CN114898564A - Intersection multi-vehicle cooperative passing method and system under unstructured scene - Google Patents
Intersection multi-vehicle cooperative passing method and system under unstructured scene Download PDFInfo
- Publication number
- CN114898564A CN114898564A CN202210812912.3A CN202210812912A CN114898564A CN 114898564 A CN114898564 A CN 114898564A CN 202210812912 A CN202210812912 A CN 202210812912A CN 114898564 A CN114898564 A CN 114898564A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- planning
- track
- area
- cooperative
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000013459 approach Methods 0.000 claims abstract description 18
- 230000001133 acceleration Effects 0.000 claims description 21
- 238000013519 translation Methods 0.000 claims description 6
- 239000011541 reaction mixture Substances 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 241000725585 Chicken anemia virus Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- QVRVXSZKCXFBTE-UHFFFAOYSA-N n-[4-(6,7-dimethoxy-3,4-dihydro-1h-isoquinolin-2-yl)butyl]-2-(2-fluoroethoxy)-5-methylbenzamide Chemical compound C1C=2C=C(OC)C(OC)=CC=2CCN1CCCCNC(=O)C1=CC(C)=CC=C1OCCF QVRVXSZKCXFBTE-UHFFFAOYSA-N 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000011064 split stream procedure Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
-
- 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
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096805—Systems involving transmission of navigation instructions to the vehicle where the transmitted instructions are used to compute a route
-
- 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
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/096833—Systems involving transmission of navigation instructions to the vehicle where different aspects are considered when computing the route
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Mathematical Physics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for intersection multi-vehicle cooperative passing under an unstructured scene, which comprise the following steps: step 1, setting a cooperation area, an approach area and a shunt area from inside to outside from the center of an intersection in an unstructured scene; step 2, setting a virtual road center line in the approach area, taking an intersection point of the virtual road center line and the outer boundary line of the coordination area as a steering point, and selecting a shunting target point on the virtual road center line at intervals from the steering point; step 3, determining the vehicle priority in each area and among different areas; step 4, according to the vehicle priority, sequentially planning a shunting track of the vehicles in the shunting area, so that the vehicles run to the virtual road center line before entering the coordination area; and 5, taking the vehicle entering the cooperative area as a waiting cooperative track planning vehicle, and performing cooperative track planning according to the vehicle priority.
Description
Technical Field
The invention relates to the technical field of intersection multi-vehicle cooperative passing control, in particular to an intersection multi-vehicle cooperative passing method and system under an unstructured scene.
Background
A signalless intersection is a complex scenario with a variety of potential collision risks. When a plurality of intelligent vehicles pass through an intersection at the same time, how to effectively improve the global traffic efficiency on the premise of ensuring safety is a very challenging problem in the field of multi-intelligent-vehicle collaborative planning. In an urban structured road scene, by designing an AIM scheme (all called as an 'Autonomous interaction Management scheme' in English and all called as an 'Intersection intelligent Management scheme' in Chinese), overall planning of a path of CAVs (all called as 'Connected and Autonomous Vehicles' in English and all called as 'intelligent Internet Vehicles' in Chinese) near an Intersection is an effective means for improving the multi-vehicle traffic safety and traffic efficiency of the Intersection. Currently, research for unstructured intersection scenes is rare. In scenes such as intelligent open-pit mines and the like, a plurality of unstructured intersections with wide roads exist. Currently, there is still a lack of a system solution for solving the problem of multi-vehicle co-planning for unstructured intersections.
The main results of the current research on AIM scheme are:
1) rule-based schemes: based on the 'predetermined' principle or the rule of first-come first-go, only one vehicle passes through the intersection each time, that is, the method only allows one vehicle to run in the intersection area at the same time, so that the safety can be guaranteed, but the traffic efficiency of the intersection is limited.
2) Based on the optimized scheme: and establishing a cost function and collision constraint, and solving an optimal control problem to obtain an optimal plan. The method regards the intersection as a continuous free space, expands the passing area of each vehicle, and obtains the global optimal solution considering all vehicle passing efficiency by a numerical calculation method, so as to improve the overall passing efficiency to the maximum extent. However, all vehicles are completely free of constraints in the aspects of routes and priority sequences, the vehicles easily overtake, change to opposite lanes and other 'light-floating' maneuvers, the passing is too disordered, only a numerical method can be adopted for solving, and the solving time is too long.
Disclosure of Invention
The invention aims to provide a method and a system for intersection multi-vehicle cooperative passing under an unstructured scene, which overcome or at least alleviate at least one of the above defects of the prior art.
In order to achieve the purpose, the invention provides a method for intersection multi-vehicle cooperative passage under an unstructured scene, which comprises the following steps:
step 1, setting a collaborative area from inside to outside from the center of the intersection in the unstructured scene
Near area of the body
And a flow splitting region
；
step 3, determining the vehicle priority in each area and among different areas;
step 4, according to the vehicle priority, sequentially aiming at the shunting areas
Inner vehicle
Planning the shunting track to ensure that the vehicle
Upon entering the collaborative area
Forward driving to the virtual road center line;
step 5, entering the cooperative area
Inner vehicle
And as a waiting collaborative trajectory planning vehicle, carrying out collaborative trajectory planning according to the vehicle priority.
Further, the step 5 specifically includes:
step 51, planning future maneuvering of the vehicle according to the cooperative track, determining an intersection through which the vehicle leaves the intersection, and obtaining an optimal collision-free track by taking all the turning points of the intersection as selectable cooperative track planning end points;
step 52, planning the future maneuver of the vehicle according to the cooperative track, and classifying the approaching area
Vehicles meeting preset requirements internally are constructed as collision sets
Judging the set of collisions
In a vehicle
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as the cooperative track; otherwise, go to step 53;
step 53, judging the maximum value of the vehicle speed of the coordinated track planning vehicle
If so, abandoning the optimal track, otherwise, determining the cooperative track according to the following rule if no collision occurs when the optimal track is defined as:
in principle one, if at least two optimal tracks do not collide, the optimal track with shorter time consumption is preferentially selected as the cooperative track;
in principle two, when all the optimal trajectories collide, the optimal trajectory with the smallest number of the spatial coincident points is selected as a collaborative trajectory, and the speed planning for avoiding collision is performed again.
Further, the preset requirements in step 52 include three requirements that need to be satisfied simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
Is located at the cooperative trackPlanning a left side of the vehicle; wherein,
representing the collaborative area
The collection of vehicles in the interior of the vehicle,
representing the proximity zone
A set of vehicles located internally on the virtual road centerline;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists between the cooperative track planning vehicle and the cooperative track planning vehicle;
a third requirement when the collaborative trajectory planning vehicle reaches the point of starting the turn, and the vehicle is
Has not exited the collaborative area
。
Further, the method of "speed planning" in step 53 specifically includes:
step 531, establishing an optimal control model (1):
in the formula (1), the reaction mixture is,
in order to be a function of the cost,
、
the weight of the weight is calculated,
is as follows
The length of each of the time steps,
is as follows
The length of each of the time steps,
、
respectively planning the acceleration and the speed of the vehicle for the vehicle collaborative track to be planned after the time is discretized;
equation (2) is the kinematic constraint of the vehicle,
、
respectively planning position coordinates, orientation angles and acceleration of the vehicle for the collaborative trajectory under a Cartesian coordinate system;
equation (3) is the vehicle side value constraint,
、
、
、
planning just-entering cooperative areas of vehicles for cooperative tracks under geodetic coordinate system respectively
Initial position of time, exit from cooperation area
Final position of time, entry into coordination area
The position coordinates and orientation angles of the start point and the end point of (1);
equation (4) is the maximum limit on vehicle speed and acceleration,
is the maximum value of the vehicle acceleration;
equation (5) is the collision restraint of the vehicle,
、
respectively under the geodetic coordinate system
Coordinate and collaborative track planning vehicle of space coincident point
Vehicle concentrating on collision
To arrive at
Time of a spatial coincidence point
The coordinates of the position are determined by the position,
is the total number of points of spatial coincidence,
、
respectively the lateral and longitudinal safety distances between the vehicles.
Further, in the step 2, "in the proximity region
The method for setting the virtual road center line specifically includes:
step 21, approaching the area
The road boundaries on the two sides in the road surface move horizontally for a preset distance in the opposite direction, and then smoothing is carried out to obtain two virtual road center lines;
and step 22, obtaining other virtual road center lines by adopting the same translation method in the step 31 on the basis of the obtained two virtual road center lines.
Further, the vehicle priority inside each zone determined in step 3 is: vehicles in the same area, the priority of each vehicle being inversely proportional to the distance of the vehicle from the center;
the vehicle priorities between the different areas determined in the step 3 are as follows: the cooperation area
The vehicles with the lowest internal priority have higher priority in different areas than the approaching area
The vehicle with the highest inner priority, the approach area
The priority of the vehicle with the lowest internal priority in different areas is higher than that of the shunting area
The vehicle with the highest internal priority.
The invention also provides a system for intersection multi-vehicle cooperative passing under the unstructured scene, which comprises:
a partitioning unit for setting a collaborative area from inside to outside starting from the center of the intersection in the unstructured scene
Near area of the body
And a flow splitting region
；
A priority determining unit for determining a priority of the vehicle inside each zone and between different zones;
a virtual road center line setting unit for setting a virtual road center line in the access area
Setting a virtual road center line, and taking the virtual road center line and the cooperation area
Taking the intersection point of the outer boundary line as a turning point, and selecting shunting target points at intervals on the virtual road center line from the turning point;
a shunting planning unit for sequentially aiming at the shunting areas according to the vehicle priority
Inner vehicle
Planning the shunting track to ensure that the vehicle
Upon entering the collaborative area
Driving to the virtual road central line;
a collaborative planning unit for entering the collaborative area
Inner vehicle
And as a waiting collaborative track planning vehicle, carrying out collaborative track planning according to the vehicle priority.
Further, the collaborative planning unit specifically includes:
the optimal track obtaining subunit is used for planning future maneuvering of the vehicle by the cooperative track, determining the intersection through which the vehicle leaves the intersection, and obtaining the optimal track without collision by taking all the turning points of the passed intersection as selectable cooperative track planning end points;
a collaborative trajectory planning subunit for future maneuvers of the collaborative trajectory planning vehicle, the approach area
Vehicles meeting preset requirements internally are constructed as collision sets
Judging the set of collisions
In a vehicle
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as the cooperative track; otherwise, judging the maximum value of the vehicle speed of the coordinated track planning vehicle
If so, abandoning the optimal track, otherwise, determining the cooperative track according to the following rule if no collision occurs when the optimal track is defined as:
in principle one, if at least two optimal tracks do not collide, the optimal track with shorter time consumption is preferentially selected as the cooperative track;
in principle two, when all the optimal tracks collide, the optimal track with the minimum number of the spatial coincident points is selected as a cooperative track, and the speed planning for avoiding collision is performed again;
wherein, the preset requirements comprise three requirements which need to be met simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
On the left side of the collaborative trajectory planning vehicle; wherein,
representing the collaborative area
Inside vehicleThe collection of vehicles is carried out by a plurality of vehicles,
representing the proximity zone
A set of vehicles located internally on the virtual road centerline;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists between the cooperative track planning vehicle and the cooperative track planning vehicle;
a third requirement when the collaborative trajectory planning vehicle reaches the point of starting the turn, and the vehicle is
Has not exited the collaborative area
。
Further, the collaborative trajectory planning subunit performs speed planning through an optimal control model (1):
in the formula (1), the reaction mixture is,
in order to be a function of the cost,
、
the weight of the weight is calculated,
is as follows
The length of each of the time steps,
is as follows
The length of each time step is equal to the length of each time step,
、
respectively planning the acceleration and the speed of the vehicle for the vehicle collaborative track to be planned after the time is discretized;
equation (2) is the kinematic constraint of the vehicle,
、
respectively planning position coordinates, orientation angles and acceleration of the vehicle for the collaborative trajectory under a Cartesian coordinate system;
equation (3) is the vehicle side value constraint,
、
、
、
respectively cooperation under the geodetic coordinate systemJust-in-time coordination area of trajectory planning vehicle
Initial position of time, exit from cooperation area
Final position of time, entry into coordination area
The position coordinates and orientation angles of the start point and the end point of (1);
equation (4) is the maximum limit on vehicle speed and acceleration,
is the maximum value of the vehicle acceleration;
equation (5) is the crash restraint of the vehicle,
、
respectively under the geodetic coordinate system
Vehicle in collision concentration by planning vehicle coordinates and coordinated tracks of space coincident points
To arrive at
Time of a spatial coincidence point
The coordinates of the position are determined by the position,
is the total number of points of spatial coincidence,
、
respectively the lateral and longitudinal safety distances between the vehicles.
Further, the method of the virtual road center line setting unit specifically includes:
first, approach the area
The road boundaries on the two sides in the inner part are translated towards the opposite direction for a preset distance, then the smoothing treatment is carried out to obtain two virtual road center lines, and the same translation method is adopted on the basis of the obtained two virtual road center lines to obtain other virtual road center lines.
Due to the adoption of the technical scheme, the invention has the following advantages:
1. aiming at a wide unstructured intersection without a signal lamp in a mining area, the invention utilizes a road boundary to generate a virtual road center line, carries out priority regulation on vehicles near the intersection, sets a target lane for the intersection when the vehicles approach the intersection, carries out pre-diversion treatment, designs a set path for different motor vehicles through a mixed A-star algorithm, finally sets performance indexes such as the distance between the vehicles, the time for passing the intersection, the comfort of acceleration and deceleration and the like as cost functions comprehensively, and carries out speed planning on the vehicles with low priority so as to achieve the purposes of avoiding collision and improving the overall traffic efficiency of the intersection.
2. The invention carries out structuralization processing on the unstructured intersection, improves the intersection passing efficiency and simultaneously fully ensures the safety and the orderly operation of the mining area work.
3. The partial optimization-partial rule method has smaller calculation amount than a pure optimization algorithm and higher passing efficiency than a pure rule algorithm.
Drawings
Fig. 1 is a schematic structural diagram of an AIM system according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of intersection multi-vehicle cooperative passing under an unstructured scene provided by an embodiment of the present invention.
Fig. 3 is a schematic diagram of the relationship of vehicle priorities according to the embodiment of the present invention.
Fig. 4 is a schematic diagram of a collision relation analysis provided in the embodiment of the present invention.
Detailed Description
In the drawings, the same or similar reference numerals are used to denote the same or similar elements or elements having the same or similar functions. Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The following terms are referred to herein, and their meanings are explained below for ease of understanding. It will be understood by those skilled in the art that the following terms may have other names, but any other names should be considered consistent with the terms set forth herein without departing from their meaning.
Arranging the AIM system: the AIM system and its communication mode near an intersection. A roadside unit is disposed at the intersection, which is capable of V2I communication with smart vehicles (chinese is collectively referred To as "communication between smart vehicles and road facilities", and english of V2I is collectively referred To as "Vehicle-To-Infrastructure"). Of course, V2V communication can be performed between intelligent vehicles (Chinese is called "communication between vehicles", and English in V2V is called "Vehicle-To-Vehicle"). And when the intelligent vehicle enters the vicinity of the intersection, the intelligent vehicle sends the coordinates of the intelligent vehicle and an application for participating in collaborative planning to the road side unit. The road side unit gives priority order to the intelligent vehicle according to a certain rule, and plans the track for the intelligent vehicle according to the order. For each smart vehicle, the roadside unit will make two plans. The first planning is shunting track planning, so that the automobiles with scattered positions originally can run on a desired path before entering the intersection, the passing is more orderly, and the calculation burden of subsequent collaborative planning can be reduced. The second planning is coordinated track planning, and specific tracks passing through the intersection are planned for the intelligent vehicle.
As shown in fig. 1, the intersection multi-vehicle cooperative passing method in the unstructured scene provided by the embodiment of the present invention includes:
step 1, setting a cooperative area from inside to outside from the center of the intersection in the unstructured scene
Near area of the body
And a flow splitting region
。
Where the center of the intersection can be, but is not limited to, choosing the geometric center of the intersection, "inside" versus "outside", closer to the center.
In one embodiment, as shown in FIG. 2, the collaborative area
Near area of the body
And a flow splitting region
The shape of the intersection can be set as a circle shown in the figure, a first preset radius, a second preset radius and a third preset radius are sequentially selected from small to large according to the size and the shape of the intersection, three circles are generated, and a cooperation area is set
Near area of the body
And a flow splitting region
Wherein:
collaborative area
The inner circular area surrounded by the outer boundary line corresponding to the first preset radius can be understood as an area included in a circle which just intersects with four right-angle-like boundaries of the intersection in the embodiment.
Proximity zone
The first ring area is formed outside the outer boundary line corresponding to the first preset radius and inside the outer boundary line corresponding to the second preset radius. The radius of the first circular area, i.e. the difference between the first preset radius and the second preset radius, is mainly determined by the size and shape of the intersection, and can be usually selected to be about 100 m.
Flow splitting region
And the second ring area is formed outside the outer boundary line corresponding to the second preset radius and inside the outer boundary line corresponding to the third preset radius. The radius of the second circular area, i.e. the difference between the third preset radius and the second preset radius, is mainly determined by the size and shape of the intersection, and usually more than 50m can be selected.
Of course, square or other geometric shaped regions may be selected as desired and are not further enumerated herein.
Thus, the preprocessing of the map of the intersection and the surrounding area is completed. The virtual road center line, the steering point and the shunting target point are all loaded into a high-precision map, so that the intelligent vehicle can be conveniently utilized.
In one embodiment, the method of "setting a virtual center line of a road in an access area" may include:
as shown in fig. 2, in the present embodiment, a scene with a road width of about four lanes is shown in fig. 2. Respectively facing the two road boundaries in the approaching area, namely translating towards the center line of the road by a preset distance, wherein the preset distance can be set to be, but is not limited to be
Wherein
in order to be a safe distance away from the vehicle,
is the mine car width. Preferably, the road boundaries on both sides are translated towards each other and then smoothed to form a first virtual road center line, as shown by the dotted line in fig. 2
As shown.
Further, two first virtual road center lines are formed
On the basis, the similar translation method is adopted, and the opposite translation is continued to obtain the other two second virtual road center lines
. The same processing is performed on the four roads meeting at the intersection, so that the virtual road center line shown in fig. 2 can be obtained, and a greater number of virtual road center lines may exist in a wider road environment. It should be noted that the number of virtual road center lines can be determined according to the relationship between the width of the unstructured road and the width of the intelligent vehicleAnd under the condition that the width of the unstructured road allows, creating virtual road center lines as much as possible to increase the mobility during passing and fully utilize the road circulation capacity.
Describing a set formed by a plurality of virtual road center lines as a set
，
Corresponding to an index of the virtual road center line,
is shown as
The strip of virtual road center line is,
the number of the virtual road center lines.
Will be provided with
The set formed by the plurality of shunting target points is described as a set
，
Correspond to as
Virtual road center line
The index of the upper diversion target point,
is shown as
Virtual road center line
To
The number of the shunt target points is equal to that of the shunt target points,
is as follows
Virtual road center line
The number of the upper shunting target points.
The virtual road center line can be set in the proximity area by other methods, which are not described in detail herein.
In one embodiment, "take the virtual road center line and the cooperation area
The method of using the intersection of the outer boundary lines as the turning point "specifically includes:
get the virtual road center line and the cooperation area
Of the outer boundary line, as in fig. 2
Dot sum
Shown by the dots, turning points
Is a first virtual road center line
And a cooperation area
Intersection of outer boundary lines, turning points
Is a second virtual road center line
And a cooperation area
The intersection of the outer boundary lines.
In one embodiment, the turning point serves as a coordinated trajectory planning starting point or ending point during turning. Such as: vehicle access coordination area
Center line and cooperation area of virtual lane driven by front vehicle
The intersection point of the outer boundary lines is the collaborative trajectory planning starting point of the vehicle. The cooperative track planning end point is determined according to the future maneuver (left turn, right turn or straight movement) to be executed by the vehicle, and the intersection passed by the vehicle leaves the intersection and the approach area of the passed intersection
And a plurality of driving virtual lane central lines exist, so that a plurality of steering points can be selected as end points, and the final collaborative trajectory planning end point is determined in the collaborative planning stage. The details of this section will be described later.
In one embodiment, the method for selecting the diversion target points at intervals on the virtual road center line may include:
the shunting target point is shown as a black square block in fig. 2, and provides a collaborative trajectory planning end point for the path search method. It should be noted that the distance between the spaced adjacent diversion target points is not less than the sum of the length of the mine card and the longitudinal safety distance.
And 3, determining the vehicle priority inside each region and between different regions.
For example, as shown in fig. 3, an axis representing the priority of vehicles is established with the center of the intersection as a starting point, and the vehicles are projected onto the axis, and the priority of the vehicles can be determined by the distance between each vehicle and the center.
Collaborative area
The set of vehicles within represents a first set of vehicles
Wherein
representing collaborative areas
Inner first
Vehicle, hereinafter simply referred to as "vehicle
”。
Numerical value of and vehicle
The corresponding priorities are inversely proportional, that is,
the larger the value of (A), the vehicle
The lower the priority of (c).
Proximity zone
The set of vehicles lying on the virtual road centerline represents the second set of vehicles
Wherein
representing an area of approach
Inner first
Vehicle, hereinafter simply referred to as "vehicle
”。
Numerical value of and vehicle
The corresponding priorities are inversely proportional, that is,
the larger the value of (A), the vehicle
The lower the priority of (c).
Proximity zone
Vehicles not located on the virtual road center line and diversion area
The set of vehicles in (1) is represented as a third set of vehicles
Wherein, in the process,
representing an area of approach
Inner first
Vehicle, hereinafter simply referred to as "vehicle
”。
Numerical value of and vehicle
The corresponding priorities are inversely proportional, that is,
the larger the value of (A), the vehicle
The lower the priority of (c).
In another embodiment, the vehicle priority inside each zone is determined as: vehicles in the same area, each vehicle having a priority inversely proportional to the distance of the vehicle from the center.
Vehicle priorities between the different zones are determined as: the cooperation area
Prioritizing different zones for vehicles with lowest internal priorityLevel above the proximity zone
The vehicle with the highest inner priority, the approach area
The priority of the vehicle with the lowest internal priority in different areas is higher than that of the shunting area
The vehicle with the highest internal priority.
Step 4, according to the vehicle priority, sequentially aiming at the shunting areas
Inner vehicle
Planning the shunting track to ensure that the vehicle
Upon entering the collaborative area
And driving to the virtual road center line.
In one embodiment, the flow splitting region
The driving position of the vehicle in (1) is randomly scattered due to the non-structural characteristics of the mining area environment and the possible local obstacle-avoiding behavior when driving on the previous road. The positions of the vehicles are more orderly, the calculation load of the central collaborative planner is reduced, and the vehicles enter the approach area
It is previously necessary to start gradually driving to the virtual lane desired by the present invention, i.e., the step2, lane symbolized by "virtual road center line". The road side planning unit will assemble the third vehicle according to the vehicle priority in the above embodiment
In a vehicle
The following planning is performed in sequence:
step 41, according to the vehicle
Current position, from the set of virtual road center lines
Selecting target lane line in the middle or near
Thereby obtaining the corresponding shunting target point set
。
Step 42, establishing an upper speed limit for the unstructured scene
And lower speed limit
Selecting a set
Middle near cooperative area
The shunting target point is a collaborative trajectory planning terminal point so as to
As a collaborative trajectory planning endpointAnd (4) planning the path by taking the self coordinate as a starting point at the boundary speed of the moment. The present embodiment employs a hybrid a-x algorithm, which generates the black, thin line traces in fig. 2. Of course, the path planning can also be performed by using the existing algorithm based on graph search, such as dijkstra, fast search random tree, etc.
Step 43, collision detection: selecting a set when detecting the presence of a temporal-spatial coincidence with a preceding vehicle diversion trajectory
Planning is carried out again until a collision-free path is generated so that the vehicle runs to the center line of the virtual lane, and planning is successful.
In step 44, since it is not desirable to take the passing action near the intersection, the terminal point at each successful planning and the diversion target point before the terminal point can no longer be used as the terminal point of the next vehicle. Then, set from the diversion target points
The shunting target points selected by the vehicles which are successfully planned are removed, and after the vehicles which are successfully planned reach the selected target shunting points and travel a safe distance forwards, the shunting target points are opened to the following vehicles, namely the vehicles are added to the set again
And (5) traversing and selecting the vehicle for subsequent vehicle planning.
Step 45, for the traversal set
Vehicles that cannot find a suitable diversion target point will be at speed in the same direction
And (5) running at a low speed, and waiting for the opening of a new shunting target point.
The planning method provided by the step 4 can enable a disordered intelligent vehicle to run according to a certain rule, reduce the calculation load for the subsequent collaborative area planning, and enable the vehicle to run orderly.
Step 5, the vehicle reaching the diversion target point is the second vehicle set in step 2
In a vehicle
These vehicles, also referred to elsewhere hereinafter as "collaborative trajectory planning vehicles", will continue to travel along the virtual road centerline as it enters the collaborative area
The starting point of (1) is a steering point of a central line of a virtual road where the starting point of (1) is located, and a cooperation area is carried out according to the priority of the vehicle
Collaborative trajectory planning within.
In order to improve the flexibility, when the coordinated trajectory is planned for three future maneuvers of left turn, right turn and straight going, a plurality of end points can be selected, and the number of the selectable end points is equal to the number of virtual road center lines on the corresponding lane (as shown in the example of fig. 2). For the collaborative trajectory planning vehicle, after the diversion of the step 4, the speed reaches the maximum passing speed of the intersection
. Collision bumper
In a vehicle
Are higher than the collaborative trajectory planning vehicle, so the vehicles
Are planned, i.e. the vehicle
The position at any point in time is known.
The step 5 specifically comprises the following steps:
step 51, planning future maneuvering of the vehicle according to the cooperative track, determining an intersection through which the vehicle leaves the intersection, and obtaining an optimal collision-free track by taking all the turning points of the passing intersection as selectable cooperative track planning end points;
step 52, planning the future maneuver of the vehicle according to the coordinated track, and connecting the approaching area
Vehicles meeting preset requirements internally are constructed as collision sets
Judging the set of collisions
In a vehicle
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as the cooperative track; otherwise, go to step 53;
step 53, judging the maximum value of the vehicle speed of the coordinated track planning vehicle
If so, abandoning the optimal track, otherwise, determining the cooperative track according to the following rule if no collision occurs when the optimal track is defined as:
in principle one, if there are at least two optimal trajectories that do not collide, the optimal trajectory with shorter time consumption is preferentially selected as the cooperative trajectory.
In principle two, when all the optimal trajectories collide, the optimal trajectory with the smallest number of the spatial coincident points is selected as a collaborative trajectory, and the speed planning for avoiding collision is performed again.
Taking the left-turn vehicle labeled (r) in FIG. 2 as an example, the vehicle (r) is about to enter the collaborative area
And planning the vehicle for the collaborative track. The method for collaborative trajectory planning provided for the collaborative trajectory planning vehicle in the embodiment specifically includes:
and step 51, planning an optimal track without collision according to the selectable number of the end points. According to the first end point, as shown in FIG. 2αAnd a second end pointβAt the maximum speed of the vehicle
Keeping constant speed, planning a first optimal track and a second optimal track of the vehicle under the condition of no collision, wherein the time for completing the first optimal track and the time for completing the second optimal track are respectively
。
And step 52, performing collision detection on the points where collision is possible, namely the coincident points of the first optimal track and the second optimal track and the vehicle track in the collision set of the first optimal track and the second optimal track on the space. And preferentially selecting the track which does not collide as a final track, and preferentially selecting the track which consumes less time when a plurality of tracks do not collide.
For example, in FIG. 2, let's say vehicle (r) is
Of the collision set
Two vehicles in the middle, namely and three, are respectively marked as
And
. The time is not considered for the moment,
、
and generating space coincidence points I, II and III respectively with the first optimal track and the second optimal track. I is on the first optimal track, and II and III are on the second optimal track. Due to the fact that at present
、 
And
the time to pass the point of spatial coincidence is known and therefore collision detection is easy. Through detection, if only one of the first optimal track and the second optimal track is not collided, the optimal track which is not collided is taken as the optimal track
The collaborative trajectory of (2); if the first optimal track and the second optimal track are not collided, selecting corresponding completion time
The track corresponding to the smaller one of the two is taken as
The cooperative trajectory of (a).
And step 53, when the two tracks are collided, selecting one track with less spatial coincident points to perform speed planning again so as to avoid collision.
In another embodiment, step 5 further comprises:
step 54, if the cooperation area
And (3) the vehicle which has the planned track and is running according to the planned track breaks down, and sends the positioning information of the vehicle to the road side unit and the vehicles in a certain range around.
Specifically, the vehicle with the planned track regards the fault vehicle as a temporary obstacle, starts a procedure of avoiding the temporary obstacle, and plans the obstacle avoidance track by using a built-in online planner. The road side unit will add the faulty vehicle as a static obstacle to the map. The roadside unit also considers the newly added static obstacles when planning the vehicles with the paths which are not planned.
For all second vehicles
The vehicles in the system repeat the process according to the priority sequence to obtain the coordinated track planning of all intelligent vehicles which are ready to pass through the intersection at the current moment.
Existing collaborative trajectory approaches can be divided into three categories: 1. based on a reservation subscription mechanism, only one vehicle passes through each time period of the intersection; 2. based on driving rules or traffic lights on the structured road; 3. and (4) planning the optimal track of the global vehicle based on solving the optimal control problem. The former two can only be applied to the structured road scene, and the last one has long calculation time and no disorder in driving. The collaborative trajectory planning method provided by the embodiment retains greater maneuverability to adapt to an unstructured scene to improve the overall peer efficiency, and reduces the calculation amount based on certain regularity so that the passing is not too disordered.
In one embodiment, the road side unit performs information interaction with the vehicles, and can know the future maneuver of each vehicle so as to judge whether a potential collision relationship exists, and the following collision situations exist in the vehicle collision at the intersection:
the first collision scenario, a cross-collision, is a collision that occurs between two vehicles traveling on a path with an intersection, as illustrated by a in FIG. 4.
The second collision scenario, the merge-in collision, is a collision that occurs between vehicles merging into the same lane, as shown in b in fig. 4.
In a third collision scenario, a rear-end collision, as in c of fig. 4, a vehicle traveling on the same lane encounters a collision between a front and rear vehicle before separation or on the way in the sequence.
For a third set of vehicles
The vehicle in (1) does not need to consider the above three collision situations for a while because the split stream processing is not completed yet and the trajectory planning is not performed yet.
For the first vehicle set
The vehicle in (1) has already completed the trajectory planning and is traveling according to the trajectory, and does not need to perform collision analysis any more.
For the second vehicle set
The vehicles in (1), i.e. the cooperative trajectory planning vehicles, have to take into account the above three collision situations, for the second set of vehicles
The preset requirements in step 52 include the following three requirements that need to be met simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
On the left side of the collaborative trajectory planning vehicle; wherein,
representing the collaborative area
The collection of vehicles in the interior of the vehicle,
representing the proximity zone
A set of vehicles located internally on the virtual road centerline;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists between the cooperative track planning vehicle and the cooperative track planning vehicle;
a third requirement when the collaborative trajectory planning vehicle reaches the point of starting the turn, and the vehicle is
Has not exited the collaborative area
。
In an embodiment, if there is a spatial coincidence point i and spatial coincidence points ii and iii on the first optimal trajectory and the second optimal trajectory in fig. 2, respectively, then the path corresponding to the first optimal trajectory is selected to perform speed planning again to obtain a new third optimal trajectory capable of avoiding collision, and the method of "speed planning" in step 53 specifically includes:
step 531, establishing an optimal control model (1):
in the formula (1), the reaction mixture is,
in the form of a cost function, the cost function,
、
weights, for example:
taking out the raw material 5, and taking out,
taking 1;
is as follows
The length of each of the time steps,
is as follows
The length of each of the time steps,
、
respectively planning the acceleration and the speed of the vehicle for the dispersed collaborative track;
equation (2) is the kinematic constraint of the vehicle,
、
respectively planning position coordinates, orientation angles and acceleration of the vehicle for the collaborative trajectory under a Cartesian coordinate system;
equation (3) is the vehicle side value constraint,
、
、
、
planning just-entering cooperative areas of vehicles for cooperative tracks under geodetic coordinate system respectively
Initial position of time, exit from cooperation area
Final position of time, entry into coordination area
The position coordinates and orientation angles of the start point and the end point of (1);
equation (4) is the maximum limit on vehicle speed and acceleration,
is the maximum value of the vehicle speed,
is the maximum value of the vehicle acceleration;
equation (5) is the collision restraint of the vehicle,
、
respectively under the geodetic coordinate system
Coordinate and collaborative track planning vehicle of space coincident point
Vehicle concentrating on collision
To arrive at
Time of a spatial coincidence point
The coordinates of the position are determined by the position,
is the total number of points of spatial coincidence,
、
respectively the lateral and longitudinal safety distances between the vehicles.
According to the detection result, if the vehicle is planned in cooperation with the track
Have all the time been
Travel, where a collision occurs at a point of spatial coincidence, at which the time of collision is known, i.e. the vehicle in the concentration of collisions (higher priority, trajectory already determined) arrives at the second
Time of a nominal collision point
Thus only needing to guarantee
At the moment, the planned vehicle and all N space coincidence points on the set path are larger than the safe distance in the x direction and the y direction
And
collision can be effectively avoided. The optimal control problem can be solved by using an interior point algorithm numerical value or other calculation methods (such as a Runge-Kutta iteration method) for solving the optimal control problem, and the effective track of the vehicle in the embodiment can be obtained.
It should be noted that, for speed planning, in addition to modeling by using a nonlinear programming problem and iterative solution by using a numerical method, modeling solution by using other methods such as integer programming may also be used, and a description thereof is not repeated.
The embodiment of the invention also provides a multi-vehicle cooperative passing system for the intersection under the unstructured scene, which comprises a partition unit, a priority determining unit, a virtual road center line setting unit, a diversion planning unit and a cooperative planning unit, wherein:
the partition unit is used for setting a cooperative area from inside to outside from the center of the intersection in the unstructured scene
Near area of the body
And a flow splitting region
。
A priority determining unit for determining a priority of the vehicle inside each zone and between different zones.
A virtual road center line setting unit for setting a virtual road center line in the access area
Setting a virtual road center line, and taking the virtual road center line and the cooperation area
And taking the intersection point of the outer boundary line as a turning point, and selecting shunting target points at intervals on the virtual road center line from the turning point.
The shunting planning unit is used for sequentially aiming at the shunting areas according to the vehicle priority
Inner vehicle
Planning the shunting track to ensure that the vehicle
Upon entering the collaborative area
And driving to the virtual road center line.
A collaborative planning unit for entering the collaborative area
Inner vehicle
And as a waiting collaborative trajectory planning vehicle, carrying out collaborative trajectory planning according to the vehicle priority.
In one embodiment, the collaborative planning unit specifically includes an optimal trajectory acquisition subunit and a collaborative trajectory planning subunit, where:
an optimal trajectory acquisition subunit for acquiring the optimal trajectory of the first image according to the first image
Inner vehicle
Determining the crossing through which the vehicle leaves the crossing, and taking all the turning points of the crossing as selectable cooperative track planning end points to obtain the optimal track without collision.
A collaborative trajectory planning subunit for future maneuvers of the collaborative trajectory planning vehicle, the approach area
Vehicles meeting preset requirements internally are constructed as collision sets
Judging the set of collisions
In a vehicle
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as the cooperative track; otherwise, judging the maximum value of the vehicle speed of the coordinated track planning vehicle
If so, abandoning the optimal track, otherwise, determining the cooperative track according to the following rule if no collision occurs when the optimal track is defined as:
in principle one, if at least two optimal tracks do not collide, the optimal track with shorter time consumption is preferentially selected as the cooperative track;
in principle two, when all the optimal tracks collide, the optimal track with the minimum number of the spatial coincident points is selected as a cooperative track, and the speed planning for avoiding collision is performed again;
wherein, the preset requirements comprise three requirements which need to be met simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
On the left side of the collaborative trajectory planning vehicle; wherein,
representing the collaborative area
The collection of vehicles in the interior of the vehicle,
representing the proximity zone
A set of vehicles located internally on the virtual road centerline;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists between the cooperative track planning vehicle and the cooperative track planning vehicle;
a third requirement when the collaborative trajectory planning vehicle reaches the point of starting the turn, and the vehicle is
Has not exited the collaborative area
。
In an embodiment, the speed planning method for the collaborative trajectory planning subunit has already been described above, and is not described herein again.
The invention can plan the cooperative track of all intelligent vehicles approaching the non-mechanization signal lamp-free intersection according to the designed AIM system. The road side unit is used for shunting the vehicles and then carrying out collaborative planning, so that the calculated amount is small, and the passing efficiency is effectively improved.
In the above embodiment, the hybrid a-algorithm is used for path planning for multiple times, and other conventional path generation methods such as the a-algorithm and the RRT algorithm may also be used when generating the path.
In fifty experiments performed by the method of the present invention, the number of coordinated vehicles set in each experiment is 24, and the length of the planning time mainly depends on the number of vehicles which cannot avoid collision in the path planning and further avoid collision by adopting the speed planning. The probability of the occurrence of a speed program per vehicle in fifty experiments was 29.2% to 45.8%. The average vehicle planning time for collision avoidance and efficient traffic is 62.67 milliseconds with only hybrid a-path planning. The average time per velocity schedule was 24.25 milliseconds. Therefore, the average planning time of each vehicle in the method is 62.67+ (29.2% -45.8%) multiplied by 24.25=62.67+ (7.08-11.10) = 69.75-73.77 milliseconds.
Compared with other similar collaborative planning methods for intersections, for example, in a numerical solution planning method for a continuous space optimal control problem adopted by plum of the university of Hunan, under an experimental scene of 24 vehicles, the planning time is different from 3.3858 seconds to 3.4572 seconds, and the average planning time of each vehicle is different from 141 milliseconds to 144 milliseconds.
Therefore, the method has the advantages that the calculation time is improved by about 50%, and the planned whole driving path has the characteristic of more orderliness.
Finally, it should be pointed out that: the above examples are only for illustrating the technical solutions of the present invention, and are not limited thereto. Those of ordinary skill in the art will understand that: modifications can be made to the technical solutions described in the foregoing embodiments, or some technical features may be equivalently replaced; such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (9)
1. A method for intersection multi-vehicle cooperative passing under an unstructured scene is characterized by comprising the following steps:
step 1, setting a cooperation area, an approach area and a shunt area from inside to outside from the center of an intersection in an unstructured scene;
step 2, setting a virtual road center line in the approach area, taking an intersection point of the virtual road center line and the outer boundary line of the coordination area as a steering point, and selecting shunting target points at intervals on the virtual road center line from the steering point;
step 3, determining the vehicle priority in each area and among different areas;
step 4, according to the priority of the vehicles, sequentially planning the shunting tracks of the vehicles in the shunting areas, and enabling the vehicles to run to the center line of the virtual road before entering the cooperative area;
and 5, taking the vehicles entering the cooperative area as waiting cooperative track planning vehicles, and performing cooperative track planning according to the vehicle priority.
2. The intersection multi-vehicle cooperative passing method under the unstructured scene as set forth in claim 1, characterized in that the step 5 specifically includes:
step 51, planning future maneuvering of the vehicle according to the cooperative track, determining an intersection through which the vehicle leaves the intersection, and obtaining an optimal collision-free track by taking all turning points of the passing intersection as selectable cooperative track planning end points;
step 52, planning the future maneuver of the vehicle according to the cooperative track, and meeting the requirement in the approaching areaThe vehicles with preset requirements are constructed into a collision set, and the vehicles in the collision set are judged
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as a cooperative track; otherwise, go to step 53; wherein, the preset requirements comprise three requirements which need to be met simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
Located on the left side of the collaborative trajectory planning vehicle; wherein,
representing a set of vehicles within a collaborative area,
representing a set of vehicles located on a virtual road centerline within the proximity zone;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists with the collaborative trajectory planning vehicle;
the third requirement is that when the vehicle reaches the point of starting to turn in cooperation with the trajectory planning, the vehicle
Not exiting the collaborative area;
step 53, judging whether the coordinated track planning vehicle is overlapped in time at the space overlapping point under the condition of constant speed running at the maximum vehicle speed, if so, abandoning the optimal track, otherwise, defining that no collision occurs, and determining the coordinated track according to the following rules:
in a first principle, if at least two optimal tracks do not collide, the optimal track with shorter time consumption is preferentially selected as a cooperative track;
and in principle two, when all the optimal tracks collide, selecting the optimal track with the minimum number of the spatial coincident points as a cooperative track, and performing speed planning for avoiding collision again.
3. The intersection multi-vehicle cooperative passing method under the unstructured scene as set forth in claim 2, characterized in that the method of "speed planning" in step 53 specifically includes:
step 531, establishing an optimal control model (1):
collision restraint s.t.
In the formula (1), the reaction mixture is,
in order to be a function of the cost,
、
the weight of the weight is calculated,
is as follows
The length of each of the time steps,
is as follows
The length of each of the time steps,
、
respectively planning the acceleration and the speed of the vehicle for the vehicle collaborative track to be planned after the time is discretized;
equation (2) is the kinematic constraint of the vehicle,
respectively planning the position coordinate, the orientation angle and the acceleration of the vehicle under the cooperation track of a Cartesian coordinate system,
planning the lateral speed of the vehicle for the coordinated trajectory in the cartesian coordinate system,
planning the longitudinal speed of the vehicle for the coordinated trajectory in the cartesian coordinate system,
planning the angular speed of the vehicle for the coordinated track under a Cartesian coordinate system;
equation (3) is the vehicle side value constraint,
、
、
、
respectively planning the initial position of the vehicle just entering the cooperation area, the final position of the vehicle exiting the cooperation area, and the position coordinates and the orientation angle of the starting point and the end point of the vehicle entering the cooperation area for the cooperation track under the geodetic coordinate system,
for vehicles in initial positions
The speed of (d);
equation (4) is the maximum limit on vehicle speed and acceleration,
is the maximum value of the acceleration of the vehicle,
an upper speed limit established for the unstructured scene;
equation (5) is the collision restraint of the vehicle,
、
respectively under the geodetic coordinate system
Vehicle in collision concentration by planning vehicle coordinates and coordinated tracks of space coincident points
To arrive at
Time of a spatial coincidence point
The coordinates of the position are determined by the position,
is the total number of points of spatial coincidence,
、
respectively the lateral and longitudinal safety distances between the vehicles.
4. The intersection multi-vehicle cooperative passing method under the unstructured scene of any one of claims 1 to 3, characterized in that the method of "setting virtual road center line in the proximity area" in the step 2 specifically comprises:
step 21, translating the road boundaries at the two sides in the approaching area towards the opposite direction for a preset distance, and then performing smoothing treatment to obtain two virtual road center lines;
and step 22, on the basis of the two obtained virtual road center lines, obtaining other virtual road center lines by adopting the translation method in the step 31.
5. The intersection multi-vehicle cooperative passing method under the unstructured scene is characterized in that the priority of the vehicles in each area determined in the step 3 is as follows: vehicles in the same area, the priority of each vehicle being inversely proportional to the distance of the vehicle from the center;
the vehicle priorities between different areas determined in step 3 are: the priority of the vehicle with the lowest priority in the collaborative area in different areas is higher than that of the vehicle with the highest priority in the approaching area, and the priority of the vehicle with the lowest priority in the approaching area in different areas is higher than that of the vehicle with the highest priority in the shunting area.
6. A crossroad multi-vehicle cooperative passing system under an unstructured scene is characterized by comprising:
the system comprises a partitioning unit, a processing unit and a control unit, wherein the partitioning unit is used for setting a cooperation area, an approach area and a shunting area from inside to outside from the center of an intersection in an unstructured scene;
a priority determining unit for determining a priority of the vehicle inside each zone and between different zones;
the virtual road center line setting unit is used for setting a virtual road center line in the approaching area, taking the intersection point of the virtual road center line and the outer boundary line of the collaborative area as a steering point, and selecting shunting target points at intervals on the virtual road center line from the steering point;
the shunting planning unit is used for sequentially planning shunting tracks of the vehicles in the shunting areas according to the priority of the vehicles so that the vehicles can drive to the center line of the virtual road before entering the cooperative area;
and the collaborative planning unit is used for taking the vehicle entering the collaborative area as a waiting collaborative trajectory planning vehicle and carrying out collaborative trajectory planning according to the vehicle priority.
7. The intersection multi-vehicle cooperative passage system under the unstructured scene as set forth in claim 6, characterized in that the cooperative planning unit specifically includes:
the optimal track acquisition subunit is used for planning future maneuvering of the vehicle in cooperation with the track, determining the intersection through which the vehicle leaves the intersection, and obtaining the optimal track without collision by taking all turning points of the passing intersection as selectable cooperative track planning end points;
a cooperative track planning subunit, configured to plan future maneuvers of the vehicle in cooperation with the track, construct vehicles meeting preset requirements in the approaching area as a collision set, and determine the vehicles in the collision set
Whether the number of the space coincident points with the optimal track is 0 or not is judged, and if yes, the optimal track is selected as a cooperative track; otherwise, judging whether the collaborative track planning vehicle is overlapped in time at a space overlapping point under the condition of constant speed running at the maximum speed of the vehicle, if so, abandoning the optimal track, otherwise, determining the collaborative track according to the following rules if no collision occurs:
in a first principle, if at least two optimal tracks do not collide, the optimal track with shorter time consumption is preferentially selected as a cooperative track;
according to a second principle, when all the optimal tracks are collided, selecting the optimal track with the minimum number of space coincident points as a cooperative track, and planning the speed for avoiding collision again;
the preset requirements comprise three requirements which need to be met simultaneously:
in the first place, the first requirement is that,
or
And a vehicle
Located on the left side of the collaborative trajectory planning vehicle; wherein,
to representCollaborative area
The collection of vehicles in the interior of the vehicle,
representing a set of vehicles located on a virtual road centerline within the proximity zone;
second request, vehicle
One of three collision situations, namely cross collision, convergent collision or rear-end collision, exists with the collaborative trajectory planning vehicle;
third requirement, when the coordinated trajectory planning vehicle reaches the starting turning point, the vehicle is turned
Has not yet exited the collaborative area.
8. The intersection multi-vehicle cooperative passage system under the unstructured scene of claim 7, characterized in that the cooperative trajectory planning subunit performs speed planning through the optimal control model (1):
collision restraint s.t.
In the formula (1), the reaction mixture is,
in order to be a function of the cost,
、
the weight of the weight is calculated,
is as follows
The length of each time step is equal to the length of each time step,
is as follows
The length of each time step is equal to the length of each time step,
、
respectively planning the acceleration and the speed of the vehicle for the vehicle collaborative track to be planned after the time is discretized;
equation (2) is the kinematic constraint of the vehicle,
、
respectively planning position coordinates, orientation angles and accelerations of the vehicle for the coordinated track under a Cartesian coordinate system,
planning the lateral speed of the vehicle for the coordinated trajectory in the cartesian coordinate system,
planning the longitudinal speed of the vehicle for the coordinated trajectory in the cartesian coordinate system,
planning the angular speed of the vehicle for the coordinated track under a Cartesian coordinate system;
equation (3) is the vehicle side value constraint,
、
、
、
respectively planning the initial position of the vehicle just entering the cooperation area, the final position of the vehicle exiting the cooperation area, and the position coordinates and the orientation angle of the starting point and the end point of the vehicle entering the cooperation area for the cooperation track under the geodetic coordinate system,
for vehicles in initial positions
The speed of (d);
equation (4) is the maximum limit on vehicle speed and acceleration,
is the maximum value of the acceleration of the vehicle,
an upper speed limit established for the unstructured scene;
equation (5) is the collision restraint of the vehicle,
、
respectively under the geodetic coordinate system
Vehicle in collision concentration by planning vehicle coordinates and coordinated tracks of space coincident points
To arrive at
Time of a spatial coincidence point
The coordinates of the position are determined by the position,
is the total number of points of spatial coincidence,
、
respectively the lateral and longitudinal safety distances between the vehicles.
9. The intersection multi-vehicle cooperative passing system under the unstructured scene of any one of claims 6 to 8, characterized in that the method of the virtual road center line setting unit specifically comprises:
firstly, the road boundaries at two sides in the approaching area are translated towards the opposite direction for a preset distance, then smoothing is carried out to obtain two virtual road center lines, and on the basis of the obtained two virtual road center lines, the same translation method is adopted to obtain other virtual road center lines.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210812912.3A CN114898564B (en) | 2022-07-12 | 2022-07-12 | Intersection multi-vehicle cooperative passing method and system under unstructured scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210812912.3A CN114898564B (en) | 2022-07-12 | 2022-07-12 | Intersection multi-vehicle cooperative passing method and system under unstructured scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114898564A true CN114898564A (en) | 2022-08-12 |
| CN114898564B CN114898564B (en) | 2022-09-30 |
Family
ID=82730114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210812912.3A Active CN114898564B (en) | 2022-07-12 | 2022-07-12 | Intersection multi-vehicle cooperative passing method and system under unstructured scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114898564B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115620543A (en) * | 2022-09-29 | 2023-01-17 | 斯润天朗（北京）科技有限公司 | Multi-vehicle cooperative passing method and device at traffic intersection and computer equipment |
| CN115662157A (en) * | 2022-10-28 | 2023-01-31 | 东风汽车集团股份有限公司 | V2X-based intersection vehicle passing control method, device, equipment and medium |
| CN115840454A (en) * | 2023-02-20 | 2023-03-24 | 江苏集萃清联智控科技有限公司 | Multi-vehicle track collaborative planning method and device for unstructured road conflict area |
| CN116630467A (en) * | 2023-03-15 | 2023-08-22 | 广州小鹏自动驾驶科技有限公司 | Virtual lane construction method, device, equipment and computer readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003467A (en) * | 2017-06-07 | 2018-12-14 | 华为技术有限公司 | A kind of method, apparatus and system preventing vehicle collision |
| CN109979217A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Cooperative intersection passing control method, device and equipment |
| CN111338335A (en) * | 2019-12-31 | 2020-06-26 | 清华大学 | Vehicle local track planning method under structured road scene |
| CN113932823A (en) * | 2021-09-23 | 2022-01-14 | 同济大学 | Unmanned multi-target-point track parallel planning method based on semantic road map |
| CN114120664A (en) * | 2022-01-05 | 2022-03-01 | 北京航空航天大学合肥创新研究院(北京航空航天大学合肥研究生院) | Mixed traffic intersection signal and vehicle track cooperative control method based on game theory |
| CN114283607A (en) * | 2020-12-21 | 2022-04-05 | 北京邮电大学 | Multi-vehicle collaborative planning method based on distributed crowd-sourcing learning |
-
2022
- 2022-07-12 CN CN202210812912.3A patent/CN114898564B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003467A (en) * | 2017-06-07 | 2018-12-14 | 华为技术有限公司 | A kind of method, apparatus and system preventing vehicle collision |
| CN109979217A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Cooperative intersection passing control method, device and equipment |
| CN111338335A (en) * | 2019-12-31 | 2020-06-26 | 清华大学 | Vehicle local track planning method under structured road scene |
| CN114283607A (en) * | 2020-12-21 | 2022-04-05 | 北京邮电大学 | Multi-vehicle collaborative planning method based on distributed crowd-sourcing learning |
| CN113932823A (en) * | 2021-09-23 | 2022-01-14 | 同济大学 | Unmanned multi-target-point track parallel planning method based on semantic road map |
| CN114120664A (en) * | 2022-01-05 | 2022-03-01 | 北京航空航天大学合肥创新研究院(北京航空航天大学合肥研究生院) | Mixed traffic intersection signal and vehicle track cooperative control method based on game theory |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115620543A (en) * | 2022-09-29 | 2023-01-17 | 斯润天朗（北京）科技有限公司 | Multi-vehicle cooperative passing method and device at traffic intersection and computer equipment |
| CN115662157A (en) * | 2022-10-28 | 2023-01-31 | 东风汽车集团股份有限公司 | V2X-based intersection vehicle passing control method, device, equipment and medium |
| CN115840454A (en) * | 2023-02-20 | 2023-03-24 | 江苏集萃清联智控科技有限公司 | Multi-vehicle track collaborative planning method and device for unstructured road conflict area |
| CN116630467A (en) * | 2023-03-15 | 2023-08-22 | 广州小鹏自动驾驶科技有限公司 | Virtual lane construction method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114898564B (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114898564B (en) | Intersection multi-vehicle cooperative passing method and system under unstructured scene | |
| CN110298122B (en) | Decision-making method for left-turning at urban intersection of unmanned vehicles based on conflict resolution | |
| US11532167B2 (en) | State machine for obstacle avoidance | |
| Li et al. | Planning and decision-making for connected autonomous vehicles at road intersections: A review | |
| US11427191B2 (en) | Obstacle avoidance action | |
| CN108256233B (en) | Intelligent vehicle trajectory planning and tracking method and system based on driver style | |
| CN113916246A (en) | A method and system for unmanned obstacle avoidance path planning | |
| CN110667585B (en) | Vehicle control device, vehicle control method, and storage medium | |
| CN108225364A (en) | A kind of pilotless automobile driving task decision system and method | |
| CN112193244B (en) | Linear Constraint-Based Motion Planning Method for Autonomous Driving Vehicles | |
| CN112820125B (en) | Passage guidance method and guidance system for intelligent networked vehicles in the case of mixed vehicles | |
| CN113932823A (en) | Unmanned multi-target-point track parallel planning method based on semantic road map | |
| JP7043295B2 (en) | Vehicle control devices, vehicle control methods, and programs | |
| WO2018123019A1 (en) | Vehicle control system, vehicle control method, and vehicle control program | |
| CN113619603B (en) | Method for planning turning track of double-stage automatic driving vehicle | |
| CN113554886B (en) | A method for resolving traffic conflict at a coordinated vehicle-road intersection | |
| CN110782707A (en) | Automobile lane change control method and device, automobile and storage medium | |
| WO2021086781A1 (en) | Obstacle avoidance action | |
| Guo et al. | Toward human-like behavior generation in urban environment based on Markov decision process with hybrid potential maps | |
| Levy et al. | Path and trajectory planning for autonomous vehicles on roads without lanes | |
| Chen et al. | A decision-making model for autonomous vehicles at intersections based on hierarchical reinforcement learning | |
| Kim et al. | Trajectory planning for autonomous highway driving using the adaptive potential field | |
| Huang et al. | Automated turning and merging for autonomous vehicles using a Nonlinear Model Predictive Control approach | |
| CN117962924A (en) | Trajectory planning method for intersection conflict situations | |
| CN117864135A (en) | Vehicle lane-changing steering assistance method based on driver's operating comfort |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |