CN116198540A

CN116198540A - Scene migration-based traveling parking decision method, device and storage medium

Info

Publication number: CN116198540A
Application number: CN202310187434.6A
Authority: CN
Inventors: 李振; 王清; 邱利宏; 孔周维; 任凡
Original assignee: Chongqing Changan Automobile Co Ltd
Current assignee: Chongqing Changan Automobile Co Ltd
Priority date: 2023-03-02
Filing date: 2023-03-02
Publication date: 2023-06-02

Abstract

The invention relates to the technical field of automatic driving, in particular to a traveling parking decision method, a traveling parking decision device and a storage medium based on scene migration, wherein the traveling parking decision method comprises the following steps: the scene design comprises a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene; judging whether to enter the matched scene based on the entry condition close to the scene; and judging whether to exit the matched scene or not based on the exit condition close to the scene. The parking integrated decision-making method can be used for communicating typical road sections such as a garage, an urban open road, an expressway and a u-turn road, realizing automatic round trip from the home garage to a working area (or a leisure area), and providing continuous and easy automatic driving vehicle experience for users. And the algorithms in various scenes can be integrated together, and different algorithm modules can be operated in a scene scheduling mode.

Description

Scene migration-based traveling parking decision method, device and storage medium

Technical Field

The invention belongs to the technical field of automatic driving, and particularly relates to a scene migration-based traveling and parking decision method, a device and a storage medium.

Background

The automatic driving vehicle is a comprehensive intelligent system integrating multiple functions of environment sensing, planning decision, behavior control, execution and the like, and covers multiple disciplinary knowledge of machinery, control, sensor technology, signal processing, pattern recognition, artificial intelligence, computer technology and the like. The intelligent vehicle with independent driving capability is developed, and has important practical significance for developing vehicle active safety auxiliary driving products with independent intellectual property rights in China, improving the intelligent level of independent brand automobiles in China, improving road traffic safety conditions and developing intelligent traffic systems. In an automatic driving vehicle, decision planning is a key technology, is a hot spot and difficult problem of current research, and is an important part for embodying intelligence by combining perception, prediction and control.

Typical car use scenario: the parking space is parked from a home garage, the vehicle automatically passes through a gate, an urban road is parked, a signal lamp is started and stopped, a protection left turn, an obstacle avoidance between lanes, a right turn at an intersection, an automatic ramp-up on an ETC, an automatic expressway, a vehicle-changing overtaking, an automatic ramp-down, a speed reduction avoidance of a bus station in a mixed area of a person and a vehicle, an unprotected left turn, a turning around at an intersection, an automatic merging, and a speed regulation according to a speed regulation plate are arranged at a parking space in a working area (or a leisure area). Typical scenarios involved are: garage, urban open road, expressway and expressway. The integrated automatic driving scheme of traveling and parking can be communicated with the road sections, and the vehicle can travel from a home garage to a garage in a working area (or a leisure area) to provide continuous and easy automatic driving vehicle experience for users. Therefore, the research of the traveling parking decision method has important significance for improving the adaptability of the complex automatic driving scene.

Disclosure of Invention

The purpose of the invention is that: aims to provide a scene migration-based parking decision method, a scene migration-based parking decision device and a storage medium, which are used for solving the problem of difficult decision planning of the existing automatic driving vehicle

In order to achieve the technical purpose, the invention adopts the following technical scheme:

in a first aspect, the present application provides a scenario migration-based approach to a row-poise decision method, including the steps of:

s110, a scene design, wherein the scene comprises a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene, and the scene design comprises an entry condition design and an exit condition design;

s120, judging whether to enter the matched scene based on the entry condition close to the scene;

s130, judging whether to exit the matched scene based on the exit condition close to the scene.

With reference to the first aspect, in some optional embodiments, the entering condition of the lane following scene is that the entering condition of other scenes except the lane following scene is not satisfied after any of the scenes is ended, and the exiting condition of the lane following scene is that the entering condition of other scenes except the lane following scene is satisfied and is not currently a lane change and borrowing state.

With reference to the first aspect, in some optional embodiments, the entering condition of the intersection scene is that the front link is an intersection, and the distance between the vehicles according to the intersection link is smaller than a first preset value, where the intersection scene includes an entering stage, an intersection probe stage and an intersection cruising stage, and the exiting condition of the intersection scene is that: and in the crossing cruising stage, the current link is not a crossing and the distance between the starting points of the vehicles according to the current link is larger than a second preset value.

With reference to the first aspect, in some optional embodiments, the garage driving scene includes a garage driving scene and a garage driving scene, an entering condition of the garage driving scene is that a distance from a garage entrance is smaller than a third preset value, and an exiting condition of the garage driving scene is that a distance from a parking space is smaller than a fourth preset value; the entering condition of the driving scene of the garage is that the parking place is completed, or the vehicle is not in the parking place and the destination is outside the garage, and the exiting condition of the garage is that the vehicle realizes that the garage turns to a target lane to realize centering and steering wheel centering.

With reference to the first aspect, in some optional embodiments, the garage parking scene includes a parking scene and a parking scene, an entry condition of the parking scene is that a distance from the parking space is smaller than a fifth preset value, and an exit condition of the parking scene is that a feedback of completion of parking is received; the entry condition of the parking scene is that a parking instruction is received and executed, and the exit condition of the parking scene is that a parking completion instruction is received.

With reference to the first aspect, in some optional embodiments, the entering condition of the u-turn scene is that the navigation gives a u-turn instruction and the distance from the u-turn point is smaller than a sixth preset value, or the navigation does not send the u-turn instruction but the vehicle is in a u-turn lane, and the distance from the u-turn point is smaller than or equal to a seventh preset value; the exit condition of the turning scene is that the vehicle turns around to a target lane to realize centering and the steering wheel is aligned.

With reference to the first aspect, in some optional embodiments, after the current scene exits, if the entering condition of other scenes is not satisfied, switching needs to be started from the lane following scene, and if multiple scenes are satisfied simultaneously, entering a scene with the highest priority according to the scene priorities, where the priorities are ranked as follows: the garage driving scene is larger than the turning scene, is larger than the intersection scene and is larger than the lane following scene.

In a second aspect, the present application provides a scene migration-based row parking decision device, which is characterized by comprising a scene design unit, an entry judgment unit and an exit judgment unit, wherein each module has the following functions:

the scene design unit is used for scene design, wherein the scene comprises a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene; the scene design comprises an entry condition design and an exit condition design;

an entry judgment unit that judges whether to enter the matched scene based on an entry condition close to the scene;

and an exit judging unit for judging whether to exit the matched scene based on the exit condition close to the scene.

In a third aspect, the present application provides a computer storage medium, wherein the computer storage medium has stored therein a computer program, and when the computer program runs on a computer, the computer program is capable of executing the scene migration-based berth decision method as described above.

The invention adopting the technical scheme has the following advantages:

1. the integrated decision method for traveling and parking based on scene migration is provided, and can be used for communicating typical road sections such as a garage, an urban open road, an expressway, a turning around and the like, realizing automatic round trip from the home garage to a working area (or a leisure area) and providing continuous and easy automatic driving vehicle experience for users;

2. in the method, algorithms in various scenes can be integrated together, and different algorithm modules can be operated in a scene scheduling mode;

3. each basic algorithm module can be multiplexed in each scene, so that the code development amount is reduced, and the maintenance cost is also reduced;

4. the newly added scene can be expanded, so that the subsequent wider complex urban scene can be dealt with.

Drawings

The invention can be further illustrated by means of non-limiting examples given in the accompanying drawings;

FIG. 1 is a flow chart of a scenario migration-based row poise decision method in an embodiment of the present application;

FIG. 2 is a view of a scene usage architecture in an embodiment of the present application;

FIG. 3 is a schematic diagram of scene switching in an embodiment of the present application;

fig. 4 is a schematic diagram of switching between a lane following scene and a garage driving scene in the embodiment of the present application;

FIG. 5 is a decision chart of a lane following scenario in an embodiment of the present application;

FIG. 6 is a decision diagram of an intersection scene in an embodiment of the present application;

fig. 7 is a decision chart of a driving scene of a garage in an embodiment of the application;

FIG. 8 is a decision chart of driving scenes of a garage in an embodiment of the application;

FIG. 9 is a parking scenario decision chart according to an embodiment of the present application;

FIG. 10 is a parking scenario decision chart according to an embodiment of the present application;

FIG. 11 is a decision diagram of a turning scene in an embodiment of the present application;

fig. 12 is a schematic diagram of a scenario migration-based berthing decision device in an embodiment of the present application.

The main reference numerals are as follows:

10: a scene design unit; 20: an entry condition judgment unit; 30: and an exit condition judgment unit.

Detailed Description

The present invention will be described in detail below with reference to the drawings and the specific embodiments, wherein like or similar parts are designated by the same reference numerals throughout the drawings or the description, and implementations not shown or described in the drawings are in a form well known to those of ordinary skill in the art. In addition, directional terms such as "upper", "lower", "top", "bottom", "left", "right", "front", "rear", etc. in the embodiments are merely directions with reference to the drawings, and are not intended to limit the scope of the present invention.

As shown in fig. 1, an embodiment of the present application provides a scenario migration-based row poise decision method, which includes the following steps:

First, the definition of a scene is: the scene can be a road section with certain characteristics in the map, such as an intersection, or a series of complex actions which the unmanned vehicle wants to complete, such as road borrowing and avoiding. The scene has the advantages that: the scenes are not interfered with each other, and parallel development and parameter adjustment can be performed; a series of complex tasks with timing or dependencies may be implemented in one scenario. The disadvantage/difficulty of the scenes is that each scene needs to process some basic cases, which may bring about redundancy of codes; the non-coverage division between scenes is very difficult; additional logic is required to perform scene recognition and to handle transitions between scenes. As shown in fig. 2, the scene is further identified through a series of data collection and processing during actual application, including map information collection, lane positioning, and track planning after identifying the scene.

With reference to fig. 3, scene recognition and switching. Defining a lane following scene as a default scene, judging whether to enter a corresponding scene mainly according to a approaching scene entering condition, switching from the default scene if other scenes are not met after the current scene is completed to exit, and entering a scene with highest priority according to scene priorities if a plurality of scenes are met simultaneously, wherein a garage driving scene is larger than a turning scene and larger than an intersection scene and larger than the lane following scene. The corresponding scene is not easy to exit after entering, and the current scene does not exit until each stage of the current scene is completed. Errors occur in the scene, and the scene itself can also decide whether to exit. For example, fig. 4 shows a switching between a lane following scene and a garage driving scene. When the distance in the direction of going to the garage is satisfied by the lane following scene and the distance from the garage in the garage driving scene is satisfied, switching the lane following scene to the garage driving scene; when the garage driving scene is located and the garage driving scene is far away from the garage, the garage driving scene is switched to the lane following scene when the garage driving scene is away from the garage, the garage is taken out of the garage and a distance is kept away from the garage. In general, the data of the entry condition and the output condition includes positioning, high-definition map, perceived traffic light, vehicle speed, and the like.

In the scene design step, the scenes comprise a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene.

In conjunction with fig. 5, in the lane following scenario:

entry conditions: as a default scene, any scene after finishing is required to start from the default scene if other scenes are not satisfied;

exit conditions: other scenes are met, and the current channel changing and channel borrowing stage is not performed;

the scene has only one stage, and the entering and exiting conditions of the stage are the same as the entering and exiting conditions of the scene.

Referring to fig. 6, in the intersection scenario:

entry conditions: the front link is an intersection, and the distance between the vehicle and the intersection link is smaller than a first preset value, wherein the first preset value is 100m;

approach stage: at the stage, determining whether to turn right, go straight and turn left according to a navigation map, finishing lane changing to a target steering lane in advance, avoiding lane crossing behaviors (except for stationary obstacles or roadside zombie vehicles), and determining whether to stop or slow down before stopping according to traffic light states, avoidance and stop marks;

crossing probe stage: after the condition that the traffic can be carried out is met (green light), a probe is slowly carried out at the intersection, so that the visual field is enlarged, the object vehicles are observed, pedestrians at the intersection are observed, and safety is confirmed;

crossing cruising phase: after the safety is confirmed, the speed can be properly increased to pass through the intersection;

exit conditions: in the crossing cruising stage, the condition that the current link is not a crossing is satisfied, and the distance between the vehicle and the starting point of the current link is greater than a second preset value (TBD), wherein the second preset value is set to be 5m.

Referring to fig. 7 and 8, the garage driving scene includes a garage driving scene and a garage driving scene. As in fig. 7, in the garage driving scenario:

entry conditions: the distance from the garage entrance < a third preset value, which in this application is set to 100m.

Approach stage: in the stage, whether to turn right, go straight and turn left is determined according to the navigation map, lane changing to a target steering lane is completed in advance, lane passing behavior (except stationary obstacles or roadside zombie vehicles) is avoided, backward and garage opposite incoming vehicles are noted in the process of turning left to a garage opening, and garage opposite incoming vehicles are noted in the process of turning right to the garage opening;

the stage of passing through the garage mouth, namely paying attention to the fact that vehicles come in the opposite direction from the garage, if a lifting rod (main purpose) is provided, the vehicles should pass through in a decelerating way or the vehicles should be stopped for waiting to pass through;

lane following stage: following the lane in the garage at this stage

Exit conditions: the distance from park_spot is < a fourth preset value, where the fourth preset value is set to 100m.

Referring to fig. 8, in the garage-out driving scenario:

entry conditions: the parking place is completed, or the parking place is not currently a parking place, the vehicle is no longer on the parking place, and the destination is outside the garage;

lane following stage: the garage runs along with the lane;

through garage exit phase (lifter): the main purpose is to pass through the lifting rod in a decelerating way or to wait for passing through a parking way;

an outlet turning stage: after passing through the lifting rod, in the process of needing to turn and merge into a main road, a probe is required to be slowly speed, a backward coming vehicle of a target lane is observed, and after safety is confirmed, the vehicle is properly accelerated to turn to the target lane to realize lane centering;

exit conditions: the vehicle can realize the turning from the garage to the target lane and the centering, and the steering wheel is aligned.

The garage parking scenes include a parking in scene and a parking out scene.

As in fig. 9, in a parking scenario:

entry conditions: the distance from the park_spot is < a fifth preset value, where the fifth preset value is 100m;

approach stage: the vehicle needs to travel along a lane at the stage, and meanwhile, empty parking spaces are detected and stored, the detected parking spaces are matched with the park_spot, and if the park_spot is not the empty parking space, the nearest empty parking space is selected as a target parking space;

and (3) parking stage: after searching the target parking space, parking into the space is started;

exit conditions: and receiving a feedback of the completion of parking.

As in fig. 10, in the parking scenario:

entry conditions: receiving an execution berthing instruction;

and (3) exiting the condition, namely receiving a berth-out completion instruction.

As in fig. 11, in the u-turn scenario:

entry conditions:

the navigation gives a turning instruction, the distance from a turning point is smaller than a sixth preset value (set as 100 m) or the navigation does not need turning, but the vehicle is on a lane which can only turn around, and the distance from the turning point is smaller than a seventh preset value (set as 5 m);

approach stage: under the stage, lane changing to a target turning lane should be completed in advance, lane passing behavior (except for stationary obstacles or roadside zombie vehicles) should be avoided, and whether parking is performed before stopping the lane is determined according to the traffic light state;

and (3) a turning phase: turning around is executed, the vehicle should pass through slowly, the backward coming vehicle of the target lane is observed, and after safety is confirmed, the vehicle is accelerated to the target lane to realize lane centering;

exit conditions: the vehicle turns around to the target lane to realize centering and the steering wheel is aligned.

When scene switching is needed, the method comprises the following steps: 1. starting from a default scene; 2. judging whether to enter a corresponding scene or not mainly according to the distance to the scene; 3. the corresponding scene is not easy to exit after entering, and the current scene does not exit until each stage of the current scene is completed.

The embodiment of the application also provides a scene migration-based traveling parking decision device, which comprises at least one software functional module stored in a storage module in the form of software or Firmware (Firmware) or solidified in an Operating System (OS) in a control device. The scene design unit 10 and other units are configured to execute executable modules stored in the storage module, for example, a software function module and a computer program module included in the scene migration-based parking decision device.

As shown in fig. 12, the apparatus includes a scene design unit 10 and an entry judgment unit 20 and an exit judgment unit 30 coupled to the scene design unit 10, respectively. The functions of each unit may be as follows:

the scene design unit 10 is used for scene design, wherein the scenes comprise a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene; the scene design comprises an entry condition design and an exit condition design;

an entry judgment unit 20 that judges whether to enter the scene that matches based on an entry condition that is close to the scene;

an exit judging unit 30 judges whether to exit the matched scene based on an exit condition close to the scene.

In this embodiment, the memory module may be, but is not limited to, a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc. In the present embodiment, the storage module may be used to store a plurality of preset values and the like in the entry judgment unit 20 and the exit judgment unit 30. Of course, the storage module may also be used to store a program, and the processing module executes the program after receiving the execution instruction. The scene design process, the entry condition and the exit condition corresponding to each scene, etc. may refer to the above display method, and will not be described in detail here.

The embodiment of the application further provides a computer storage medium, in which a computer program is stored, which when executed on a computer, causes the computer to execute the scene migration-based berth decision method as described in the above embodiment.

From the foregoing description of the embodiments, it will be apparent to those skilled in the art that the present application may be implemented in hardware, or by means of software plus a necessary general hardware platform, and based on this understanding, the technical solution of the present application may be embodied in the form of a software product, where the software product may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disc, a mobile hard disk, etc.), and includes several instructions to cause a computer device (may be a personal computer, an electronic device, or a network device, etc.) to perform the methods described in the respective implementation scenarios of the present application.

In summary, the present application provides a scenario migration-based traveling parking decision method, apparatus and storage medium, where in the traveling parking decision method, the steps of the method include: the scene design comprises a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene; judging whether to enter the matched scene based on the entry condition close to the scene; and judging whether to exit the matched scene or not based on the exit condition close to the scene. The parking integrated decision-making method can be used for communicating typical road sections such as a garage, an urban open road, an expressway and a u-turn road, realizing automatic round trip from the home garage to a working area (or a leisure area), and providing continuous and easy automatic driving vehicle experience for users. And the algorithms in various scenes can be integrated together, and different algorithm modules can be operated in a scene scheduling mode.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus, system, and method may be implemented in other manners as well. The above-described apparatus, systems, and method embodiments are merely illustrative, for example, flow charts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. In addition, the functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.

The invention provides a scene migration-based row parking decision method, a scene migration-based row parking decision device and a storage medium. The description of the specific embodiments is only intended to aid in understanding the method of the present invention and its core ideas. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the invention can be made without departing from the principles of the invention and these modifications and adaptations are intended to be within the scope of the invention as defined in the following claims.

Claims

1. The traveling and parking decision method based on scene migration is characterized by comprising the following steps:

2. The scene migration-based traveling parking decision method according to claim 1, wherein the entry condition of the lane following scene is that the entry condition of the other scenes except the lane following scene is not satisfied after any of the scenes is ended, the exit condition of the lane following scene is that the entry condition of the other scenes except the lane following scene is satisfied, and the current vehicle is not in a lane change and borrow state.

3. The scene migration-based traveling parking decision method according to claim 2, wherein the entry condition of the intersection scene is that a front link is an intersection, the distance between vehicles according to the intersection link is smaller than a first preset value, the intersection scene comprises an entry stage, an intersection probe stage and an intersection cruising stage, and the exit condition of the intersection scene is that: and in the crossing cruising stage, the current link is not a crossing and the distance between the starting points of the vehicles according to the current link is larger than a second preset value.

4. The scene migration-based parking decision method according to claim 3, wherein the garage driving scene comprises a garage driving scene and a garage driving scene, the entry condition of the garage driving scene is that the distance from the garage entrance is smaller than a third preset value, and the exit condition of the garage driving scene is that the distance from the parking space is smaller than a fourth preset value; the entering condition of the driving scene of the garage is that the parking place is completed, or the vehicle is not in the parking place and the destination is outside the garage, and the exiting condition of the garage is that the vehicle realizes that the garage turns to a target lane to realize centering and steering wheel centering.

5. The scene migration-based row parking decision method according to claim 4, wherein the garage parking scene comprises a parking scene and a parking scene, the parking scene has an entry condition that a distance from the parking space is smaller than a fifth preset value, and an exit condition of the parking scene is that a feedback of parking completion is received; the entry condition of the parking scene is that a parking instruction is received and executed, and the exit condition of the parking scene is that a parking completion instruction is received.

6. The scene migration-based traveling-parking decision method according to claim 5, wherein the entry condition of the turning scene is that navigation gives a turning instruction and the distance from a turning point is smaller than a sixth preset value, or that navigation does not give the turning instruction but the vehicle is in a turning lane and the distance from the turning point is smaller than or equal to a seventh preset value; the exit condition of the turning scene is that the vehicle turns around to a target lane to realize centering and the steering wheel is aligned.

7. The scene migration-based traveling parking decision method according to claim 6, wherein switching from the lane following scene is required if the entry condition of other scenes is not satisfied after the current scene exits, and the scene with the highest priority is entered according to the scene priorities if a plurality of scenes are satisfied at the same time, wherein the priority is ordered such that the garage traveling scene is larger than the turning scene and larger than the intersection scene and larger than the lane following scene.

8. The scene migration-based berthing decision device is characterized by comprising a scene design unit (10), an entry judgment unit (20) and an exit judgment unit (30), wherein each module has the following functions:

the scene design unit (10) is used for scene design, wherein the scene comprises a lane following scene, an intersection scene, a garage driving scene, a garage parking scene and a turning scene, and the scene design comprises an entry condition design and an exit condition design;

an entry judgment unit (20) that judges whether to enter the scene that matches, based on an entry condition that is close to the scene;

an exit judgment unit (30) judges whether to exit the matched scene based on an exit condition close to the scene.

9. A computer storage medium, wherein a computer program is stored in the computer storage medium, and when the computer program is run on a computer, the computer program is capable of executing the scene migration-based berthing decision method according to any of claims 1-7.