WO2020135881A1 - 自动驾驶车辆的车道选择方法、系统及车辆 - Google Patents

自动驾驶车辆的车道选择方法、系统及车辆 Download PDF

Info

Publication number
WO2020135881A1
WO2020135881A1 PCT/CN2019/130061 CN2019130061W WO2020135881A1 WO 2020135881 A1 WO2020135881 A1 WO 2020135881A1 CN 2019130061 W CN2019130061 W CN 2019130061W WO 2020135881 A1 WO2020135881 A1 WO 2020135881A1
Authority
WO
WIPO (PCT)
Prior art keywords
lane
target
road
normal
vehicle
Prior art date
Application number
PCT/CN2019/130061
Other languages
English (en)
French (fr)
Inventor
甄龙豹
刘洪亮
张凯
葛建勇
常仕伟
卜玉帅
王天培
刘宏伟
Original Assignee
长城汽车股份有限公司
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
Priority claimed from CN201811636519.3A external-priority patent/CN110619758B/zh
Priority claimed from CN201811635557.7A external-priority patent/CN110619757A/zh
Priority claimed from CN201811643942.6A external-priority patent/CN110675654A/zh
Application filed by 长城汽车股份有限公司 filed Critical 长城汽车股份有限公司
Publication of WO2020135881A1 publication Critical patent/WO2020135881A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • 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
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems

Definitions

  • the present invention relates to the technical field of automobiles, and in particular, to a lane selection method, system, and vehicle for autonomous vehicles.
  • Self-driving car refers to: through visual sensors, lidar, millimeter wave radar, ultrasonic radar, monitoring devices and positioning systems and other devices cooperate with each other to realize the automatic operation of the vehicle without any human active operation.
  • the vehicle operation was liberated. It uses computers, sensors, information fusion, communication, artificial intelligence, and automatic control technologies to plan vehicle travel routes and arrive at preset locations in real time.
  • the autonomous vehicle encounters road changes during driving, such as (there are obstacles in front of the road, the ramp is wider, the ramp is narrower, the ramp is bifurcated, and the ramp merges and merges), it is necessary to change lanes to drive.
  • the speed and distance of vehicles or obstacles determine whether it is possible to change lanes.
  • due to road changes if the driving path is not properly planned, it may cause frequent lane changes and have a high safety risk.
  • the present invention aims to propose a lane selection method for autonomous vehicles.
  • the lane selection method of the self-driving vehicle can plan the priority lane of the self-driving vehicle when the self-driving vehicle is driving on the highway ramp (such as the ramp becomes wider) to avoid the inaccurate map positioning, which can cause the automatic driving vehicle to be on the ramp.
  • the planning method of target lane selection conforms to human driving habits.
  • a lane selection method for an autonomous vehicle includes the following steps: acquiring road information in front of the vehicle driving; judging whether the road ahead of the vehicle is the target road according to the road information; if it is, driving in the vehicle Before the target road, select a target lane from the target road according to driving rules; control the vehicle to change lanes to the target lane.
  • the target road includes a ramp-widening road
  • selecting a target lane from the ramp-widening road according to driving rules includes: determining whether the rightmost lane in the ramp-widening road is normal; if so, then The rightmost lane of the ramp widening road is used as the target lane; the target road is a ramp narrowing road, and the target lane is selected from the ramp narrowing road according to driving rules, including: judging the ramp Whether the right-most lane on the narrowed road is normal; if it is, the right-most lane on the narrowed road is used as the target lane; the target road is a ramp-forked road, and then further based on the navigation information
  • a target branch is selected from the ramp bifurcation road, and before the vehicle enters the target branch, a target lane is selected from the target branch according to driving rules.
  • the selecting the target lane from the widened ramp road according to the driving rules further includes: if the rightmost lane on the widened ramp road is abnormal; then the normal lane near the rightmost lane will be approached As the target lane; the selection of the target lane from the narrowing road on the ramp according to the driving rules, further including: if the rightmost lane on the narrowing road is abnormal; then the rightmost lane will be approached The normal lane on the left is used as the target lane; the selection of the target lane from the target branch according to the driving rules includes: judging whether the rightmost lane in the target branch is normal; if so, the target lane The rightmost lane in the branch serves as the target lane.
  • the method further includes: determining the distance between the ramp widened road and the vehicle; when the distance is less than the preset distance, according to The driving rule selects a target lane from the ramp widening road.
  • the ramp widening road includes a ramp widening road on the left, a ramp widening road on the right, and a ramp widening road on both sides.
  • the method further includes: determining the distance between the narrowed road on the ramp and the vehicle; when the distance is less than the preset distance, according to The driving rule selects a target lane from the narrowed road on the ramp.
  • the narrowing ramp road includes a narrowing ramp on the left, a narrowing ramp on the right, and a narrowing ramp on both sides; the selection of the target lane from the target branch according to the driving rules includes: Whether the rightmost lane in the target branch is normal; if so, the rightmost lane in the target branch is used as the target lane.
  • the selection of the target lane from the target branch according to the driving rules further includes: if the rightmost lane in the target branch is abnormal; the normal lane adjacent to the rightmost lane is used as the target Lane.
  • the target lane from the target branch according to the driving rules it further includes: determining the distance between the target branch and the vehicle; when the distance is less than the preset distance, according to the driving rules Select the target lane in the target branch.
  • the lane selection method of an autonomous driving vehicle can plan the autonomous driving vehicle to preferentially drive the target lane when driving on a freeway ramp (eg, the ramp becomes wider) to avoid automatic positioning due to inaccurate map positioning.
  • a freeway ramp eg, the ramp becomes wider
  • the target lane selection planning method conforms to human driving habits.
  • the second object of the present invention is to propose a lane selection method for an autonomous driving vehicle. This method can perform a timely and reasonable lane change when the autonomous driving vehicle is driving on the main road of a highway and when the main road becomes wider. Measures to improve driving safety.
  • a lane selection method for an autonomous driving vehicle includes the following steps: acquiring road information in front of the vehicle driving; judging whether the road ahead of the vehicle driving is widening the main road according to the road information; if so, according to the current The target lane is selected by the driving lane, the number of widened main lanes, and the abnormality of each main lane; the vehicle is controlled to change lanes to the target lane.
  • selecting the target lane according to the current driving lane, the number of main roads after widening, and the abnormal situation of each main road includes: When the second lane is normal, use the second lane as the target lane, and when the second lane is abnormal and the first lane is normal, use the first lane as the target lane; when the main lane is widened
  • select the target lane according to the current driving lane, the number of main lanes after widening, and the abnormal situation of each main lane including: when the second lane is normal, the second lane As the target lane; when the second lane is abnormal, it is further determined whether the third lane is normal; if it is normal and the current driving lane of the vehicle is not the first lane, the third lane is used as the target lane Otherwise, when the first lane is normal and the current driving lane of the vehicle is the first lane, the
  • the automatic driving vehicle of the present invention when the automatic driving vehicle runs on the main road of the expressway, when the main road widens, a timely and reasonable lane change measure can be taken to improve driving safety.
  • the third object of the present invention is to propose a lane selection method for autonomous vehicles. This method can avoid and avoid the narrowing of the main road when the autonomous vehicle is driving on the main road of the highway. There are too many lane changes of vehicles to improve driving safety.
  • a lane selection method for an autonomous driving vehicle includes the following steps: acquiring road information ahead of the vehicle driving; judging whether the road ahead of the vehicle driving is narrowing the main road based on the road information; if so, further determining the current Whether the driving lane of the vehicle is a narrowing lane; if it is, the normal main lane closest to the current driving lane is taken as the target lane; and the vehicle is controlled to change lanes to the target lane.
  • the lane selection method of the automatic driving vehicle when the automatic driving vehicle is driving on the main road of the expressway, when the main road becomes narrow, it can be avoided in a timely manner, and the vehicle can avoid excessive lane changes and improve driving Safety.
  • the fourth object of the present invention is to propose a lane selection method for autonomous vehicles.
  • This method can plan the priority lane of the autonomous driving vehicle when driving on the main highway of the highway (such as the main road bifurcation), to avoid the vehicle from changing the lane too much, and improve the driving safety and reliability of the autonomous driving vehicle. .
  • a lane selection method for an autonomous driving vehicle includes the following steps: acquiring road information in front of the vehicle driving; judging whether the road ahead of the vehicle driving is based on the road information whether the road in front of the main road is bifurcated; The information selects a target branch from the main road bifurcation road; selects a target lane according to the number of main roads in the target branch and the abnormality of each main road; and controls the vehicle to change lanes to the target lane.
  • the selecting the target lane according to the number of main lanes in the target branch and the abnormality of each main lane includes: when the number of main lanes in the target branch is 1 lane and the first lane is normal, Use the first lane as the target lane; when the number of main lanes in the target branch is 2 lanes and the second lane is normal, use the second lane as the target lane, otherwise, judge the first lane further Whether the lane is normal, and when the first lane is normal, use the first lane as the target lane; when the number of main lanes in the target branch is 3 lanes and the second lane is normal, change the second lane
  • the lane is used as the target lane, otherwise it is further determined whether the third lane is normal, and when the third lane is normal, the third lane is used as the target lane, otherwise, it is further determined whether the first lane is normal and is located in the first lane When normal, the first lane is used as the target lane; when the target branch is the right branch and the
  • the lane selection method of the self-driving vehicle of the present invention can plan the automatic driving vehicle to preferentially drive the target lane when the self-driving vehicle is driving on the main road of the highway (such as the main road bifurcation), to avoid the vehicle from changing the lane too large and improving Driving safety and reliability of autonomous vehicles.
  • FIG. 1 is a flowchart of a method for selecting a lane of an autonomous driving vehicle according to an embodiment of the invention
  • FIG. 2 is a schematic diagram of a ramp widening in the lane selection method of an automatic driving vehicle according to an embodiment of the present invention
  • 3A-3C are flowcharts of the lane selection method of an autonomous driving vehicle according to an embodiment of the present invention when the second lane, the third lane, and the fourth lane are respectively;
  • FIG. 4 is a flowchart of a method for selecting a lane of an autonomous driving vehicle according to another embodiment of the invention.
  • 5A-5C are flowcharts of a method for selecting a lane for an autonomous driving vehicle according to another embodiment of the present invention when the second lane, the third lane, and the fourth lane are respectively;
  • FIG. 6 is a structural block diagram of a lane selection system for an autonomous driving vehicle according to an embodiment of the invention.
  • FIG. 7 is a flowchart of a lane selection method of an autonomous driving vehicle according to an embodiment of the invention.
  • FIG. 8 is a schematic diagram of narrowing a ramp in the lane selection method of an autonomous driving vehicle according to an embodiment of the present invention.
  • 9A-9C are flowcharts of a method for selecting a lane of an autonomous driving vehicle according to an embodiment of the present invention when the number of lanes is 2, 3, and 4, respectively;
  • FIG. 10 is a flowchart of a method for selecting a lane of an autonomous driving vehicle according to another embodiment of the invention.
  • FIG. 11 is a structural block diagram of a lane selection system for an autonomous driving vehicle according to an embodiment of the invention.
  • FIG. 12 is a flowchart of a lane selection method of an autonomous driving vehicle according to an embodiment of the invention.
  • FIG. 13 is a schematic diagram of ramp bifurcation in the lane selection method of an autonomous driving vehicle according to an embodiment of the present invention.
  • 14A-14C are flowcharts of target lane selection by applying the lane selection method of the self-driving vehicle according to the embodiment of the present invention for different numbers of forked ramps;
  • 15 is a flowchart of a lane selection method of an autonomous driving vehicle according to another embodiment of the invention.
  • 16A-16C are flow charts of target lane selection by applying the lane selection method of the autonomous driving vehicle according to the embodiment of the present invention for different numbers of bifurcated main lanes;
  • 17 is a structural block diagram of a lane selection system of an autonomous driving vehicle according to an embodiment of the present invention.
  • the lane selection method of an automatic driving vehicle includes: acquiring road information ahead of the vehicle driving; judging whether the road ahead of the vehicle driving is the target road according to the road information; if so, then Before the vehicle enters the target road, select a target lane from the target road according to the driving rules; control the vehicle to change lanes to the target lane.
  • selecting the target lane from the ramp-widening road according to the driving rules includes: determining whether the rightmost lane on the ramp-widening road is normal; if so, then Use the rightmost lane of the ramp-on road as the target lane; if the target road is a ramp-on narrow road, then select the target lane from the ramp-on narrow road according to the driving rules, including: Whether the rightmost lane on the ramp-narrowing road is normal; if it is, then the right-most lane on the ramp-narrowing road is used as the target lane; the target road is a ramp-forked road, further according to the navigation
  • the information selects a target branch from the ramp bifurcation road, and before the vehicle enters the target branch, selects a target lane from the target branch according to driving rules.
  • the selecting a target lane from the ramp-up road according to a driving rule further includes: if the right-most lane on the ramp-up road is abnormal; The normal lane of the right lane is used as the target lane; the selection of the target lane from the narrowed road on the ramp according to the driving rules also includes: if the rightmost lane on the narrowed road on the ramp is abnormal; The left normal lane of the rightmost lane is used as the target lane; the selection of the target lane from the target branch according to the driving rules includes: determining whether the rightmost lane in the target branch is normal; if it is , Then the rightmost lane in the target branch is used as the target lane.
  • the method before selecting the target lane from the ramp-up road according to the driving rules, the method further includes: determining the distance between the ramp-up road and the vehicle; when the distance is less than When the distance is preset, the target lane is selected from the ramp widening road according to the driving rules.
  • the ramp-widening road includes a ramp-on-left road, a ramp-on-right road, and a ramp-on-two-side road.
  • the method before selecting the target lane from the narrowing road on the ramp according to the driving rules, the method further includes: determining the distance between the narrowing road on the ramp and the vehicle; when the distance is less than When the distance is preset, the target lane is selected from the narrowed road on the ramp according to the driving rules.
  • the ramp-on narrowing road includes a ramp-on narrowing road on the left, a ramp-on narrowing road on the right, and a ramp-on-narrowing road on both sides;
  • the target is selected from the target branches according to driving rules
  • the lane includes: judging whether the rightmost lane in the target branch is normal; if it is, then taking the rightmost lane in the target branch as the target lane.
  • the selecting the target lane from the target branch according to the driving rule further includes: if the rightmost lane in the target branch is abnormal; The normal lane is used as the target lane.
  • the method before selecting a target lane from the target branch according to the driving rules, the method further includes: determining the distance between the target branch and the vehicle; when the distance is less than a preset distance , Select the target lane from the target branch according to the driving rules.
  • FIG. 1 is a flowchart of a lane selection method of an automatic driving vehicle according to an embodiment of the present invention.
  • the lane selection method of an autonomous driving vehicle includes the following steps:
  • S101 Acquire road information ahead of the vehicle.
  • S102 Determine, according to the road information, whether the road ahead of the vehicle is a ramp-widening road.
  • S104 Control the vehicle to change lanes to the target lane.
  • ramp wide refers to autonomous driving vehicles driving on ramps, the number of roads in front increases, which includes three cases of widening on the left side, widening on the right side, and widening on both sides, namely: widening ramps include the left side Ramp widened roads, right ramp widened roads, and ramp widened roads on both sides.
  • the lane selection method of the autonomous driving vehicle can plan the autonomous driving vehicle to preferentially drive the target lane when driving on the highway ramp (for example: the ramp becomes wider) to avoid the automatic positioning of the target lane due to inaccurate map positioning.
  • the target lane selection planning method conforms to human driving habits.
  • the selecting a target lane from the ramp-up road according to the driving rules includes: determining whether the rightmost lane on the ramp-up road is normal; if so, widening the ramp The rightmost lane in the road serves as the target lane.
  • the selecting a target lane from the widened ramp road according to the driving rules further includes: if the rightmost lane on the widened ramp road is abnormal; then the normal lane near the rightmost lane will be approached As the target lane.
  • the method further includes: determining the distance between the ramp-up road and the vehicle; when the distance is less than a preset distance, Select a target lane from the ramp widening road according to the driving rules.
  • FIG. 2 a schematic diagram of the left ramp widening road is shown.
  • the principle of selecting the target lane for widening the ramp is as follows:
  • the lane change can be performed when the distance is preset in advance, where the preset distance is, for example, 500 meters. It should be noted that 500 meters is only exemplary, and other distances such as 300 meters, 400 meters, and 600 meters can also be set.
  • the vehicle runs along the current road, where it is necessary to determine whether the second lane is normal, If it is normal, the target lane is the second lane. If it is not normal, it is further determined whether the left lane, that is: the first lane (ie: 1 lane) is normal. If the first lane is normal, the first lane is regarded as the target lane .
  • the normal lane refers to a lane that can pass smoothly, such as obstacle-free objects, etc., which is opposite to the abnormal lane, where the abnormal refers to the presence of static objects in the target lane (such as roadblocks, road cones, and vehicles that cannot move accidents, etc.)
  • the lane where the lane is impassable where the lane is impassable.
  • priority is given to judging whether the third lane (ie, the three lanes, in this example, the rightmost lane) is normal, and if normal, the third lane is used as the target lane, Otherwise, it is determined whether the second lane is normal. If it is normal, the second lane is regarded as the target lane; otherwise, it is further determined whether the first lane is normal. If it is normal, the first lane is regarded as the target lane.
  • priority is given to determining whether the fourth lane (that is, the four lanes, in this example, the rightmost lane) is normal. If it is normal, the fourth lane is used as the target lane. Otherwise, determine whether the third lane is normal. If it is normal, then use the third lane as the target lane. Otherwise, determine whether the second lane is normal. If it is normal, then use the second lane as the target lane. Otherwise, determine whether the first lane is normal. If it is normal, the first lane is taken as the target lane.
  • setting the 0 lane means that all lanes are abnormal, and the target lane selection is invalid. At this time, it can be handed over to the person to control the vehicle.
  • the above is the method of road selection when the ramp is widened, and the method of road selection when the main road is widened is described below.
  • the main road is, for example, the main road of the expressway.
  • an embodiment of the present invention discloses a lane selection method for an autonomous driving vehicle, which is characterized by including the following steps:
  • S402 Determine, according to the road information, whether the road ahead of the vehicle travels is a road with a wider main road;
  • the automatic driving vehicle of the present invention when the automatic driving vehicle runs on the main road of the expressway, when the main road widens, a timely and reasonable lane change measure can be taken to improve driving safety.
  • the second lane as an example, that is, when the number of main lanes after the main road becomes wider and the road becomes wider is 2, according to the current driving lane, the widened
  • the number of main lanes and the abnormal situation of each main lane select the target lane, including: when the second lane is normal, the second lane is used as the target lane, and when the second lane is abnormal and the first lane is normal, the target lane is selected.
  • the first lane is used as the target lane.
  • each Selecting the target lane for the abnormal situation of the main lane includes: when the second lane is normal, the second lane is used as the target lane; when the second lane is abnormal, it is further determined whether the third lane is normal; if it is normal and the vehicle Of the current driving lane is not the first lane, the third lane is used as the target lane, otherwise when the first lane is normal and the current driving lane of the vehicle is the first lane, the first lane As the target lane; if both the second lane and the third lane are abnormal, the first lane is used as the target lane when the first lane is normal.
  • the fourth lane As shown in FIG. 5C, taking the fourth lane as an example, that is, when the main road becomes wider and the number of main roads after widening is 4, according to the current driving lane, the number of main roads after widening, and each Select the target lane for the abnormal situation of the main lane, including: when the second lane is normal, the third lane is abnormal, the fourth lane is normal and the current driving lane is the fourth lane, the fourth lane is regarded as the target lane, otherwise , Using the second lane as the target lane; when the second lane is abnormal, the third lane is abnormal, the fourth lane is abnormal, and the first lane is normal, the first lane is used as the target lane; when the second lane is abnormal 3.
  • the fourth lane is normal, the first lane is normal, and the current driving lane is the second lane, the first lane is regarded as the target lane; otherwise, the fourth lane is regarded as the target lane;
  • the third lane is normal, the first lane is normal and the current driving lane is the first lane, the first lane is regarded as the target lane, otherwise, the third lane is regarded as the target Lane.
  • setting the 0 lane means that all lanes are abnormal, and the target lane selection is invalid. At this time, it can be handed over to the person to control the vehicle.
  • the first to fourth lanes are the lanes that are sorted sequentially from left to right.
  • the lane selection system 100 of an autonomous driving vehicle according to an embodiment of the present invention includes: an acquisition module 110, an identification module 120, and a control module 130.
  • the obtaining module 110 is used to obtain road information in front of the vehicle driving; the identifying module 120 is used to determine whether the road ahead of the vehicle is a ramp-widening road according to the road information; the control module 130 is used in the identifying module 120 When identifying that the road ahead of the vehicle is a ramp widening road, before the vehicle enters the ramp widening road, select a target lane from the ramp widening road according to the driving rules, and control the vehicle to The target lane changes lanes.
  • the lane selection system of the self-driving vehicle can plan the priority lane of the self-driving vehicle when the self-driving vehicle runs on a freeway ramp (eg, the ramp becomes wider) to avoid the inaccurate map positioning.
  • a freeway ramp eg, the ramp becomes wider
  • the risk of collisions with other vehicles caused by large lateral deviations of autonomous driving vehicles on the ramp ensures that the vehicles can drive at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • control module 130 is configured to: determine whether the rightmost lane on the ramp widened road is normal; if so, change the rightmost lane on the ramp widened road As the target lane.
  • control module 130 is further configured to: if the rightmost lane on the widened ramp is abnormal; then use the normal lane adjacent to the rightmost lane as the target lane .
  • control module 130 is further configured to determine the distance between the ramp-on-road and the vehicle before selecting a target lane from the ramp-on-road on the basis of driving rules, When the distance is less than the preset distance, a target lane is selected from the ramp widening road according to driving rules.
  • an embodiment of the present invention discloses another lane selection system for an autonomous driving vehicle, which includes: a first acquisition module for acquiring road information ahead of the vehicle; and a judgment module for judging the vehicle based on the road information Whether the road in front of the vehicle driving is the main road widening road; the first control module is used to: when the road in front of the vehicle driving is the main road widening road, according to the current driving lane and the number of widened main roads And the abnormal situation of each main lane selects the target lane and controls the vehicle to change lanes to the target lane.
  • the first control module when the number of main roads after the main road becomes wider and the road becomes wider is 2, the first control module is used to: when the second lane is normal, use the second lane as In the target lane, when the second lane is abnormal and the first lane is normal, the first lane is used as the target lane.
  • the first control module when the number of main lanes after the main road becomes wider and the road becomes wider is 3, the first control module is used to: when the second lane is normal, use the second lane as The target lane; when the second lane is abnormal, further determine whether the third lane is normal; if it is normal and the current driving lane of the vehicle is not the first lane, the third lane is used as the target lane, Otherwise, when the first lane is normal and the current driving lane of the vehicle is the first lane, the first lane is used as the target lane; if both the second lane and the third lane are abnormal, the When a lane is normal, the first lane is used as the target lane.
  • the first control module when the number of main roads after the main road becomes wider and the road becomes wider is 4, the first control module is used to: when the second lane is normal, the third lane is abnormal, and the fourth When the lane is normal and the current driving lane is the fourth lane, the fourth lane is regarded as the target lane; otherwise, the second lane is regarded as the target lane; when the second lane is abnormal, the third lane is abnormal, the fourth When the lane is abnormal and the first lane is normal, the first lane is used as the target lane; when the second lane is abnormal, the third lane is abnormal, the fourth lane is normal, the first lane is normal, and the current driving lane is the second lane When the first lane is used as the target lane, otherwise, the fourth lane is used as the target lane; when the second lane is abnormal, the third lane is normal, the first lane is normal and the current driving lane is the first lane At this time, the first lane is regarded as the target lane;
  • the lane selection system of the self-driving vehicle of the present invention when the self-driving vehicle is driving on the main road of the expressway, when the main road becomes wider, it can take timely and reasonable lane-changing measures to improve driving safety.
  • an embodiment of the present invention discloses a vehicle provided with a lane selection system for an autonomous driving vehicle as in any of the above embodiments.
  • the vehicle can plan the priority lane of the self-driving vehicle when driving on a freeway ramp (eg, the ramp becomes wider) to avoid large lateral deviations caused by the automatic driving vehicle on the ramp due to inaccurate map positioning
  • the resulting danger of collision with other vehicles ensures that the vehicle can travel at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • FIG. 7 is a flowchart of a lane selection method of an autonomous driving vehicle according to an embodiment of the present invention.
  • the lane selection method of an autonomous driving vehicle includes the following steps:
  • S701 Acquire road information ahead of the vehicle.
  • S702 Determine, based on the road information, whether the road ahead of the vehicle is a narrow-ramp road.
  • the driving rule is the right driving rule.
  • the narrowing of the ramp refers to the autonomous driving vehicle driving on the ramp, the number of roads in front decreases, which includes three cases of narrowing on the left side, narrowing on the right side, and narrowing on both sides, that is: the narrowing road on the ramp includes the left side Ramp narrowing road, right ramp narrowing road and both ramp narrowing roads.
  • the lane selection method for an autonomous driving vehicle can plan for the autonomous driving vehicle to preferentially drive the target lane when driving on a freeway ramp (eg, the ramp becomes narrower) to avoid inaccurate map positioning.
  • a freeway ramp eg, the ramp becomes narrower
  • the risk of collisions with other vehicles caused by large lateral deviations of autonomous driving vehicles on the ramp ensures that the vehicles can drive at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • static objects such as roadblocks, road cones, unmovable accident vehicles, etc.
  • the selecting a target lane from the narrowing road on the ramp according to the driving rule includes: determining whether the rightmost lane on the narrowing road on the ramp is normal; if it is, narrowing the ramp The rightmost lane in the road serves as the target lane.
  • the selection of the target lane from the narrowed road on the ramp according to the driving rule further includes: if the rightmost lane on the narrowed road on the ramp is abnormal; then it will approach the left side of the rightmost lane The normal lane is used as the target lane.
  • the method further includes: determining the distance between the narrowing road on the ramp and the vehicle; when the distance is less than a preset distance, Select the target lane from the narrowing ramp according to the driving rules
  • FIG. 8 a schematic diagram of a narrow road on the right ramp is shown.
  • the principle of selecting the target lane for narrowing the ramp is as follows:
  • the lane change can be performed when the distance is preset in advance, where the preset distance is, for example, 500 meters. It should be noted that 500 meters is only exemplary, and other distances such as 300 meters, 400 meters, and 600 meters can also be set.
  • the normal lane refers to a lane that can pass smoothly, such as obstacle-free objects, etc., which is opposite to the abnormal lane, where the abnormal refers to the presence of static objects in the target lane (such as roadblocks, road cones, and vehicles that cannot move accidents, etc.)
  • the lane where the lane is impassable is impassable.
  • the priority is to determine whether the 3 lanes are normal, if it is normal, it becomes 3 lanes, if 3 If the lane is abnormal, it is further determined whether the two lanes are normal, and if it is normal, it becomes the two lanes. Similarly, if the right narrows, priority is given to determining whether lane 2 is normal. If it is normal, it becomes lane two. If lane two is abnormal, it is further determined whether lane one is normal. If it is normal, lane one becomes.
  • setting the 0 lane means that all lanes are abnormal, and the target lane selection is invalid. At this time, it can be handed over to the person to control the vehicle.
  • the above is the method of road selection when the ramp is narrowed, and the method of road selection when the main road is narrowed is described below.
  • the main road is, for example, the main road of the expressway.
  • an embodiment of the present invention discloses a lane selection method for an autonomous driving vehicle, which is characterized by including the following steps:
  • S1002 Determine, according to the road information, whether the road ahead of the vehicle is narrowing the main road;
  • the lane selection method of the self-driving vehicle further includes: if the current driving lane is not a narrowing lane, keeping the current driving lane.
  • the principle of lane change is that if the current driving lane disappears ahead, that is, the current driving lane is a narrowing lane, the normal main lane closest to the current driving lane is taken as the target lane.
  • four lanes namely: 1 lane, 2 lanes, 3 lanes and 4 lanes, assuming that the vehicle is driving in 1 lane and the 1 lane in front disappears, first determine whether the 2 lane is normal, if it is normal, it becomes 2 lanes, if If the 2-lane is abnormal, it is further determined whether the 3-lane is normal. If it is normal, it becomes the 3-lane. If the 3-lane is also abnormal, it is further determined whether the 4-lane is normal. If it is normal, it becomes the 4-lane driving.
  • the lane selection method of the automatic driving vehicle when the automatic driving vehicle is driving on the main road of the expressway, when the main road becomes narrow, it can be avoided in a timely manner, and the vehicle can avoid excessive lane changes and improve driving Safety.
  • the lane selection system 1100 of an autonomous driving vehicle according to an embodiment of the present invention includes: an acquisition module 1110, an identification module 1120, and a control module 1130.
  • the obtaining module 1110 is used to obtain road information in front of the vehicle driving; the identifying module 1120 is used to determine whether the road ahead of the vehicle is narrowing on a ramp according to the road information; the control module 1130 is used in the identifying module 1120 When identifying that the road ahead of the vehicle is a narrowing ramp, before the vehicle enters the narrowing ramp, select a target lane from the narrowing ramp according to the driving rules, and control the vehicle Change lanes to the target lane.
  • the lane selection system of the self-driving vehicle can plan the priority lane of the self-driving vehicle when the self-driving vehicle is driving on the highway ramp (eg, the ramp is narrowed) to avoid the inaccurate map positioning.
  • the risk of collisions with other vehicles caused by large lateral deviations of autonomous driving vehicles on the ramp ensures that the vehicles can drive at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • control module 1130 is configured to: determine whether the rightmost lane on the narrowing road is normal; if so, narrow the rightmost lane on the narrowing road As the target lane.
  • control module 1130 is further configured to: if the rightmost lane on the narrowed road is abnormal; then use the left normal lane adjacent to the rightmost lane as the Target lane.
  • control module 1130 is further configured to determine the distance between the ramp-narrowing road and the vehicle before selecting a target lane from the ramp-narrowing road according to driving rules, When the distance is less than the preset distance, a target lane is selected from the narrowing road on the ramp according to the driving rules.
  • an embodiment of the present invention discloses another lane selection system for an autonomous driving vehicle, which includes: a first acquisition module for acquiring road information in front of the vehicle; a first identification module for determining the road information according to the road information Determine whether the road ahead of the vehicle is narrowing the main lane; the first control module is used to further determine whether the current driving lane is the narrowing lane when the road ahead of the vehicle is narrowing the main lane If yes, the normal main lane closest to the current driving lane is taken as the target lane, and the vehicle is controlled to change lanes to the target lane.
  • the first control module is also used to maintain the current driving lane when the current driving lane is not a narrowing lane.
  • an embodiment of the present invention discloses a vehicle provided with a lane selection system for an autonomous driving vehicle as in any of the above embodiments.
  • the vehicle can plan the priority lane of the autonomous driving vehicle when driving on a freeway ramp (such as a narrowing ramp) to avoid large lateral deviations caused by the autonomous driving vehicle on the ramp due to inaccurate map positioning
  • a freeway ramp such as a narrowing ramp
  • the resulting danger of collision with other vehicles ensures that the vehicle can travel at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • FIG. 12 is a flowchart of a lane selection method of an autonomous driving vehicle according to an embodiment of the present invention.
  • the lane selection method of an autonomous driving vehicle includes the following steps:
  • S1201 Acquire road information ahead of the vehicle.
  • S1202 Determine whether the road ahead of the vehicle is a ramp road based on the road information.
  • S1203 If yes, further select a target branch from the ramp bifurcation road according to navigation information, and select a target lane from the target branch according to driving rules before the vehicle enters the target branch.
  • the lane selection method of an autonomous driving vehicle can plan the autonomous driving vehicle to preferentially drive the target lane when driving on a freeway ramp (such as a ramp bifurcation) to avoid inaccurate map positioning.
  • a freeway ramp such as a ramp bifurcation
  • the risk of collisions with other vehicles caused by large lateral deviations of autonomous driving vehicles on the ramp ensures that the vehicles can drive at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • static objects such as roadblocks, road cones, unmovable accident vehicles, etc.
  • the selection of the target lane from the target branch according to the driving rule includes: determining whether the rightmost lane in the target branch is normal; if so, the rightmost lane in the target branch The lane serves as the target lane.
  • the selection of the target lane from the target branch according to the driving rule further includes: if the rightmost lane in the target branch is abnormal; the normal lane adjacent to the rightmost lane is used as the target Lane.
  • the method before selecting the target lane from the target branch according to the driving rules, the method further includes: determining the distance between the target branch and the vehicle; when the distance is less than the preset distance, according to the driving rules Select the target lane from the target branch.
  • FIG. 13 a schematic diagram of a ramp bifurcation road is shown.
  • the principle of selecting the target lane for ramp divergence is as follows:
  • the preset distance is, for example, 500 meters. It should be noted that 500 meters is only an example. In other examples, it can also be set to 300 meters, 400 meters, and 600. Meters and other distances.
  • the original ramp is 4 ramps, namely: the first lane (ie: 1 lane), the second lane (ie: 2 lanes), the third lane (ie: 3 lanes) and the fourth lane ( That is: 4 lanes), as shown in FIG. 14A, when driving to the right after bifurcation, no matter how many lanes are on the right branch, the normal lane on the far right is the target lane, for example: there is 1 lane on the right branch, Namely: the fourth lane, first determine whether the fourth lane is normal, if it is normal, then use the fourth lane as the target lane, otherwise, determine whether the third lane, the second lane, and the first lane are normal in sequence.
  • the third lane is the target lane. If the third lane is abnormal and the second lane is normal, the second lane is the target lane. If the second lane is abnormal and the first lane is normal, the first lane is the target lane. Lane.
  • the first lane when driving to the left after bifurcation and when there is one lane on the left branch, that is, the first lane, first determine whether the first lane is normal, if it is normal, then use the first lane as the target lane, Otherwise, determine whether the second lane, the third lane, and the fourth lane are normal in sequence. If the second lane is normal, the second lane is regarded as the target lane. If the second lane is abnormal and the third lane is normal, the third lane is normal. The lane is the target lane. If the third lane is abnormal and the fourth lane is normal, the fourth lane is regarded as the target lane.
  • setting the 0 lane means that all lanes are abnormal, and the target lane selection fails, at this time, it can be handed over to the person to control the vehicle.
  • the normal lane refers to a lane that can pass smoothly, for example: obstacle-free objects, etc.
  • the lane selection method of the self-driving vehicle according to the embodiment of the present invention can change lanes reasonably when encountering a ramp bifurcation, thereby improving driving safety and driving reliability.
  • the above is the method of road selection at the time of the ramp divergence, and the method of road selection at the time of the main road divergence is described below.
  • the main road is, for example, the main road of the expressway.
  • an embodiment of the present invention discloses a lane selection method for an autonomous driving vehicle, which is characterized by including the following steps:
  • S1502 Determine, according to the road information, whether the road ahead of the vehicle travels is a bifurcated main road;
  • S1504 Select a target lane according to the number of main lanes in the target branch and the abnormal situation of each main lane;
  • S1505 Control the vehicle to change lanes to the target lane.
  • selecting the target lane according to the number of main lanes in the target branch and the abnormality of each main lane includes: when the number of main lanes in the target branch is 1 lane And when the first lane is normal, the first lane is regarded as the target lane; when the number of main lanes in the target branch is 2 lanes and the second lane is normal, the second lane is regarded as the target lane Lane, otherwise it is further determined whether the first lane is normal, and when the first lane is normal, the first lane is used as the target lane; when the number of main lanes in the target branch is 3 lanes and the second lane is normal When the second lane is used as the target lane, otherwise it is further judged whether the third lane is normal, and when the third lane is normal, the third lane is regarded as the target lane, otherwise it is further judged whether the first lane is Normal, and when the first lane is normal, use the first lane as the target lane
  • the target branch is the right branch (that is, to the right) and the lanes in the target branch are abnormal
  • the normal lane in the left branch closest to the right branch is used as the State the target lane.
  • the method further includes: when the target branch is a left branch (that is, to the left) and the lanes in the target branch are abnormal, the centered normal lane in the right branch is used as the target Lane.
  • the centering principle is, for example: when there is one lane in the right branch, the lane is the center lane, when there are two lanes in the right branch, the right lane is preferentially the center lane, when the right branch.
  • the centered lane is regarded as the target lane, otherwise, it is determined whether the rightmost lane in the right branch is normal, if it is normal, the rightmost lane is regarded as the target lane Otherwise, when the left-most lane in the right branch is normal, the left-most lane is used as the target lane.
  • the left branch has two lanes, namely: the first lane and the second lane, first determine whether the second lane is normal, if it is normal, the second lane is used as the target lane, otherwise, it is further determined whether the first lane is normal, If it is normal, the first lane is taken as the target lane, otherwise, it is judged whether the fourth lane is normal, if it is normal, then the fourth lane is regarded as the target lane, otherwise, it is judged whether the third lane is normal, if it is normal, then the third lane The lane serves as the target lane.
  • the priority is to determine whether the right lane (ie: the fourth lane) in the right branch is normal. If it is abnormal, then determine the left lane in the right branch (Ie: third lane) is normal.
  • the left branch has 1 lane, that is: the first lane, first determine whether the first lane is normal, if normal, the first lane is used as the target lane, otherwise, the third lane is further judged (ie: inside the right branch The middle lane) is normal, if it is normal, the third lane is taken as the target lane, otherwise, it is judged whether the fourth lane (ie the rightmost lane in the right branch) is normal, if it is normal, the fourth lane is taken as For the target lane, the negative one determines whether the second lane (ie, the leftmost lane in the right branch) is normal. If it is normal, the fourth lane is regarded as the target lane.
  • the first lane, the second lane and the third lane there are 3 lanes on the left branch, namely: the first lane, the second lane and the third lane, then first determine whether the second lane is normal, if it is normal, then use the second lane as the target lane, otherwise, further judge the third Whether the lane is normal, if it is normal, the third lane is regarded as the target lane, otherwise, it is judged whether the first lane is normal, if it is normal, then the first lane is regarded as the target lane, otherwise, whether the fourth lane is normal, if normal, Then take the fourth lane as the target lane.
  • the first to fourth lanes are the lanes that are sequentially ordered from left to right.
  • setting the 0 lane means that all lanes are abnormal, and the target lane selection is invalid. At this time, it can be handed over to the person to control the vehicle.
  • the lane selection method of the automatic driving vehicle according to the embodiment of the present invention can change lanes reasonably when encountering a main road bifurcation, thereby improving driving safety and driving reliability.
  • the lane selection system 1700 of an autonomous driving vehicle according to an embodiment of the present invention includes: an acquisition module 1710, an identification module 1720, and a control module 1730.
  • the obtaining module 1710 is used to obtain the road information ahead of the vehicle driving; the recognition module 1720 is used to determine whether the road ahead of the vehicle is a ramp bifurcation road according to the road information; the control module 1730 is used in the recognition module When recognizing that the road ahead of the vehicle driving is a ramp bifurcation road, further select a target branch from the ramp bifurcation road according to navigation information, and before the vehicle enters the target branch, select the target branch from the target according to the driving rules Select the target lane in the branch.
  • the lane selection system of the self-driving vehicle can plan the priority lane of the self-driving vehicle when driving the highway ramp (such as a ramp bifurcation) of the self-driving vehicle to avoid inaccurate map positioning.
  • the risk of collisions with other vehicles caused by large lateral deviations of autonomous driving vehicles on the ramp ensures that the vehicles can drive at a faster speed under the premise of safety, and the target lane selection planning method conforms to human driving habits.
  • static objects such as roadblocks, road cones, unmovable accident vehicles, etc.
  • control module 1730 is configured to: determine whether the rightmost lane in the target branch is normal; if so, use the rightmost lane in the target branch as the target Lane.
  • control module 1730 is further configured to: if the rightmost lane in the target branch is abnormal; use the normal lane adjacent to the rightmost lane as the target lane.
  • control module 1730 is further used to determine the distance between the target branch and the vehicle before selecting the target lane from the target branch according to driving rules. When the distance is less than the preset distance, the target lane is selected from the target branch according to the driving rules.
  • an embodiment of the present invention discloses another lane selection system for an autonomous driving vehicle, which includes: a first acquisition module for acquiring road information in front of the vehicle; a first identification module for determining the road information according to the road information Judging whether the road ahead of the vehicle is diverging from the main road; the first control module is used to further diverge from the main road according to the navigation information when the road ahead of the vehicle is diverging from the main road.
  • the road selects a target branch, and selects a target lane according to the number of main lanes in the target branch and the abnormality of each main lane, and controls the vehicle to change lanes to the target lane.
  • the first control module is configured to: when the number of main lanes in the target branch is 1 lane and the first lane is normal, use the first lane as the target lane ; When the number of main lanes in the target branch is 2 lanes and the second lane is normal, the second lane is used as the target lane, otherwise it is further determined whether the first lane is normal, and when the first lane is normal , Use the first lane as the target lane; when the number of main lanes in the target branch is 3 lanes and the second lane is normal, use the second lane as the target lane, otherwise, determine the first lane Whether the third lane is normal, and when the third lane is normal, the third lane is regarded as the target lane, otherwise, it is further judged whether the first lane is normal, and when the first lane is normal, the first lane is regarded as all State the target lane.
  • the first control module is further used to: when the target branch is the right branch and the lanes in the target branch are abnormal, then the one in the left branch closest to the right branch The normal lane is used as the target lane.
  • the first control module is further configured to: when the target branch is the left branch and the lanes in the target branch are abnormal, then use the centered normal lane in the right branch as the Target lane; when the target branch is the left branch and the lanes in the target branch are abnormal and the number of lanes in the left branch is 2, the normal lane in the right branch that is farthest from the left branch is taken as the target lane .
  • the lane selection method of the automatic driving vehicle according to the embodiment of the present invention can change lanes reasonably when encountering a main road bifurcation, thereby improving driving safety and driving reliability.
  • an embodiment of the present invention discloses a vehicle provided with a lane selection system for an autonomous driving vehicle as in any of the above embodiments.
  • the vehicle can make a reasonable lane change when encountering a road bifurcation (such as a main road bifurcation and/or a ramp bifurcation), thereby improving driving safety and driving reliability.
  • a road bifurcation such as a main road bifurcation and/or a ramp bifurcation
  • a "computer-readable medium” may be any device that can contain, store, communicate, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.
  • computer-readable media include the following: electrical connections (electronic devices) with one or more wires, portable computer cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable and editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer-readable medium may even be paper or other suitable medium on which the program can be printed, because, for example, by optically scanning the paper or other medium, followed by editing, interpretation, or other appropriate if necessary Process to obtain the program electronically and then store it in computer memory.

Abstract

一种自动驾驶车辆的车道选择方法、系统及车辆。其中，自动驾驶车辆的车道选择方法包括：获取车辆行驶前方的道路信息（S101）；根据道路信息判断车辆行驶前方的道路是否为匝道变宽道路（S102）；如果是，则在车辆驶入匝道变宽道路之前，根据行驶规则从匝道变宽道路中选择目标车道（S103）；控制车辆向目标车道变道（S104）。利用该车道选择方法，当自动驾驶车辆在高速公路匝道行驶时，可以对自动驾驶车辆优先行驶的目标车道进行规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶时出现较大横向偏差而与其它车辆碰撞的危险，并且目标车道选择规划方式符合人的驾驶习惯。

Description

自动驾驶车辆的车道选择方法、系统及车辆
相关申请的交叉引用
本申请要求长城汽车股份有限公司于2018年12月29日提交的中国专利申请号为“201811643942.6”、“201811636519.3”和“201811635557.7”的优先权。
技术领域
本发明涉及汽车技术领域，特别涉及一种自动驾驶车辆的车道选择方法、系统及车辆。
背景技术
自动驾驶汽车指：通过视觉传感器、激光雷达、毫米波雷达、超声波雷达、监控装置和定位系统等设备相互协同合作，实现没有任何人类主动的操作的情况下，自动地操作车辆将驾驶员从繁重的车辆操作中解放出来。它集中运用了计算机、传感器、信息融合、通讯、人工智能及自动控制等技术，实时规划车辆出行路线并到达预设地点。
自动驾驶车辆在行驶过程中如遇到道路变化，如(道路前方有障碍物、匝道变宽、匝道变窄、匝道分叉、匝道合并交汇时)需要换道行驶，相关技术中，根据周围的车辆或者障碍物的速度和距离等情况判断是否能够换道，然而，由于道路变化，如果不是合理的规划行驶路径，可能导致频繁换道行驶，存在较高的安全隐患。
发明内容
有鉴于此，本发明旨在提出一种自动驾驶车辆的车道选择方法。该自动驾驶车辆的车道选择方法可以对自动驾驶车辆高速公路匝道(如：匝道变宽)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，并且目标车道选择规划方式符合人的驾驶习惯。
为达到上述目的，本发明的技术方案是这样实现的：
一种自动驾驶车辆的车道选择方法，包括以下步骤：获取车辆行驶前方的道路信息；根据所述道路信息判断所述车辆行驶前方的道路是否为目标道路；如果是，则在所述车辆驶入所述目标道路之前，根据行驶规则从所述目标道路中选择目标车道；控制所述车辆向所述目标车道变道。
进一步的，所述目标道路包括匝道变宽道路，根据行驶规则从所述匝道变宽道路中选择目标车道，包括：判断所述匝道变宽道路中的最右侧车道是否正常；如果是，则将所述匝道变宽道路中的最右侧车道作为所述目标车道；所述目标道路为匝道变窄道路，根据行驶规则从所述匝道变窄道路中选择目标车道，包括：判断所述匝道变窄道路中的最右侧车道是否正常；如果是，则将所述匝道变窄道路中的最右侧车道作为所述目标车道；所述目标道路为匝道分叉道路，则进一步根据导航信息从所述匝道分叉道路选择目标分支，并在所述车辆驶入所述目标分支之前，根据行驶规则从所述目标分支中选择目标车道。
进一步的，所述根据行驶规则从所述匝道变宽道路中选择目标车道，还包括：如果所述匝道变宽道路中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道；所述根据行驶规则从所述匝道变窄道路中选择目标车道，还包括：如果所述匝道变窄道路中的最右侧车道异常；则将临近所述最右侧车道的左侧正常车道作为所述目标车道；所述根据行驶规则从所述目标分支中选择目标车道，包括：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
进一步的，在根据行驶规则从所述匝道变宽道路中选择目标车道之前，还包括：判断所述匝道变宽道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变宽道路中选择目标车道。
进一步的，所述匝道变宽道路包括左侧匝道变宽道路、右侧匝道变宽道路和两侧匝道变宽道路。
进一步的，在根据行驶规则从所述匝道变窄道路中选择目标车道之前，还包括：判断所述匝道变窄道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变窄道路中选择目标车道。
进一步的，所述匝道变窄道路包括左侧匝道变窄道路、右侧匝道变窄道路和两侧匝道变窄道路；所述根据行驶规则从所述目标分支中选择目标车道，包括：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
进一步的，所述根据行驶规则从所述目标分支中选择目标车道，还包括：如果所述目标分支中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
进一步的，在根据行驶规则从所述目标分支中选择目标车道之前，还包括：判断所述目标分支与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述目标分支中选择目标车道。
本发明实施例的自动驾驶车辆的车道选择方法，可以对自动驾驶车辆高速公路匝道(如：匝道变宽)行驶时，自动驾驶车辆优先行驶目标车道进行规划，避免由于地图定位 不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，并且目标车道选择规划方式符合人的驾驶习惯。
本发明的第二个目的在于提出一种自动驾驶车辆的车道选择方法，该方法可以在自动驾驶车辆在高速公路主道行驶时，当主道出现变宽情况时，可以进行及时且合理的换道措施，提升行车安全。
为达到上述目的，本发明的技术方案是这样实现的：
一种自动驾驶车辆的车道选择方法，包括以下步骤：获取车辆行驶前方的道路信息；根据所述道路信息判断所述车辆行驶前方的道路是否为主道变宽道路；如果是，则根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道；控制所述车辆向所述目标车道变道。
进一步的，当所述主道变宽道路变宽后的主道数量为2时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：在第二车道正常时将所述第二车道作为所述目标车道，在所述第二车道异常且第一车道正常时将所述第一车道作为所述目标车道；当所述主道变宽道路变宽后的主道数量为3时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：当第二车道正常时将所述第二车道作为所述目标车道；当所述第二车道异常时，进一步判断第三车道是否正常；如果正常且车辆的当前的行驶车道不为第一车道，则将所述第三车道作为所述目标车道，否则当第一车道正常且车辆的当前的行驶车道为第一车道时，则将所述第一车道作为所述目标车道；如果所述第二车道和所述第三车道均异常，则在第一车道正常时将所述第一车道作为所述目标车道；当所述主道变宽道路变宽后的主道数量为4时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：当第二车道正常、第三车道异常、第四车道正常且当前的行驶车道为第四车道时，将所述第四车道作为所述目标车道，否则，将第二车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道异常且第一车道正常时，将所述第一车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道正常、第一车道正常且当前的行驶车道为第二车道时，将所述第一车道作为所述目标车道，否则，将第四车道作为所述目标车道；当第二车道异常、第三车道正常、第一车道正常且当前的行驶车道为第一车道时，将所述第一车道作为所述目标车道，否则，将所述第三车道作为所述目标车道。
本发明的自动驾驶车辆的车道选择方法，自动驾驶车辆在高速公路主道行驶时，当主道出现变宽情况时，可以进行及时且合理的换道措施，提升行车安全。
本发明的第三个目的在于提出一种自动驾驶车辆的车道选择方法，该方法可以对自动 驾驶车辆在高速公路主道行驶时，当主道出现变窄情况时，可以进行及时的规避，且避免车辆出现过多的变道，提升行车安全。
为达到上述目的，本发明的技术方案是这样实现的：
一种自动驾驶车辆的车道选择方法，包括以下步骤：获取车辆行驶前方的道路信息；根据所述道路信息判断所述车辆行驶前方的道路是否为主道变窄道路；如果是，则进一步判断当前的行驶车道是否为变窄车道；如果是，则将最为临近所述当前的行驶车道的正常主道作为目标车道；控制所述车辆向所述目标车道变道。
本发明实施例的自动驾驶车辆的车道选择方法，自动驾驶车辆在高速公路主道行驶时，当主道出现变窄情况时，可以进行及时的规避，且避免车辆出现过多的变道，提升行车安全。
本发明的第四个目的在于提出一种自动驾驶车辆的车道选择方法。该方法可以对自动驾驶车辆高速公路主道(如：主道分叉)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免车辆变道过大，提升自动驾驶车辆的行车安全和可靠性。
为达到上述目的，本发明的技术方案是这样实现的：
一种自动驾驶车辆的车道选择方法，包括以下步骤：获取车辆行驶前方的道路信息；根据所述道路信息判断所述车辆行驶前方的道路是否为主道分叉道路；如果是，则进一步根据导航信息从所述主道分叉道路选择目标分支；根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道；控制所述车辆向所述目标车道变道。
进一步的，所述根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道，包括：当所述目标分支内的主道数量为1条车道且第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为2条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为3条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第三车道是否正常，并在第三车道正常时，将所述第三车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支为右分支且目标分支内的车道均异常时，则将左分支内的最为临近所述右分支的正常车道作为所述目标车道；当所述目标分支为左分支且目标分支内的车道均异常时，则将右分支内居中的正常车道作为所述目标车道。
本发明的自动驾驶车辆的车道选择方法，可以对自动驾驶车辆高速公路主道(如：主道分叉)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免车辆变道过大，提升 自动驾驶车辆的行车安全和可靠性。
附图说明
构成本发明的一部分的附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1为本发明一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图2为本发明一个实施例所述的自动驾驶车辆的车道选择方法中匝道变宽的示意图；
图3A-图3C分别为第二车道、第三车道、第四车道时本发明一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图4为本发明另一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图5A-图5C分别为第二车道、第三车道、第四车道时本发明另一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图6为本发明一个实施例所述的自动驾驶车辆的车道选择系统的结构框图；
图7为本发明一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图8为本发明一个实施例所述的自动驾驶车辆的车道选择方法中匝道变窄的示意图；
图9A-图9C分别为车道数为2、3和4时本发明实施例所述的自动驾驶车辆的车道选择方法的流程图；
图10为本发明另一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图11为本发明一个实施例所述的自动驾驶车辆的车道选择系统的结构框图；
图12为本发明一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图13为本发明一个实施例所述的自动驾驶车辆的车道选择方法中匝道分叉的示意图；
图14A-图14C分别针对不同分叉的匝道数量应用本发明实施例所述的自动驾驶车辆的车道选择方法进行目标车道选择的流程图；
图15为本发明另一个实施例所述的自动驾驶车辆的车道选择方法的流程图；
图16A-图16C分别为针对不同分叉的主道数量应用本发明实施例所述的自动驾驶车辆的车道选择方法进行目标车道选择的流程图；
图17为本发明一个实施例所述的自动驾驶车辆的车道选择系统的结构框图。
具体实施方式
需要说明的是，在不冲突的情况下，本发明中的实施例及实施例中的特征可以相互组合。
下面将参考附图并结合实施例来详细说明本发明。
其中，根据本发明一个实施例的自动驾驶车辆的车道选择方法，包括：获取车辆行驶前方的道路信息；根据所述道路信息判断所述车辆行驶前方的道路是否为目标道路；如果是，则在所述车辆驶入所述目标道路之前，根据行驶规则从所述目标道路中选择目标车道；控制所述车辆向所述目标车道变道。
具体来说，目标道路包括匝道变宽道路，则根据行驶规则从所述匝道变宽道路中选择目标车道，包括：判断所述匝道变宽道路中的最右侧车道是否正常；如果是，则将所述匝道变宽道路中的最右侧车道作为所述目标车道；所述目标道路为匝道变窄道路，则根据行驶规则从所述匝道变窄道路中选择目标车道，包括：判断所述匝道变窄道路中的最右侧车道是否正常；如果是，则将所述匝道变窄道路中的最右侧车道作为所述目标车道；所述目标道路为匝道分叉道路，则进一步根据导航信息从所述匝道分叉道路选择目标分支，并在所述车辆驶入所述目标分支之前，根据行驶规则从所述目标分支中选择目标车道。
在本发明的一个实施例中，所述根据行驶规则从所述匝道变宽道路中选择目标车道，还包括：如果所述匝道变宽道路中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道；所述根据行驶规则从所述匝道变窄道路中选择目标车道，还包括：如果所述匝道变窄道路中的最右侧车道异常；则将临近所述最右侧车道的左侧正常车道作为所述目标车道；所述根据行驶规则从所述目标分支中选择目标车道，包括：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
在本发明的一个实施例中，在根据行驶规则从所述匝道变宽道路中选择目标车道之前，还包括：判断所述匝道变宽道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变宽道路中选择目标车道。
在本发明的一个实施例中，所述匝道变宽道路包括左侧匝道变宽道路、右侧匝道变宽道路和两侧匝道变宽道路。
在本发明的一个实施例中，在根据行驶规则从所述匝道变窄道路中选择目标车道之前，还包括：判断所述匝道变窄道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变窄道路中选择目标车道。
在本发明的一个实施例中，所述匝道变窄道路包括左侧匝道变窄道路、右侧匝道变窄道路和两侧匝道变窄道路；所述根据行驶规则从所述目标分支中选择目标车道，包括：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
在本发明的一个实施例中，所述根据行驶规则从所述目标分支中选择目标车道，还包括：如果所述目标分支中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
在本发明的一个实施例中，在根据行驶规则从所述目标分支中选择目标车道之前，还包括：判断所述目标分支与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述目标分支中选择目标车道。
以下分别对本发明的匝道变宽道路、匝道变窄道路和匝道分叉道路三种道路的具体实现方式进行描述。
[实施例1]
图1是根据本发明一个实施例的自动驾驶车辆的车道选择方法的流程图。
如图1所示，根据本发明一个实施例的自动驾驶车辆的车道选择方法，包括如下步骤：
S101：获取车辆行驶前方的道路信息。
S102：根据所述道路信息判断所述车辆行驶前方的道路是否为匝道变宽道路。
S103：如果是，则在所述车辆驶入所述匝道变宽道路之前，根据行驶规则从所述匝道变宽道路中选择目标车道。
S104：控制所述车辆向所述目标车道变道。
其中，匝道变宽指的是自动驾驶车辆在匝道上面行驶，前方道路数量增多，其包括左侧变宽、右侧变宽及两侧变宽三种情况，即：匝道变宽道路包括左侧匝道变宽道路、右侧匝道变宽道路和两侧匝道变宽道路。
本发明实施例的自动驾驶车辆的车道选择方法，可以对自动驾驶车辆高速公路匝道(如：匝道变宽)行驶时，自动驾驶车辆优先行驶目标车道进行规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在具体示例中，所述根据行驶规则从所述匝道变宽道路中选择目标车道，包括：判断所述匝道变宽道路中的最右侧车道是否正常；如果是，则将所述匝道变宽道路中的最右侧车道作为所述目标车道。
进一步地，所述根据行驶规则从所述匝道变宽道路中选择目标车道，还包括：如果所 述匝道变宽道路中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
更进一步地，在根据行驶规则从所述匝道变宽道路中选择目标车道之前，还包括：判断所述匝道变宽道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变宽道路中选择目标车道。
如图2所示，示出了左侧匝道变宽道路的示意图。其中，匝道变宽的目标车道选择原则如下：
可提前预设距离时进行变道，其中，预设距离例如为500米，需要说明的是，500米仅是示例性的，也可以设定为300米、400米、600米等其它距离。
如图3A所示，当车道数为2，目标车道为最右侧车道，即：第二车道(即：2车道)，车辆沿着当前的道路行驶，其中，需要判断第二车道是否正常，如果正常，则目标车道为第二车道，如果不正常，则进一步判断左侧车道，即：第一车道(即：1车道)是否正常，如果第一车道正常，则将第一车道作为目标车道。
其中，正常车道指：可以顺利通过的车道，例如：无障碍物等，与异常车道相反，其中，异常指由于目标车道存在静态物体目标(如：路障、路锥、无法移动事故车辆等)引起的该车道无法通行的情况的车道。
如图3B所示，当车道数为3时，优先判断第三车道(即：3车道，在该示例中，为最右侧车道)是否正常，如果正常，则将第三车道作为目标车道，否者判断第二车道是否正常，如果正常，则将第二车道作为目标车道，否则进一步判断第一车道是否正常，如果正常，则将第一车道作为目标车道。
如图3C所示，当车道数为4是，优先判断第四车道(即：4车道，在该示例中，为最右侧车道)是否正常，如果正常，则将第四车道作为目标车道，否者判断第三车道是否正常，如果正常，则将第三车道作为目标车道，否则进一步判断第二车道是否正常，如果正常，则将第二车道作为目标车道，否则进一步判断第一车道是否正常，如果正常，则将第一车道作为目标车道。
其中，在图3A至图3C中，置0车道表示所有车道均异常，目标车道选择失效，此时，可以交给人员来操控车辆。
以上是匝道变宽时的道路选择的方法，以下描述主道变宽时的道路选择的方法。其中，主道例如为高速公路的主道。
如图4所示，本发明的实施例公开了一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
S401：获取车辆行驶前方的道路信息；
S402：根据所述道路信息判断所述车辆行驶前方的道路是否为主道变宽道路；
S403：如果是，则根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道；
S404：控制所述车辆向所述目标车道变道。
本发明的自动驾驶车辆的车道选择方法，自动驾驶车辆在高速公路主道行驶时，当主道出现变宽情况时，可以进行及时且合理的换道措施，提升行车安全。
作为一个具体的示例，如图5A所示，以第二车道为例，即：当所述主道变宽道路变宽后的主道数量为2时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：在第二车道正常时将所述第二车道作为所述目标车道，在所述第二车道异常且第一车道正常时将所述第一车道作为所述目标车道。
如图5B所示，以第三车道为例，即：当所述主道变宽道路变宽后的主道数量为3时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：当第二车道正常时将所述第二车道作为所述目标车道；当所述第二车道异常时，进一步判断第三车道是否正常；如果正常且车辆的当前的行驶车道不为第一车道，则将所述第三车道作为所述目标车道，否则当第一车道正常且车辆的当前的行驶车道为第一车道时，则将所述第一车道作为所述目标车道；如果所述第二车道和所述第三车道均异常，则在第一车道正常时将所述第一车道作为所述目标车道。
如图5C所示，以第四车道为例，即：当所述主道变宽道路变宽后的主道数量为4时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：当第二车道正常、第三车道异常、第四车道正常且当前的行驶车道为第四车道时，将所述第四车道作为所述目标车道，否则，将第二车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道异常且第一车道正常时，将所述第一车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道正常、第一车道正常且当前的行驶车道为第二车道时，将所述第一车道作为所述目标车道，否则，将第四车道作为所述目标车道；当第二车道异常、第三车道正常、第一车道正常且当前的行驶车道为第一车道时，将所述第一车道作为所述目标车道，否则，将所述第三车道作为所述目标车道。
其中，在图5A至图5C中，置0车道表示所有车道均异常，目标车道选择失效，此时，可以交给人员来操控车辆。
在以上描述中，第一车道至第四车道为依次从左到右进行排序的车道。
图6是根据本发明一个实施例的自动驾驶车辆的车道选择系统的结构框图。如图6所 示，根据本发明一个实施例的自动驾驶车辆的车道选择系统100，包括：获取模块110、识别模块120和控制模块130。
其中，获取模块110用于获取车辆行驶前方的道路信息；识别模块120用于根据所述道路信息判断所述车辆行驶前方的道路是否为匝道变宽道路；控制模块130用于在所述识别模块120识别车辆行驶前方的道路为匝道变宽道路时，在所述车辆驶入所述匝道变宽道路之前，根据行驶规则从所述匝道变宽道路中选择目标车道，并控制所述车辆向所述目标车道变道。
本发明实施例的自动驾驶车辆的车道选择系统，可以对自动驾驶车辆高速公路匝道(如：匝道变宽)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在本发明的一个实施例中，所述控制模块130用于：判断所述匝道变宽道路中的最右侧车道是否正常；如果是，则将所述匝道变宽道路中的最右侧车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块130还用于：如果所述匝道变宽道路中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块130还用于在根据行驶规则从所述匝道变宽道路中选择目标车道之前，判断所述匝道变宽道路与所述车辆之间的距离，当所述距离小于预设距离时，根据行驶规则从所述匝道变宽道路中选择目标车道。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述。
进一步地，本发明的实施例公开了另一种自动驾驶车辆的车道选择系统，包括：第一获取模块，用于获取车辆行驶前方的道路信息；判断模块，用于根据所述道路信息判断所述车辆行驶前方的道路是否为主道变宽道路；第一控制模块，用于在所述车辆行驶前方的道路为主道变宽道路时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，并控制所述车辆向所述目标车道变道。
在本发明的一个实施例中，当所述主道变宽道路变宽后的主道数量为2时，所述第一控制模块用于：在第二车道正常时将所述第二车道作为所述目标车道，在所述第二车道异 常且第一车道正常时将所述第一车道作为所述目标车道。
在本发明的一个实施例中，当所述主道变宽道路变宽后的主道数量为3时，所述第一控制模块用于：当第二车道正常时将所述第二车道作为所述目标车道；当所述第二车道异常时，进一步判断第三车道是否正常；如果正常且车辆的当前的行驶车道不为第一车道，则将所述第三车道作为所述目标车道，否则当第一车道正常且车辆的当前的行驶车道为第一车道时，则将所述第一车道作为所述目标车道；如果所述第二车道和所述第三车道均异常，则在第一车道正常时将所述第一车道作为所述目标车道。
在本发明的一个实施例中，当所述主道变宽道路变宽后的主道数量为4时，所述第一控制模块用于：当第二车道正常、第三车道异常、第四车道正常且当前的行驶车道为第四车道时，将所述第四车道作为所述目标车道，否则，将第二车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道异常且第一车道正常时，将所述第一车道作为所述目标车道；当第二车道异常、第三车道异常、第四车道正常、第一车道正常且当前的行驶车道为第二车道时，将所述第一车道作为所述目标车道，否则，将第四车道作为所述目标车道；当第二车道异常、第三车道正常、第一车道正常且当前的行驶车道为第一车道时，将所述第一车道作为所述目标车道，否则，将所述第三车道作为所述目标车道。
本发明的自动驾驶车辆的车道选择系统，自动驾驶车辆在高速公路主道行驶时，当主道出现变宽情况时，可以进行及时且合理的换道措施，提升行车安全。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述。
进一步地，本发明的实施例公开了一种车辆，设置有如上述任意一个实施例中的自动驾驶车辆的车道选择系统。该车辆可以对自动驾驶车辆高速公路匝道(如：匝道变宽)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险，和/或，该车辆可以在自动驾驶车辆在高速公路主道行驶时，当主道出现变宽情况时，可以进行及时且可靠的换道措施，提升行车安全。
[实施例二]
下面将参考附图并结合实施例来详细说明本发明。
图7是根据本发明一个实施例的自动驾驶车辆的车道选择方法的流程图。
如图7所示，根据本发明一个实施例的自动驾驶车辆的车道选择方法，包括如下步骤：
S701：获取车辆行驶前方的道路信息。
S702：根据所述道路信息判断所述车辆行驶前方的道路是否为匝道变窄道路。
S703：如果是，则在所述车辆驶入所述匝道变窄道路之前，根据行驶规则从所述匝道变窄道路中选择目标车道。其中，行驶规则为靠右行驶规则。
S704：控制所述车辆向所述目标车道变道。
其中，匝道变窄指的是自动驾驶车辆在匝道上面行驶，前方道路数量减少，其包括左侧变窄、右侧变窄及两侧变窄三种情况，即：匝道变窄道路包括左侧匝道变窄道路、右侧匝道变窄道路和两侧匝道变窄道路。
本发明实施例的自动驾驶车辆的车道选择方法，可以对自动驾驶车辆高速公路匝道(如：匝道变窄)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在具体示例中，所述根据行驶规则从所述匝道变窄道路中选择目标车道，包括：判断所述匝道变窄道路中的最右侧车道是否正常；如果是，则将所述匝道变窄道路中的最右侧车道作为所述目标车道。
进一步地，所述根据行驶规则从所述匝道变窄道路中选择目标车道，还包括：如果所述匝道变窄道路中的最右侧车道异常；则将临近所述最右侧车道的左侧正常车道作为所述目标车道。
更进一步地，在根据行驶规则从所述匝道变窄道路中选择目标车道之前，还包括：判断所述匝道变窄道路与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述匝道变窄道路中选择目标车道
如图8所示，示出了右侧匝道变窄道路的示意图。其中，匝道变窄的目标车道选择原则如下：
可提前预设距离时进行变道，其中，预设距离例如为500米，需要说明的是，500米 仅是示例性的，也可以设定为300米、400米、600米等其它距离。
如果原目标车道为道路变窄车道，则将最为临近该车道的正常车道作为为目标车道。其中，正常车道指：可以顺利通过的车道，例如：无障碍物等，与异常车道相反，其中，异常指由于目标车道存在静态物体目标(如：路障、路锥、无法移动事故车辆等)引起的该车道无法通行的情况的车道。
其中，当匝道数为2，3，4时，匝道变窄目标车道选择流程如图9A-图9C所示，当自动驾驶车辆原行驶匝道车道数变多时，目标车道选择依据此流程原则进行选择。
具体地说，如图9A所示，当为两车道时，即：1车道和2车道，变窄后，例如，左变窄，则判断2车道是否正常，如果正常则变为2车道；同理，如果右变窄，则判断1车道是否正常，如果正常则变为1车道。
如图9B所示，当为三车道时，即：1车道、2车道和3车道，此时，如果左变窄，则优先判断3车道是否正常，如果正常，则变为3车道，如果3车道异常，则进一步判断2车道是否正常，如果正常，则变为2车道。同理，右变窄，则优先判断2车道是否正常，如果正常，则变为2车道，如果2车道异常，则进一步判断1车道是否正常，如果正常，则变为1车道。
如图9C所示，当为四车道时，即：1车道、2车道、3车道和4车道，此时，如果左变窄，则优先判断4车道是否正常，如果正常，则变为4车道，如果4车道异常，则进一步判断3车道是否正常，如果正常，则变为3车道，如果3车道异常，则进一步判断2车道是否正常，如果正常，则变为2车道。同理，右变窄，则优先判断3车道是否正常，如果正常，则变为3车道，如果3车道异常，则进一步判断2车道是否正常，如果正常，则变为2车道，如果2车道异常，则进一步判断1车道是否正常，如果正常，则变为1车道。
其中，在图9A至图9C中，置0车道表示所有车道均异常，目标车道选择失效，此时，可以交给人员来操控车辆。
以上是匝道变窄时的道路选择的方法，以下描述主道变窄时的道路选择的方法。其中，主道例如为高速公路的主道。
如图10所示，本发明的实施例公开了一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
S1001：获取车辆行驶前方的道路信息；
S1002：根据所述道路信息判断所述车辆行驶前方的道路是否为主道变窄道路；
S1003：如果是，则进一步判断当前的行驶车道是否为变窄车道；
S1004：如果是，则将最为临近所述当前的行驶车道的正常主道作为目标车道；
S1005：控制所述车辆向所述目标车道变道。
进一步地，自动驾驶车辆的车道选择方法，还包括：如果当前的行驶车道不为变窄车道，则保持当前的行驶车道行驶。
具体地说，无论几个主道，变道原则为如果当前的行驶车道在前方消失，即：当前的行驶车道为变窄车道，则将最为临近当前的行驶车道的正常主道作为目标车道。
例如：四车道，即：1车道、2车道、3车道和4车道，假设车辆在1车道行驶，前方1车道消失，则首先判断2车道是否正常，如果正常，则变为2车道行驶，如果2车道异常，则进一步判断3车道是否正常，如果正常，则变为3车道，如果3车道也异常，则进一步判断4车道是否正常，如果正常，则变为4车道行驶。
本发明实施例的自动驾驶车辆的车道选择方法，自动驾驶车辆在高速公路主道行驶时，当主道出现变窄情况时，可以进行及时的规避，且避免车辆出现过多的变道，提升行车安全。
图11是根据本发明一个实施例的自动驾驶车辆的车道选择系统的结构框图。如图11所示，根据本发明一个实施例的自动驾驶车辆的车道选择系统1100，包括：获取模块1110、识别模块1120和控制模块1130。
其中，获取模块1110用于获取车辆行驶前方的道路信息；识别模块1120用于根据所述道路信息判断所述车辆行驶前方的道路是否为匝道变窄道路；控制模块1130用于在所述识别模块1120识别所述车辆行驶前方的道路为匝道变窄道路时，在所述车辆驶入所述匝道变窄道路之前，根据行驶规则从所述匝道变窄道路中选择目标车道，并控制所述车辆向所述目标车道变道。
本发明实施例的自动驾驶车辆的车道选择系统，可以对自动驾驶车辆高速公路匝道(如：匝道变窄)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在本发明的一个实施例中，所述控制模块1130用于：判断所述匝道变窄道路中的最右侧车道是否正常；如果是，则将所述匝道变窄道路中的最右侧车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块1130还用于：如果所述匝道变窄道路中的最右侧车道异常；则将临近所述最右侧车道的左侧正常车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块1130还用于在根据行驶规则从所述匝道变窄道路中选择目标车道之前，判断所述匝道变窄道路与所述车辆之间的距离，当所述距离小于预设距离时，根据行驶规则从所述匝道变窄道路中选择目标车道。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述。
进一步地，本发明的实施例公开了另一种自动驾驶车辆的车道选择系统，包括：第一获取模块，用于获取车辆行驶前方的道路信息；第一识别模块，用于根据所述道路信息判断所述车辆行驶前方的道路是否为主道变窄道路；第一控制模块，用于在所述车辆行驶前方的道路为主道变窄道路时，进一步判断当前的行驶车道是否为变窄车道，如果是，则将最为临近所述当前的行驶车道的正常主道作为目标车道，并控制所述车辆向所述目标车道变道。
在本发明的一个实施例中，所述第一控制模块还用于在当前的行驶车道不为变窄车道时，保持当前的行驶车道行驶。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述
进一步地，本发明的实施例公开了一种车辆，设置有如上述任意一个实施例中的自动驾驶车辆的车道选择系统。
该车辆可以对自动驾驶车辆高速公路匝道(如：匝道变窄)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险，和/或，该车辆可以在自动驾驶车辆在高速公路主道行驶时，当主道出现变窄情况时，可以进行及时的规避，且避免车辆出现过多的变道，提升行车安全。
[实施例三]
下面将参考附图并结合实施例来详细说明本发明。
图12是根据本发明一个实施例的自动驾驶车辆的车道选择方法的流程图。
如图12所示，根据本发明一个实施例的自动驾驶车辆的车道选择方法，包括如下步骤：
S1201：获取车辆行驶前方的道路信息。
S1202：根据所述道路信息判断所述车辆行驶前方的道路是否为匝道分叉道路。
S1203：如果是，则进一步根据导航信息从所述匝道分叉道路选择目标分支，并在所述车辆驶入所述目标分支之前，根据行驶规则从所述目标分支中选择目标车道。
S1204：控制所述车辆向所述目标车道变道。
本发明实施例的自动驾驶车辆的车道选择方法，可以对自动驾驶车辆高速公路匝道(如：匝道分叉)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在具体示例中，所述根据行驶规则从所述目标分支中选择目标车道，包括：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
进一步地，所述根据行驶规则从所述目标分支中选择目标车道，还包括：如果所述目标分支中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
在具体示例中，在根据行驶规则从所述目标分支中选择目标车道之前，还包括：判断所述目标分支与所述车辆之间的距离；当所述距离小于预设距离时，根据行驶规则从所述目标分支中选择目标车道。
如图13所示，示出了匝道分叉道路的示意图。其中，匝道分叉的目标车道选择原则如下：
可提前预设距离时进行变道，其中，预设距离例如为500米，需要说明的是，500米仅是示例性的，在其它示例中，也可以设定为300米、400米、600米等其它距离。
作为一个具体的示例，假设原匝道为4条匝道，即：第一车道(即：1车道)、第二车道(即：2车道)、第三车道(即：3车道)和第四车道(即：4车道)，如图14A所示，当分叉后向右行驶时，无论右分支有几条车道，均以最右侧的正常车道为目标车道，例如：右分支有1条车道，即：第四车道，则首先判断第四车道是否正常，如果正常，则将第四车道作为目标车道，否则，依次判断第三车道、第二车道和第一车道是否正常，其中，如 果第三车道正常，则将第三车道作为目标车道，如果第三车道异常而第二车道正常，则将第二车道作为目标车道，如果第二车道异常而第一车道正常，则将第一车道作为目标车道。
如图14B所示，当分叉后向左行驶且当左分支有1条车道时，即：第一车道，则首先判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否则，依次判断第二车道、第三车道和第四车道是否正常，其中，如果第二车道正常，则将第二车道作为目标车道，如果第二车道异常而第三车道正常，则将第三车道作为目标车道，如果第三车道异常而第四车道正常，则将第四车道作为目标车道。
如图14C所示，当分叉后向左行驶且当左分支有2条车道时，即：第一车道和第二车道，则首先判断第二车道是否正常，如果正常，则将第二车道作为目标车道，否则，判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否则，依次判断第三车道和第四车道是否正常，其中，如果第三车道正常，则将第三车道作为目标车道，如果第三车道异常而第四车道正常，则将第四车道作为目标车道。
再次结合图14C所示，当分叉后向左行驶且当左分支有3条车道时，即：第一车道、第二车道和第三车道，则首先判断第三车道是否正常，如果正常，则将第三车道作为目标车道，否则，判断第二车道是否正常，如果正常，则将第二车道作为目标车道，否则，判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否则，如果第四车道正常，则将第四车道作为目标车道。
其中，在图14A至图14C中，置0车道表示所有车道均异常，目标车道选择失效，此时，可以交给人员来操控车辆。
在以上描述中，正常车道指：可以顺利通过的车道，例如：无障碍物等，以异常车道相反，其中，异常指由于目标车道存在静态物体目标(如：路障、路锥、无法移动事故车辆等)引起的该车道无法通行的情况的车道。
本发明实施例的自动驾驶车辆的车道选择方法，可以在遇到匝道分叉时，进行合理的变道，从而提升行车安全和行车的可靠性。
以上是匝道分叉时的道路选择的方法，以下描述主道分叉时的道路选择的方法。其中，主道例如为高速公路的主道。
如图15所示，本发明的实施例公开了一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
S1501：获取车辆行驶前方的道路信息；
S1502：根据所述道路信息判断所述车辆行驶前方的道路是否为主道分叉道路；
S1503：如果是，则进一步根据导航信息从所述主道分叉道路选择目标分支；
S1504：根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道；
S1505：控制所述车辆向所述目标车道变道。
具体地，结合图16A至图16C所示，根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道，包括：当所述目标分支内的主道数量为1条车道且第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为2条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为3条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第三车道是否正常，并在第三车道正常时，将所述第三车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道。
进一步地，如图16A所示，当所述目标分支为右分支(即：向右)且目标分支内的车道均异常时，则将左分支内的最为临近所述右分支的正常车道作为所述目标车道。
如图16B和图16C所示，还包括：当所述目标分支为左分支(即：向左)且目标分支内的车道均异常时，则将右分支内的居中的正常车道作为所述目标车道。
其中，居中原则为例如：当右分支内有1条车道时，则该车道为居中的车道，当右分支内有2条车道时，则优先将右侧车道作为居中的车道，当右分支内有3条车道时，则如果居中的车道正常时，则将居中的车道作为目标车道，否则，接着判断右分支内的最右侧车道是否正常，如果正常，则将最右侧车道作为目标车道，否则在右分支内的最左侧车道正常时，将最左侧车道作为目标车道。
例如：左分支有两条车道，即：第一车道和第二车道，则首先判断第二车道是否正常，如果正常，则将第二车道作为目标车道，否则，进一步判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否者，判断第四车道是否正常，如果正常，则将第四车道作为目标车道，否者判断第三车道是否正常，如果正常，则将第三车道作为目标车道。即：在该种情况下，即：右分支内有两条车道，则优先判断右分支内的右侧车道(即：第四车道)是否正常，异常时，再判断右分支内的左侧车道(即：第三车道)是否正常。
再如：左分支有1条车道，即：第一车道，则首先判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否则，进一步判断第三车道(即：右分支内的中间车道)是否正常，如果正常，则将第三车道作为目标车道，否者，判断第四车道(即：右分支内的最右侧车道)是否正常，如果正常，则将第四车道作为目标车道，否者判断第二车道(即：右分支内的最左侧车道)是否正常，如果正常，则将第四车道作为目标车道。
另外，左分支有3条车道，即：第一车道、第二车道和第三车道，则首先判断第二车 道是否正常，如果正常，则将第二车道作为目标车道，否则，进一步判断第三车道是否正常，如果正常，则将第三车道作为目标车道，否者，判断第一车道是否正常，如果正常，则将第一车道作为目标车道，否者判断第四车道是否正常，如果正常，则将第四车道作为目标车道。
在以上描述中，第一车道至第四车道为从左至右依次排序的车道。
其中，在图16A至图16C中，置0车道表示所有车道均异常，目标车道选择失效，此时，可以交给人员来操控车辆。
本发明实施例的自动驾驶车辆的车道选择方法，可以在遇到主道分叉时，进行合理的变道，从而提升行车安全和行车的可靠性。
图17是根据本发明一个实施例的自动驾驶车辆的车道选择系统的结构框图。如图17所示，根据本发明一个实施例的自动驾驶车辆的车道选择系统1700，包括：获取模块1710、识别模块1720和控制模块1730。
其中，获取模块1710用于获取车辆行驶前方的道路信息；识别模块1720用于根据所述道路信息判断所述车辆行驶前方的道路是否为匝道分叉道路；控制模块1730用于在所述识别模块识别所述车辆行驶前方的道路为匝道分叉道路时，进一步根据导航信息从所述匝道分叉道路选择目标分支，并在所述车辆驶入所述目标分支之前，根据行驶规则从所述目标分支中选择目标车道。
本发明实施例的自动驾驶车辆的车道选择系统，可以对自动驾驶车辆高速公路匝道(如：匝道分叉)行驶时，自动驾驶车辆优先行驶目标车道进行了规划，避免由于地图定位不准，造成自动驾驶车辆在匝道行驶出现较大横向偏差造成的与其它车辆碰撞的危险，保证车辆在安全的前提下可以以一个较快的速度行驶，并且目标车道选择规划方式符合人的驾驶习惯。当自动驾驶车辆行驶前方道路由静态物体目标(如：路障、路锥、无法移动事故车辆等)或者隧道入口红灯等工况引起的车道异常无法通行，可以提前主动引导自动驾驶车辆逐渐向应急车道靠拢或者驶离高速公路，避免车辆发生碰撞危险。
在本发明的一个实施例中，所述控制模块1730用于：判断所述目标分支中的最右侧车道是否正常；如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块1730还用于：如果所述目标分支中的最右侧车道异常；则将临近所述最右侧车道的正常车道作为所述目标车道。
在本发明的一个实施例中，所述控制模块1730还用于在根据行驶规则从所述目标分支中选择目标车道之前，判断所述目标分支与所述车辆之间的距离，当所述距离小于预设距离时，根据行驶规则从所述目标分支中选择目标车道。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述。
进一步地，本发明的实施例公开了另一种自动驾驶车辆的车道选择系统，包括：第一获取模块，用于获取车辆行驶前方的道路信息；第一识别模块，用于根据所述道路信息判断所述车辆行驶前方的道路是否为主道分叉道路；第一控制模块，用于在所述车辆行驶前方的道路为主道分叉道路时，进一步根据导航信息从所述主道分叉道路选择目标分支，并根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道，以及控制所述车辆向所述目标车道变道。
在本发明的一个实施例中，所述第一控制模块用于：当所述目标分支内的主道数量为1条车道且第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为2条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；当所述目标分支内的主道数量为3条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第三车道是否正常，并在第三车道正常时，将所述第三车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道。
在本发明的一个实施例中，所述第一控制模块还用于：当所述目标分支为右分支且目标分支内的车道均异常时，则将左分支内的最为临近所述右分支的正常车道作为所述目标车道。
在本发明的一个实施例中，所述第一控制模块还用于：当所述目标分支为左分支且目标分支内的车道均异常时，则将右分支内的居中的正常车道作为所述目标车道；当所述目标分支为左分支且目标分支内的车道均异常且左分支内的车道数量为2时，则将右分支内的最为远离所述左分支的正常车道作为所述目标车道。
本发明实施例的自动驾驶车辆的车道选择方法，可以在遇到主道分叉时，进行合理的变道，从而提升行车安全和行车的可靠性。
需要说明的是，本发明实施例的自动驾驶车辆的车道选择系统的具体实现方式与本发明实施例的自动驾驶车辆的车道选择方法的具体实现方式类似，具体请参见方法部分的描述，为了减少冗余，此处不做赘述。
进一步地，本发明的实施例公开了一种车辆，设置有如上述任意一个实施例中的自动驾驶车辆的车道选择系统。该车辆可以在遇到道路分叉(如主道分叉和/或匝道分叉)时， 进行合理的变道，从而提升行车安全和行车的可靠性。
另外，根据本发明实施例的车辆的其它构成以及作用对于本领域的普通技术人员而言都是已知的，为了减少冗余，此处不做赘述。
需要说明的是，在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，"计算机可读介质"可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
尽管上面已经示出和描述了本申请的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本申请的限制，本领域的普通技术人员在本申请的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (14)

  1. 一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
    获取车辆行驶前方的道路信息；
    根据所述道路信息判断所述车辆行驶前方的道路是否为目标道路；
    如果是，则在所述车辆驶入所述目标道路之前，根据行驶规则从所述目标道路中选择目标车道；
    控制所述车辆向所述目标车道变道。
  2. 根据权利要求1所述的自动驾驶车辆的车道选择方法，其特征在于，所述目标道路包括匝道变宽道路，根据行驶规则从所述匝道变宽道路中选择目标车道，包括：
    判断所述匝道变宽道路中的最右侧车道是否正常；
    如果是，则将所述匝道变宽道路中的最右侧车道作为所述目标车道；
    所述目标道路为匝道变窄道路，根据行驶规则从所述匝道变窄道路中选择目标车道，包括：
    判断所述匝道变窄道路中的最右侧车道是否正常；
    如果是，则将所述匝道变窄道路中的最右侧车道作为所述目标车道；
    所述目标道路为匝道分叉道路，则进一步根据导航信息从所述匝道分叉道路选择目标分支，并在所述车辆驶入所述目标分支之前，根据行驶规则从所述目标分支中选择目标车道。
  3. 根据权利要求2所述的自动驾驶车辆的车道选择方法，其特征在于，所述根据行驶规则从所述匝道变宽道路中选择目标车道，还包括：
    如果所述匝道变宽道路中的最右侧车道异常；
    则将临近所述最右侧车道的正常车道作为所述目标车道；
    所述根据行驶规则从所述匝道变窄道路中选择目标车道，还包括：
    如果所述匝道变窄道路中的最右侧车道异常；
    则将临近所述最右侧车道的左侧正常车道作为所述目标车道；
    所述根据行驶规则从所述目标分支中选择目标车道，包括：
    判断所述目标分支中的最右侧车道是否正常；
    如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
  4. 根据权利要求2或3任一项所述的自动驾驶车辆的车道选择方法，其特征在于，在根据行驶规则从所述匝道变宽道路中选择目标车道之前，还包括：
    判断所述匝道变宽道路与所述车辆之间的距离；
    当所述距离小于预设距离时，根据行驶规则从所述匝道变宽道路中选择目标车道。
  5. 根据权利要求2-4任一项所述的自动驾驶车辆的车道选择方法，其特征在于，所述匝道变宽道路包括左侧匝道变宽道路、右侧匝道变宽道路和两侧匝道变宽道路。
  6. 根据权利要求2或3所述的自动驾驶车辆的车道选择方法，其特征在于，在根据行驶规则从所述匝道变窄道路中选择目标车道之前，还包括：
    判断所述匝道变窄道路与所述车辆之间的距离；
    当所述距离小于预设距离时，根据行驶规则从所述匝道变窄道路中选择目标车道。
  7. 根据权利要求2或3所述的自动驾驶车辆的车道选择方法，其特征在于，所述匝道变窄道路包括左侧匝道变窄道路、右侧匝道变窄道路和两侧匝道变窄道路；
    所述根据行驶规则从所述目标分支中选择目标车道，包括：
    判断所述目标分支中的最右侧车道是否正常；
    如果是，则将所述目标分支中的最右侧车道作为所述目标车道。
  8. 根据权利要求2或3所述的自动驾驶车辆的车道选择方法，其特征在于，所述根据行驶规则从所述目标分支中选择目标车道，还包括：
    如果所述目标分支中的最右侧车道异常；
    则将临近所述最右侧车道的正常车道作为所述目标车道。
  9. 根据权利要求2-3和7-8中任一项所述的自动驾驶车辆的车道选择方法，其特征在于，在根据行驶规则从所述目标分支中选择目标车道之前，还包括：
    判断所述目标分支与所述车辆之间的距离；
    当所述距离小于预设距离时，根据行驶规则从所述目标分支中选择目标车道。
  10. 一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
    获取车辆行驶前方的道路信息；
    根据所述道路信息判断所述车辆行驶前方的道路是否为主道变宽道路；
    如果是，则根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道；
    控制所述车辆向所述目标车道变道。
  11. 根据权利要求10所述的自动驾驶车辆的车道选择方法，其特征在于，当所述主道变宽道路变宽后的主道数量为2时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：
    在第二车道正常时将所述第二车道作为所述目标车道，在所述第二车道异常且第一车道正常时将所述第一车道作为所述目标车道；
    当所述主道变宽道路变宽后的主道数量为3时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：
    当第二车道正常时将所述第二车道作为所述目标车道；
    当所述第二车道异常时，进一步判断第三车道是否正常；
    如果正常且车辆的当前的行驶车道不为第一车道，则将所述第三车道作为所述目标车道，否则当第一车道正常且车辆的当前的行驶车道为第一车道时，则将所述第一车道作为所述目标车道；
    如果所述第二车道和所述第三车道均异常，则在第一车道正常时将所述第一车道作为所述目标车道；
    当所述主道变宽道路变宽后的主道数量为4时，根据当前的行驶车道、变宽后的主道数量以及每个主道的异常情况选择目标车道，包括：
    当第二车道正常、第三车道异常、第四车道正常且当前的行驶车道为第四车道时，将所述第四车道作为所述目标车道，否则，将第二车道作为所述目标车道；
    当第二车道异常、第三车道异常、第四车道异常且第一车道正常时，将所述第一车道作为所述目标车道；
    当第二车道异常、第三车道异常、第四车道正常、第一车道正常且当前的行驶车道为第二车道时，将所述第一车道作为所述目标车道，否则，将第四车道作为所述目标车道；
    当第二车道异常、第三车道正常、第一车道正常且当前的行驶车道为第一车道时，将所述第一车道作为所述目标车道，否则，将所述第三车道作为所述目标车道。
  12. 一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
    获取车辆行驶前方的道路信息；
    根据所述道路信息判断所述车辆行驶前方的道路是否为主道变窄道路；
    如果是，则进一步判断当前的行驶车道是否为变窄车道；
    如果是，则将最为临近所述当前的行驶车道的正常主道作为目标车道；
    控制所述车辆向所述目标车道变道。
  13. 一种自动驾驶车辆的车道选择方法，其特征在于，包括以下步骤：
    获取车辆行驶前方的道路信息；
    根据所述道路信息判断所述车辆行驶前方的道路是否为主道分叉道路；
    如果是，则进一步根据导航信息从所述主道分叉道路选择目标分支；
    根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道；
    控制所述车辆向所述目标车道变道。
  14. 根据权利要求13所述的自动驾驶车辆的车道选择方法，其特征在于，所述根据所述目标分支内的主道数量以及每个主道的异常情况选择目标车道，包括：
    当所述目标分支内的主道数量为1条车道且第一车道正常时，将所述第一车道作为所述目标车道；
    当所述目标分支内的主道数量为2条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；
    当所述目标分支内的主道数量为3条车道且第二车道正常时，将所述第二车道作为所述目标车道，否则进一步判断第三车道是否正常，并在第三车道正常时，将所述第三车道作为所述目标车道，否则进一步判断第一车道是否正常，并在第一车道正常时，将所述第一车道作为所述目标车道；
    当所述目标分支为右分支且目标分支内的车道均异常时，则将左分支内的最为临近所述右分支的正常车道作为所述目标车道；
    当所述目标分支为左分支且目标分支内的车道均异常时，则将右分支内居中的正常车道作为所述目标车道。
PCT/CN2019/130061 2018-12-29 2019-12-30 自动驾驶车辆的车道选择方法、系统及车辆 WO2020135881A1 (zh)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
CN201811635557.7 2018-12-29
CN201811636519.3A CN110619758B (zh) 2018-12-29 2018-12-29 自动驾驶车辆的车道选择方法、系统及车辆
CN201811635557.7A CN110619757A (zh) 2018-12-29 2018-12-29 自动驾驶车辆的车道选择方法、系统及车辆
CN201811643942.6A CN110675654A (zh) 2018-12-29 2018-12-29 自动驾驶车辆的车道选择方法、系统及车辆
CN201811636519.3 2018-12-29
CN201811643942.6 2018-12-29

Publications (1)

Publication Number Publication Date
WO2020135881A1 true WO2020135881A1 (zh) 2020-07-02

Family

ID=71129689

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/130061 WO2020135881A1 (zh) 2018-12-29 2019-12-30 自动驾驶车辆的车道选择方法、系统及车辆

Country Status (1)

Country Link
WO (1) WO2020135881A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112839855A (zh) * 2020-12-31 2021-05-25 华为技术有限公司 一种轨迹预测方法与装置

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009047491A (ja) * 2007-08-16 2009-03-05 Hitachi Ltd ナビゲーション装置
CN102640198A (zh) * 2009-12-14 2012-08-15 爱信艾达株式会社 车辆引导装置、车辆引导方法和车辆引导程序
CN106461406A (zh) * 2014-06-10 2017-02-22 歌乐株式会社 车道选择装置、车辆控制系统及车道选择方法
CN107923760A (zh) * 2015-07-27 2018-04-17 日产自动车株式会社 路径引导装置以及路径引导方法
CN108177653A (zh) * 2016-12-08 2018-06-19 本田技研工业株式会社 车辆控制装置
CN109074740A (zh) * 2016-03-16 2018-12-21 本田技研工业株式会社 车辆控制系统、车辆控制方法及车辆控制程序
CN110619757A (zh) * 2018-12-29 2019-12-27 长城汽车股份有限公司 自动驾驶车辆的车道选择方法、系统及车辆
CN110619758A (zh) * 2018-12-29 2019-12-27 长城汽车股份有限公司 自动驾驶车辆的车道选择方法、系统及车辆

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009047491A (ja) * 2007-08-16 2009-03-05 Hitachi Ltd ナビゲーション装置
CN102640198A (zh) * 2009-12-14 2012-08-15 爱信艾达株式会社 车辆引导装置、车辆引导方法和车辆引导程序
CN106461406A (zh) * 2014-06-10 2017-02-22 歌乐株式会社 车道选择装置、车辆控制系统及车道选择方法
CN107923760A (zh) * 2015-07-27 2018-04-17 日产自动车株式会社 路径引导装置以及路径引导方法
CN109074740A (zh) * 2016-03-16 2018-12-21 本田技研工业株式会社 车辆控制系统、车辆控制方法及车辆控制程序
CN108177653A (zh) * 2016-12-08 2018-06-19 本田技研工业株式会社 车辆控制装置
CN110619757A (zh) * 2018-12-29 2019-12-27 长城汽车股份有限公司 自动驾驶车辆的车道选择方法、系统及车辆
CN110619758A (zh) * 2018-12-29 2019-12-27 长城汽车股份有限公司 自动驾驶车辆的车道选择方法、系统及车辆

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112839855A (zh) * 2020-12-31 2021-05-25 华为技术有限公司 一种轨迹预测方法与装置

Similar Documents

Publication Publication Date Title
WO2022105579A1 (zh) 一种基于自动驾驶的控制方法、装置、车辆以及相关设备
WO2020135742A1 (zh) 自动驾驶车辆的横向决策系统及横向决策确定方法
US10126751B2 (en) Lane change support device
EP3333820B1 (en) Driving plan device, travel support device, and driving plan method
EP3950453B1 (en) Driving assistance method and driving assistance device
CN110619758B (zh) 自动驾驶车辆的车道选择方法、系统及车辆
KR101417522B1 (ko) 고속도로 자율주행 시스템 및 방법
WO2017126250A1 (ja) 運転支援方法及び装置
EP3327693A1 (en) Scene evaluation device, travel support device, and scene evaluation method
JP6575612B2 (ja) 運転支援方法及び装置
CN106428007A (zh) 车辆的自主行驶控制装置和自主行驶控制方法
KR20190133623A (ko) 적어도 하나의 자동차의 가이드를 지원하기 위한 방법, 보조 시스템 및 자동차
CA2992967A1 (en) Scene determination device, travel assistance apparatus, and scene determination method
CN110614992A (zh) 车辆自动驾驶时避障的方法、系统及车辆
CN111399512A (zh) 行车控制方法、行车控制装置和车辆
KR20190138312A (ko) 운전 지원 장치 및 운전 지원 방법
CN110619757A (zh) 自动驾驶车辆的车道选择方法、系统及车辆
RU2756878C1 (ru) Способ помощи при движении и устройство помощи при движении
CN110673595B (zh) 车辆自动驾驶时避障的方法、系统及车辆
US11731616B2 (en) Autonomous driving system for preventing collision of cut-in vehicle and autonomous driving method thereof
WO2020135881A1 (zh) 自动驾驶车辆的车道选择方法、系统及车辆
KR102624829B1 (ko) 차량의 위치 정보를 이용한 주행 안내 제공 방법, 장치 및 컴퓨터프로그램
JP7304875B2 (ja) 自動運転制御方法及び自動運転制御システム
CN114475649A (zh) 自动驾驶控制装置和自动驾驶控制方法
KR20210068223A (ko) 자율주행 주행경로에 필요한 신호등 체계 분석 시스템

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: 19902701

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19902701

Country of ref document: EP

Kind code of ref document: A1