考轨迹时，根据所述自动驾驶车辆在所述第一路点时的行驶速度、所述静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的初始横向规划约束；
    第三确定单元，用于至少根据第一对象和所述第一路点之间的第一距离和所述初始横向规划约束，确定所述自动驾驶车辆在所述第一路点时的所述第一侧的实际横向规划约束；所述第一对象处于所述第一道路上，且为所述自动驾驶车辆关注的对象。
  15. 根据权利要求14所述的装置，其特征在于，所述第一侧的车道线为可换道线；所述静态横向规划约束包括第一软静态横向规划约束和第一硬静态横向规划约束；其中，所述第一软静态横向规划约束小于所述第一路点所在车道的二分之一宽度，所述第一硬静态横向规划约束大于所述第一路点所在车道的二分之一宽度；
    所述初始横向规划约束包括第一软初始横向规划约束和第一硬初始横向规划约束；其中，所述第一软初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一软静态横向规划约束、所述自动驾驶车辆的车身宽度确定；所述第一硬初始横向规划约束由所述自动驾驶车辆在所述第一路点时的行驶速度、所述第一硬静态横向规划约束、所述自动驾驶车辆的车身宽度确定。
  16. 根据权利要求15所述的装置，其特征在于，所述第一软静态横向规划约束由所述第一路点所在车道的二分之一宽度减去第一预设值得到，所述第一硬静态横向规划约束由所述第一软静态横向规划约束加上第二预设值得到。
  17. 根据权利要求15或16所述的装置，其特征在于，所述第一路点位于所述第一道路上的第一车道，所述第一对象位于与所述第一车道相邻的第二车道，且所述第二车道位于所述第一车道的所述第一侧；所述实际横向规划约束包括硬实际横向规划约束；
    所述第三确定单元还用于：
    根据所述第一距离、所述第一对象的运动速度、所述第一对象的运动方向、所述自动驾驶车辆的运动速度、所述自动驾驶车辆的运动方向，确定所述第一对象和所述自动驾驶车辆之间的碰撞风险度；
    当所述碰撞风险度小于预设的安全阈值时，缩小所述第一硬初始横向规划约束，以使所述第一硬初始横向规划约束小于或等于所述第一路点所在车道的二分之一宽度；
    确定缩小后的所述第一硬初始横向规划约束，为所述硬实际横向规划约束。
  18. 根据权利要求17所述的装置，其特征在于，所述碰撞风险度包括碰撞时间TTC和 /或头车时距HWT。
  19. 根据权利要求15或16所述的装置，其特征在于，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；
    所述第三确定单元还用于：
    将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；
    当所述第一加和≤所述第一侧的横向扩张宽度，或者，所述第一加与和预设的第一安全距离的相加和≤所述横向扩张宽度时，将所述第一软初始横向规划约束和所述横向扩张宽度进行相加，得到第二加和；
    确定所述第二加和为所述硬实际横向规划约束。
  20. 根据权利要求15或16所述的装置，其特征在于，所述第一对象与所述第一路点处于同一车道，且在所述自动驾驶车辆行驶期间所述自动驾驶车辆逐渐靠近所述第一对象；所述第一侧的车道线为可换道线，所述第一侧的对侧的车道线为不可换道线；所述第一距离包括所述第一对象和所述第一路点之间的第一横向偏移，所述第一横向偏移为所述第一对象和所述第一路点之间在第一方向上的距离；所述实际横向规划约束包括硬实际横向规划约束，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向；
    所述第三确定单元还用于：
    将第一差值和所述自动驾驶车辆的宽度相加，得到第一加和；所述第一差值由所述第一横向偏移减去所述第一软初始横向规划约束得到；
    当所述第一加和＞所述第一侧的横向扩张宽度，或者，所述第一加和和预设的第一安全距离的相加和＞所述横向扩张宽度时，确定所述第一硬初始横向规划约束为所述硬实际横向规划约束。
  21. 根据权利要求20所述的装置，其特征在于，所述装置还包括：
    第四确定单元，用于确定所述第一路点和所述第一对象在所述自动驾驶车辆位于所述第一路点时的行驶方向上的第二距离，以及确定所述自动驾驶车辆在第一时长内行驶的第一长度；
    第五确定单元，用于当第三距离小于所述第一长度时，确定所述第三距离为所述自动驾驶车辆在所述第一路点时的纵向规划约束；或者，当所述第三距离大于所述第一长度时，确定所述第一长度为所述自动驾驶车辆在所述第一路点时的纵向规划约束；其中，所述第三距离等于所述第二距离，或者所述第三距离由所述第二距离减去预设的第二安全距离得到；
    第六确定单元，用于当所述自动驾驶车辆位于所述第一路点时，根据所述纵向规划约束范围内的路况信息，确定所述自动驾驶车辆的行驶策略。
  22. 根据权利要求14所述的装置，其特征在于，所述第二确定单元还用于：
    当所述行驶速度≤预设的第一速度阈值时，确定所述初始横向规划约束等于所述静态横向规划约束；
    当所述行驶速度≥预设的第二速度阈值时，确定所述初始横向规划约束等于二分之一的所述车身宽度；
    当所述第一速度阈值＜所述行驶速度＜所述第二速度阈值时，根据所述行驶速度，按照速度-横向规划约束曲线，确定所述初始横向规划约束；其中，在所述横向规划约束-速度曲线上，速度的大小和横向规划约束的大小呈负相关。
  23. 根据权利要求14所述的装置，其特征在于；所述装置还包括：第七确定单元，用于当所述自动驾驶车辆执行避让动作时，根据所述多个路点中第二路点的所述第一侧的初始横向规划约束，确定所述自动驾驶车辆的当前位置的所述第一侧的实际横向规划约束；其中，所述自动驾驶车辆从所述第二路点开始执行所述避让动作。
  24. 根据权利要求23所述的装置，其特征在于，所述第七确定还用于：
    当所述自动驾驶车辆的当前速度≥所述自动驾驶车辆在所述第二路点时的速度时，确定所述第二路点的所述第一侧的实际横向规划约束为所述当前位置的所述第一侧的实际横向规划约束；
    当所述自动驾驶车辆的当前速度＜所述自动驾驶车辆在所述第二路点时的速度时，根据所述当前速度、所述第二路点的所述第一侧的静态横向规划约束、所述自动驾驶车辆的车身宽度，确定所述当前位置的所述第一侧的实际横向规划约束。
  25. 根据权利要求14所述的装置，其特征在于，所述装置还包括第八确定单元，用于当所述自动驾驶车辆执行从第三车道到第四车道的换道动作时，将所述自动驾驶车辆的所述第一侧的静态横向规划约束逐渐从第一静态横向规划约束变化到第二静态横向规划约束；其中，所述第一静态横向规划约束由所述第三车道的所述第一侧的车道线确定，所述第二静态横向规划约束由所述第四车道的所述第一侧的车道线确定。
  26. 根据权利要求14所述的装置，其特征在于，所述装置还包括控制单元，用于在所述实际横向规划约束的范围内，控制所述自动驾驶车辆在所述第一路点时的所述第一侧的横向位移，所述横向位移为第一方向上的位移，所述第一方向垂直于所述自动驾驶车辆在所述第一路点时的行驶方向。
  27. 一种自动驾驶车辆的横向规划约束确定装置，其特征在于，包括：
    存储器，用于存储程序；
    处理器，用于执行所述存储器存储的程序；当所述存储器存储的程序被执行时，所述处理器执行权利要求1-13任一项所述的方法。
  28. 一种自动驾驶车辆，其特征在于，包括权利要求14-26任一项所述的横向规划约束 确定装置。
  29. 一种计算机可读存储介质，其特征在于，所述计算机可读介质存储用于计算设备执行的指令，所述计算设备执行所述指令时，实现如权利要求1-13任一项所述的方法。
PCT/CN2021/097295 2020-11-24 2021-05-31 一种横向规划约束确定方法及装置 WO2022110740A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21896249.6A EP4250048A1 (en) 2020-11-24 2021-05-31 Method and apparatus for determining transverse planning constraint

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011331404.0 2020-11-24
CN202011331404.0A CN114550474B (zh) 2020-11-24 2020-11-24 一种横向规划约束确定方法及装置

Publications (1)

Publication Number Publication Date
WO2022110740A1 true WO2022110740A1 (zh) 2022-06-02

Family

ID=81660614

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/097295 WO2022110740A1 (zh) 2020-11-24 2021-05-31 一种横向规划约束确定方法及装置

Country Status (3)

Country Link
EP (1) EP4250048A1 (zh)
CN (1) CN114550474B (zh)
WO (1) WO2022110740A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115311635A (zh) * 2022-07-26 2022-11-08 阿波罗智能技术(北京)有限公司 车道线处理方法、装置、设备及存储介质

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116674557B (zh) * 2023-07-31 2023-10-31 福思(杭州)智能科技有限公司 车辆自主变道动态规划方法、装置和域控制器
CN117151596B (zh) * 2023-11-01 2023-12-29 领先未来科技集团有限公司 物联网仓储agv物流管理方法、系统及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014210456A (ja) * 2013-04-17 2014-11-13 本田技研工業株式会社 車両の操舵制御装置
CN106314419A (zh) * 2015-07-01 2017-01-11 丰田自动车株式会社 自动驾驶控制装置
CN111123952A (zh) * 2019-12-31 2020-05-08 华为技术有限公司 一种轨迹规划方法及装置
CN111383474A (zh) * 2018-12-29 2020-07-07 长城汽车股份有限公司 自动驾驶车辆的决策系统及方法
CN111443709A (zh) * 2020-03-09 2020-07-24 北京百度网讯科技有限公司 车辆变道路线规划方法、装置、终端和存储介质

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2404195B1 (de) * 2009-03-04 2018-09-19 Continental Teves AG & Co. OHG Verfahren zur automatischen erkennung eines fahrmanövers eines kraftfahrzeugs und ein dieses verfahren umfassendes fahrerassistenzsystem
US20150175162A1 (en) * 2012-09-07 2015-06-25 Toyota Jidosha Kabushiki Kaisha Inter-vehicle distance control apparatus
EP2848487B1 (en) * 2013-09-12 2020-03-18 Volvo Car Corporation Manoeuvre generation for automated drive
CN104019825A (zh) * 2014-06-23 2014-09-03 中国北方车辆研究所 一种路径规划的确定方法
CN105751999B (zh) * 2016-03-31 2018-07-20 汪家琳 全智能全自动无人驾驶汽车
CN107121980B (zh) * 2017-03-17 2019-07-09 北京理工大学 一种基于虚拟约束的自动驾驶车辆路径规划方法
JP6503585B2 (ja) * 2017-09-01 2019-04-24 本田技研工業株式会社 車両制御システム、車両制御方法、及びプログラム
US10571921B2 (en) * 2017-09-18 2020-02-25 Baidu Usa Llc Path optimization based on constrained smoothing spline for autonomous driving vehicles
US11378961B2 (en) * 2018-04-17 2022-07-05 Baidu Usa Llc Method for generating prediction trajectories of obstacles for autonomous driving vehicles
CN110667578B (zh) * 2018-12-29 2021-09-17 毫末智行科技有限公司 自动驾驶车辆的横向决策系统及横向决策确定方法
CN109799821A (zh) * 2019-01-25 2019-05-24 汉腾汽车有限公司 一种基于状态机的自动驾驶控制方法
CN111267843A (zh) * 2019-06-28 2020-06-12 长城汽车股份有限公司 车辆行驶控制方法及设备
CN111739342B (zh) * 2020-03-26 2022-03-01 腾讯科技（深圳）有限公司 用于避让侧前方车辆的方法、装置、介质以及车辆
CN111829545B (zh) * 2020-09-16 2021-01-08 深圳裹动智驾科技有限公司 自动驾驶车辆及其运动轨迹的动态规划方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014210456A (ja) * 2013-04-17 2014-11-13 本田技研工業株式会社 車両の操舵制御装置
CN106314419A (zh) * 2015-07-01 2017-01-11 丰田自动车株式会社 自动驾驶控制装置
CN111383474A (zh) * 2018-12-29 2020-07-07 长城汽车股份有限公司 自动驾驶车辆的决策系统及方法
CN111123952A (zh) * 2019-12-31 2020-05-08 华为技术有限公司 一种轨迹规划方法及装置
CN111443709A (zh) * 2020-03-09 2020-07-24 北京百度网讯科技有限公司 车辆变道路线规划方法、装置、终端和存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115311635A (zh) * 2022-07-26 2022-11-08 阿波罗智能技术(北京)有限公司 车道线处理方法、装置、设备及存储介质
CN115311635B (zh) * 2022-07-26 2023-08-01 阿波罗智能技术(北京)有限公司 车道线处理方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN114550474A (zh) 2022-05-27
CN114550474B (zh) 2023-03-03
EP4250048A1 (en) 2023-09-27

Similar Documents

Publication Publication Date Title
JP7255782B2 (ja) 障害物回避方法、障害物回避装置、自動運転装置、コンピュータ可読記憶媒体及びプログラム
CN111688663B (zh) 机动交通工具及控制其自动驾驶操作的方法
JP6965297B2 (ja) 車両制御装置、車両制御方法、および車両制御プログラム
JP6569186B2 (ja) 車両制御装置、車両制御方法、および車両制御プログラム
US20220212690A1 (en) Vehicle-mounted device information display method, apparatus, and vehicle
WO2022110740A1 (zh) 一种横向规划约束确定方法及装置
JP6270227B2 (ja) 車両制御装置、車両制御方法、および車両制御プログラム
CN109195845B (zh) 车辆控制系统、车辆控制方法及存储介质
WO2017141396A1 (ja) 車両制御装置、車両制御方法、および車両制御プログラム
WO2017175377A1 (ja) 車両制御システム、車両制御方法、および車両制御プログラム
JP7059652B2 (ja) 車両制御システム
JPWO2017138513A1 (ja) 車両制御装置、車両制御方法、および車両制御プログラム
JPWO2017158768A1 (ja) 車両制御システム、車両制御方法、および車両制御プログラム
JP2017081426A (ja) 車両制御装置、車両制御方法、および車両制御プログラム
JP7172287B2 (ja) 自動運転システム
JP6729326B2 (ja) 自動運転装置
JP6580115B2 (ja) 自動運転車両の走行制御装置
US10976739B2 (en) Vehicle control device
JP2018037100A (ja) 車両制御システム、車両制御方法、および車両制御プログラム
JP2017191551A (ja) 車両制御システム、車両制御方法、および車両制御プログラム
US20210043090A1 (en) Electronic device for vehicle and method for operating the same
JP2021160714A (ja) 車両制御装置および車両制御方法
US11644830B1 (en) Autonomous vehicle remote teleoperations system with scenario selection
JP7077870B2 (ja) 自動運転システム
JP2018176959A (ja) 自動運転システム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21896249

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021896249

Country of ref document: EP

Effective date: 20230619

NENP Non-entry into the national phase

Ref country code: DE