Vehicle control method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a vehicle control method and device, electronic equipment and a computer readable storage medium. The method comprises the steps that vehicle collection information of a to-be-controlled vehicle at the current moment and map information of the current position of the to-be-controlled vehicle are acquired; performing information conversion on the vehicle acquisition information and the map information to obtain a lane mark of the current position, lane-level road section data of the current position and a road mark of the current position; determining a target driving scene to which the to-be-controlled vehicle belongs at the current moment based on the lane identifier, the lane-level road section data and the road sign; determining a planned track of the to-be-controlled vehicle based on the target driving scene and the lane-level road section data; and controlling the to-be-controlled vehicle to run according to the planned track. According to the invention, the adaptability of the to-be-controlled vehicle in a complex driving scene can be improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a vehicle control method, device, electronic device, and computer-readable storage medium. Background Art

[0002] As the core component of autonomous vehicles, the decision-making system plays a key role in improving driving safety and environmental adaptability. The decision-making system can output driving behaviors that meet preset requirements based on the perception of the external environment and the judgment of the internal state, and the performance of the driving behavior will directly affect the final effect of autonomous driving. Therefore, the decision-making system occupies a very important position in autonomous driving.

[0003] Most of the research on autonomous driving decision modules in related technologies is based on a single goal, such as generating behavioral intentions or trajectory outputs based on environmental information. This approach is difficult to comprehensively handle multiple mode switching requirements in complex and dynamic traffic scenarios. In addition, the decision frameworks in related technologies are mostly designed for a single scenario and are difficult to adapt to diverse driving needs. The decision-making method based on deep learning needs to rely on a large amount of high-quality driving data, and it is difficult to quickly iterate and update in the absence of transparency and explainability. Therefore, autonomous driving vehicles cannot cope with complex traffic environments and are difficult to adapt to diverse driving needs. Summary of the invention

[0004] The embodiments of the present application provide a vehicle control method, device, electronic device and computer-readable storage medium, which can improve the adaptability of the vehicle to be controlled in complex driving scenarios.

[0005] The technical solution of the embodiment of the present application is implemented as follows:

[0006] An embodiment of the present application provides a vehicle control method, which includes: obtaining vehicle collection information of a vehicle to be controlled at a current moment and map information of the current position of the vehicle to be controlled; performing information conversion on the vehicle collection information and the map information to obtain a lane marking at the current position, lane-level road section data at the current position, and a road sign at the current position; determining a target driving scenario to which the vehicle to be controlled belongs at the current moment based on the lane marking, the lane-level road section data, and the road sign; determining a planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data; and controlling the vehicle to be controlled to travel according to the planned trajectory.

[0007] An embodiment of the present application provides a vehicle control device, including: an information acquisition module, used to acquire vehicle collection information of a vehicle to be controlled at a current moment and map information of the current position of the vehicle to be controlled; an information conversion module, used to convert the vehicle collection information and the map information to obtain the lane identification of the current position, the lane-level road section data of the current position and the road sign of the current position; a scene determination module, used to determine the target driving scene to which the vehicle to be controlled belongs at the current moment based on the lane identification, the lane-level road section data and the road sign; a trajectory determination module, used to determine the planned trajectory of the vehicle to be controlled based on the target driving scene and the lane-level road section data; and a control module, used to control the vehicle to be controlled to travel according to the planned trajectory.

[0008] An embodiment of the present application provides an electronic device, which includes: a memory for storing computer-executable instructions or computer programs; and a processor for executing the computer-executable instructions or computer programs stored in the memory to implement the vehicle control method provided in the embodiment of the present application.

[0009] An embodiment of the present application provides a computer-readable storage medium storing a computer program or computer-executable instructions for implementing the vehicle control method provided in the embodiment of the present application when executed by a processor.

[0010] An embodiment of the present application provides a computer program product, including a computer program or computer executable instructions. When the computer program or computer executable instructions are executed by a processor, the vehicle control method provided by the embodiment of the present application is implemented.

[0011] The embodiments of the present application have the following beneficial effects:

[0012] First, the vehicle collection information and map information of the vehicle to be controlled at the current moment are obtained; then, the lane marking, lane-level section data and road signs are obtained from the vehicle collection information and map information; then, the target driving scene to which the vehicle to be controlled belongs at the current moment is determined through the lane marking, lane-level section data and road signs. In this way, the specific driving scene of the vehicle to be controlled at the current moment can be determined, so as to divide the driving scenes of the vehicle to be controlled in complex traffic environments; then, according to the target driving scene and lane-level section data, the planned trajectory of the vehicle to be controlled is determined; finally, the vehicle to be controlled is controlled to travel according to the planned trajectory. In this way, corresponding response plans can be adopted for different driving scenarios, so as to plan the vehicle's driving trajectory in a refined manner, thereby improving the adaptability of the vehicle to be controlled in complex driving scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is an optional flow chart of a vehicle control method provided in an embodiment of the present application;

[0014] Figure 2 is another optional flow chart of the vehicle control method provided in the embodiment of the present application;

[0015] Figure 3 It is a schematic diagram of the implementation process of information conversion provided by the embodiment of the present application;

[0016] Figure 4 This is a schematic diagram of an implementation process of determining a planning trajectory provided in an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of a preset behavior tree provided in an embodiment of the present application;

[0018] Figure 6 is another implementation flow diagram of determining a planning trajectory provided in an embodiment of the present application;

[0019] Figure 7 It is a schematic diagram of the implementation process of autonomous driving hierarchical decision-making provided in an embodiment of the present application;

[0020] Figure 8 is a schematic diagram of scene switching provided by an embodiment of the present application;

[0021] Fig. 9 It is a schematic diagram of the architecture of hierarchical decision-making for autonomous driving provided in an embodiment of the present application;

[0022] Fig.10 is a structural block diagram of a vehicle control device provided in an embodiment of the present application;

[0023] Fig.11 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0025] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0026] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0027] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0028] Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meanings as those commonly understood by those skilled in the art. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0029] The relevant data collection and processing in the embodiments of this application should be strictly in accordance with the requirements of relevant laws and regulations when applied in examples, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of authorization of laws and regulations and the personal information subject.

[0030] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0031] 1) In response to: used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed may be executed.

[0032] 2) Autonomous driving: refers to the ability of a vehicle to autonomously control its driving through computer systems, sensors and algorithms without intervention from a human driver.

[0033] 3) Task layer: The highest level of the autonomous driving decision-making module, the task layer defines the overall tasks and goals that the vehicle needs to complete. The task layer is responsible for breaking down high-level tasks into more specific subtasks and assigning these subtasks to the scenario layer and behavior layer.

[0034] 4) Scenario layer: The middle layer of the autonomous driving decision module, the scenario layer processes specific driving scenarios and environmental conditions. The scenario layer generates a driving strategy suitable for the current scenario based on the subtasks assigned by the task layer and the current environmental information (such as road type, traffic flow, weather conditions, etc.). The scenario layer needs to identify and understand different driving scenarios and adjust the strategy to adapt to changes in driving scenarios.

[0035] 5) Behavior layer: The lowest level of the autonomous driving decision-making module, the scenario layer directly controls the specific behavior of the vehicle. The behavior layer generates specific control commands such as acceleration, deceleration, and steering based on the strategies provided by the scenario layer. The behavior layer needs to respond to environmental changes in real time and ensure that the vehicle's behavior complies with safety standards and traffic regulations.

[0036] In order to better understand the vehicle control method provided in the embodiment of the present application, the vehicle control method in the related art is first described below.

[0037] In related technologies, the research on autonomous driving decision modules is mostly based on a single goal, such as generating behavioral intentions or trajectory outputs based on environmental information, but usually ignores the collaborative working mechanism of the task layer, scenario layer and behavior layer. The autonomous driving decision-making method based on a single goal is difficult to comprehensively handle multiple mode switching requirements in complex and dynamic traffic scenarios.

[0038] Based on the problems existing in the related art, the embodiments of the present application provide a vehicle control method, device, electronic device, computer-readable storage medium and computer program product, which can improve the adaptability of the vehicle to be controlled in complex driving scenarios. Specifically, first, the vehicle collection information and map information of the vehicle to be controlled at the current moment are obtained; then, the lane identification, lane-level road section data and road signs are obtained from the vehicle collection information and map information; then, the target driving scene to which the vehicle to be controlled belongs at the current moment is determined through the lane identification, lane-level road section data and road signs, so that the specific driving scene of the vehicle to be controlled at the current moment can be determined, thereby refining the decision-making problem in a complex traffic environment; then, according to the target driving scene and lane-level road section data, the planned trajectory of the vehicle to be controlled is determined; finally, the vehicle to be controlled is controlled to travel according to the planned trajectory, so that corresponding response plans can be taken for different driving scenarios, so as to finely plan the driving trajectory of the vehicle, thereby improving the adaptability of the vehicle to be controlled in complex driving scenarios.

[0039] The following describes exemplary applications of electronic devices provided in the embodiments of the present application. The electronic devices provided in the embodiments of the present application can be implemented as various types of terminals such as laptop computers, tablet computers, desktop computers, set-top boxes, smart phones, smart speakers, smart watches, smart TVs, and vehicle-mounted terminals, and can also be implemented as servers. The embodiments of the present application do not impose any restrictions on the specific types of electronic devices.

[0040] The vehicle control method provided in each embodiment of the present application can be executed by an electronic device, wherein the electronic device can be a server or a terminal, that is, the vehicle control method provided in each embodiment of the present application can be executed by a server, can be executed by a terminal, or can be executed through interaction between a server and a terminal.

[0041] The vehicle control method provided in the embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0042] Figure 1 1 is an optional flow chart of a vehicle control method provided in an embodiment of the present application, and the method can be applied to an electronic device. The following will take the electronic device as an example for illustrative description. Figure 1 As shown, the method includes the following steps S101 to S105:

[0043] Step S101, obtaining vehicle collection information of the vehicle to be controlled at the current moment and map information of the current location of the vehicle to be controlled.

[0044] The vehicle to be controlled can be a vehicle with autonomous driving functions. For example, the vehicle to be controlled can be an unmanned delivery robot, or an autonomous driving vehicle with driver assistance, etc. Autonomous driving vehicles can be divided into levels 0 to 5 according to the degree of automation, where level 0 represents no automation and level 5 represents full automation. Full human driving (Level L0): A vehicle that is completely controlled by a human driver. Assisted driving (Level L1): Controlled by a human driver, but some functions of the vehicle can be automated, such as adaptive cruise control and automatic braking in emergencies. Partially autonomous driving (Level L2): The vehicle can automate steering, acceleration and deceleration, but a human driver is required to monitor road conditions and be ready to take over control at any time. Conditional autonomous driving (Level L3): The vehicle is capable of autonomous driving, but human driver intervention is required in certain situations. Highly autonomous driving (Level L4): The vehicle can be fully autonomous in certain specific scenarios without the need for human driver monitoring, such as an autonomous truck on a highway. Fully autonomous driving (Level L5): The vehicle can be fully autonomous regardless of the environment and conditions, without the need for a human driver.

[0045] Vehicle collection information can be collected by sensors installed on the vehicle. For example, image sensors collect image data including color images, depth images, stereo images, etc. Image sensors can be used to identify road signs, traffic lights, pedestrians, vehicles and other obstacles; radar sensors can collect information such as the distance between the vehicle and the obstacle, the vehicle's speed and the vehicle's azimuth; laser radar can collect three-dimensional point cloud data of the surrounding environment, providing accurate distance to obstacles and obstacle shape information; ultrasonic sensors can collect distance information of close-range obstacles, and ultrasonic sensors can be used for parking assistance to detect the distance between the vehicle and surrounding objects; global positioning systems can collect vehicle geographic location information.

[0046] The map information can be obtained from the navigation software used by the vehicle to be controlled. The map configured by the navigation software can be a navigation map, a high-precision map, a map dedicated to autonomous driving, etc.

[0047] Step S102, converting the vehicle collection information and the map information to obtain the lane identification of the current location, the lane-level road section data of the current location and the road sign of the current location.

[0048] The information collected by the vehicle includes image data, point cloud data, and longitude and latitude data. The server cannot process different data structure types in a unified manner, and needs to convert this information into the data structure types required for subsequent scene decisions and behavior decisions. The original high-precision map information usually contains very detailed geographic features and road attribute information, including lane geometry, traffic signs, traffic lights, intersection layouts, etc. In order to realize the path planning of the vehicle in actual driving, it is necessary to convert the high-precision map information to generate a more simplified routing map information. For example, the continuous lane geometry is simplified into straight line segments, and the complex layout of the intersection is simplified into basic connection relationships; key features related to path planning and driving decisions are extracted from the high-precision map, such as lane centerlines, lane boundaries, traffic signs, intersection types, etc.; nodes (such as the start and end points of the intersection) and edges (such as lane segments) in the routing map are constructed.

[0049] By converting the vehicle collection information and map information, the lane marking of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road sign of the current position can be obtained through the converted vehicle collection information and map information. Lane markings can be marks or indications on the road to distinguish different lanes or indicate the driving rules of lanes. Lane markings usually include road markings, arrows, text markings, etc., which are part of the traffic sign system and are used to guide the correct use of roads and ensure traffic order and safety. For example: white dotted lines, white solid lines, yellow solid lines, yellow dotted lines, arrow marks, text markings, etc. Road signs can be signs installed beside or above the road to provide information, instructions, warnings or prohibitions. Road signs can be graphic symbols, text or a combination of the two. For example: road signs, location distance signs, straight signs, left or right turn signs, construction signs ahead, school signs ahead, no-entry signs, speed limit signs, lane restriction signs, yield signs, emergency parking strip signs, pedestrian crossing signs, etc. Lane-level road segment data can be detailed attribute information about each lane on the road stored in the map database. For example: Lane centerline: the centerline coordinates of the lane (usually expressed in latitude and longitude or local coordinate system); Lane boundary line: the coordinates of the left and right boundary lines of the lane; Lane width: the width of the lane (in meters); Lane curvature: the degree of curvature of the lane (such as straight line, curve, curve radius, etc.); Lane slope: the slope information of the lane (such as uphill, downhill, flat road); Lane type: the purpose or type of the lane (such as straight lane, left turn lane, right turn lane, bus lane, emergency lane, etc.); Lane direction: the driving direction of the lane (such as one-way, two-way); Lane number: the unique identifier of the lane (such as "lane 1", "lane 2"); Lane connection relationship: the connection relationship between the current lane and the adjacent lane (such as merge, separate, cross, etc.). By parsing the lane-level road segment data, lane markings and road signs can be obtained.

[0050] Step S103, based on the lane identification, lane-level road section data and road signs, determine the target driving scene to which the vehicle to be controlled belongs at the current moment.

[0051] Target driving scenarios can include free navigation scenarios and multi-lane scenarios.

[0052] A multi-lane scenario refers to a road where the vehicle is currently traveling with two or more parallel lanes. Multi-lane scenarios usually occur in busy traffic areas such as highways, urban main roads, and intersections. Vehicles can travel in different lanes, and different lanes may have different functions, such as straight lanes, left turn lanes, right turn lanes, and bus lanes. In a multi-lane scenario, vehicles may need to frequently change lanes, turn, or enter ramps. Traffic signs and markings are complex.

[0053] Free navigation scenarios may include on-lane scenarios, off-lane scenarios, parking scenarios, and large curvature road section scenarios, among which on-lane scenarios may include parking-start scenarios and ramp-merging-into-main-road scenarios, and off-lane scenarios may include pull-over parking scenarios and main-road-merging-into-ramp scenarios. Detailed description below. The parking-start scenario may refer to a scenario in which a vehicle enters a driving state from a stationary state (such as parking or waiting) and merges into traffic. The parking-start scenario may occur at a traffic light, a stop sign, or a congested road section. The ramp-merging-into-main-road scenario may refer to the process in which a vehicle enters a highway or a main-road of a city expressway from a ramp (usually a low-speed driving or acceleration lane). The pull-over parking scenario may refer to a scenario in which a vehicle decelerates from a driving state and pulls over to park. The pull-over parking scenario may occur at the destination, temporary parking, or in an emergency. The parking scenario may refer to a scenario in which a vehicle completes parking in a parking lot or roadside parking space. Parking scenarios usually require precise vehicle control and environmental perception. The large curvature road section scenario may refer to a scenario in which a vehicle is driving on a road section with a small curve radius. Large curvature road scenes can occur on mountain roads, urban curves, or highway ramps. The vehicle needs to adjust the speed and direction according to the curvature and slope of the curve.

[0054] Based on lane markings, lane-level road section data and road signs, we can obtain information such as the relationship between the vehicle to be controlled and the lane, the purpose of the lane, driving rules, etc. at the current moment, and then determine the target driving scenario to which the vehicle to be controlled belongs.

[0055] Step S104, determining a planned trajectory of the vehicle to be controlled based on the target driving scenario and lane-level road section data.

[0056] The planned trajectory of a vehicle refers to the path and speed distribution that the vehicle is expected to travel in the future. The planned trajectory can include two parts: path (spatial information) and speed (time information). The path can be the path that the vehicle will travel in the future, which can be represented by a series of discrete points (such as longitude and latitude coordinates or points in a local coordinate system). The speed information can be the speed distribution of the vehicle at each path point, including acceleration, deceleration, and constant speed. The speed information needs to take into account traffic rules (such as speed limits), vehicle dynamics (such as acceleration limits), and the surrounding environment (such as the speed of the vehicle in front).

[0057] Step S105, controlling the vehicle to be controlled to travel according to the planned trajectory.

[0058] After determining the planned trajectory of the vehicle to be controlled, the trajectory tracking algorithm can be used to determine the steering angle, throttle opening, driving speed and other information of the controlled vehicle, and then the vehicle is controlled to travel along the planned trajectory according to the above information. For example, the trajectory tracking algorithm outputs a steering angle of 5 degrees and a throttle opening of 30%. The vehicle control system transmits the steering angle to the steering motor and the throttle opening to the throttle motor, so that the vehicle travels along the planned trajectory.

[0059] In an embodiment of the present application, first, the vehicle collection information and map information of the vehicle to be controlled at the current moment are obtained; then, the lane marking, lane-level road section data and road signs are obtained from the vehicle collection information and map information; then, the target driving scene to which the vehicle to be controlled at the current moment belongs is determined through the lane marking, lane-level road section data and road signs, so that the specific driving scene in which the vehicle to be controlled is located at the current moment can be determined, thereby dividing the driving scenes in which the vehicle to be controlled in a complex traffic environment is located; then, according to the target driving scene and the lane-level road section data, the planned trajectory of the vehicle to be controlled is determined; finally, the vehicle to be controlled is controlled to travel according to the planned trajectory, so that corresponding response plans can be adopted for different driving scenarios, thereby finely planning the vehicle's driving trajectory, and improving the adaptability of the vehicle to be controlled in complex driving scenarios.

[0060] Figure 2 is another optional flow chart of the vehicle control method provided in the embodiment of the present application, such as Figure 2 As shown, the method includes the following steps S201 to S208:

[0061] Step S201: The terminal receives an operation input by a user.

[0062] The operation input by the user includes a selection operation or an input operation, wherein the selection operation is used to select a task instruction to be sent to the vehicle to be controlled, or the input operation is used to input a task instruction to be sent to the vehicle to be controlled.

[0063] Step S202: The terminal determines N task instructions for the vehicle to be controlled in response to the operation input by the user.

[0064] In the embodiment of the present application, there are multiple tasks corresponding to the task instructions for the vehicle to be controlled, and different tasks correspond to different task instructions. The user can input corresponding operations for different tasks. The terminal can determine the task instruction corresponding to the operation based on the operation input by the user.

[0065] Step S203: The terminal sends N task instructions for the vehicle to be controlled to the server.

[0066] Here, N is an integer greater than or equal to 1.

[0067] In an embodiment of the present application, the terminal sends N task instructions for the vehicle to be controlled to the server, and the server can control the vehicle to be controlled according to the received N task instructions for the vehicle to be controlled. Of course, in some embodiments, the server can also actively control the vehicle to be controlled, that is, the server can actively control these vehicles to be controlled.

[0068] The vehicle to be controlled can complete the autonomous driving decision through the collaborative working mechanism of the task layer, scene layer and behavior layer. The functions of the task layer, scene layer and behavior layer can be realized by the server. In the initialization stage, the task layer can determine the current high-level task goal according to the input task instruction, such as lane driving task, free navigation task, demonstration task, teaching task, etc. Different task instructions can be set in a targeted manner according to different task objectives. Lane driving task can refer to the behavior of a vehicle driving along a fixed lane on the road according to lane markings and traffic rules. Free navigation task can refer to the behavior of a vehicle autonomously planning a path and driving according to the destination and real-time environmental information without the constraints of a fixed lane. Demonstration task can refer to the demonstration of the functions and performance of the autonomous driving system through a pre-set path. It is usually used for technology display, testing or teaching. Teaching task can refer to the use of manual operation or demonstration to let the autonomous driving system learn driving rules or behaviors in a specific scenario.

[0069] Step S204, the server instructs the vehicle to be controlled to drive in a lane in response to any task instruction, and obtains vehicle collection information of the vehicle to be controlled at the current moment and map information of the current position of the vehicle to be controlled.

[0070] In the embodiment of the present application, when any of the N task instructions for the vehicle to be controlled received by the server instructs the vehicle to be controlled to drive in a lane, the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current position of the vehicle to be controlled are obtained. It should be noted that the "obtaining the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current position of the vehicle to be controlled" in step S204 is the same as the above-mentioned step S101, and the implementation details of the "obtaining the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current position of the vehicle to be controlled" in step S204 are not repeated in the embodiment of the present application.

[0071] Through step 203 to step 204, the mission objectives of the vehicle to be controlled can be finely divided, which is convenient for the vehicle to be controlled to make quick decisions and plans under various mission objectives, thereby improving the adaptability of the vehicle to be controlled in complex driving scenarios.

[0072] Step S205, the server converts the vehicle collection information and the map information to obtain the lane identification of the current location, the lane-level road section data of the current location and the road sign of the current location.

[0073] In some embodiments, see Figure 3 , Figure 3 It is shown that step S205 can be implemented by the following steps S2051 to S2053:

[0074] Step S2051, converting the vehicle collected information to obtain the position information of the current position of the vehicle to be controlled.

[0075] In an embodiment of the present application, the vehicle to be controlled may be equipped with a variety of sensors, including laser radar, camera, millimeter wave radar, ultrasonic radar, accelerometer and gyroscope. Sensors can collect environmental information around the vehicle to be controlled from different angles and in different ways. Various sensors convert the collected environmental information into raw data (i.e., vehicle collection information). For example, laser radar obtains distance information by measuring light reflection time and generates point cloud data; camera captures optical image; millimeter wave radar detects the speed and direction of objects, etc. The raw data usually needs to be converted into a usable data format through a series of signal processing steps. First, the raw data can be preprocessed: including denoising, calibration, etc. to improve data quality. For example, remove outliers in point cloud data, or correct lens distortion of the camera. Then, useful features can be extracted from the raw data. For image data, it can include edge detection, color recognition, shape recognition, etc. For radar data, it can include speed and direction estimation. Then, the data collected by different sensors can be integrated to obtain a more comprehensive environmental model. Finally, the processed data is encapsulated into a data structure suitable for the autonomous driving decision system of the vehicle to be controlled, such as the vehicle's position, speed, acceleration, the position and motion state of surrounding objects, etc. The data structure types required by the autonomous driving decision system may include: perception data: objects describing the vehicle's surrounding environment, including position, shape, speed and other attributes; vehicle state data: including the vehicle's own speed, acceleration, position, orientation and driving status; position data: such as the latitude and longitude of the vehicle's current location, etc.

[0076] Step S2052: convert the map information to obtain lane identification, lane-level road section data and road signs.

[0077] In the embodiment of the present application, the map information is converted into a data structure suitable for the automatic driving decision system of the vehicle to be controlled, which may be lane-level road section data. For example, the map information comes from a high-precision map, which is usually stored in a specific format (such as the open driving scene description format (OpenDRIVE), the navigation data standard (Navigation Data Standard, NDS), the lane network description format (Lanelet2), etc.). A map parsing tool or library (such as the OpenDRIVE parser) can be used to read the map file and extract the geometric information and semantic information in the map. Lane line information is usually stored in a geometric form in a high-precision map, and the map information is converted to obtain lane-level road section data. The lane-level road section data is parsed to obtain information such as lane markings and road signs. For example, the center line and boundary line (left boundary and right boundary) of the lane line. The geometric coordinates of the lane line (such as a series of points or curves) can be extracted, and the coordinates describe the shape and position of the lane line. The type of lane line (such as solid line, dashed line, double yellow line, etc.), color (such as white, yellow) and width can also be extracted. The connection relationship between lanes can also be extracted, such as how lanes fork, merge or cross. Lane markings and road signs (such as arrows, text, stop lines, etc.) are also stored in high-precision maps in geometric shapes (such as polygons or point sets). The location and shape information of lane markings and road signs can be extracted. The types of lane markings and road signs (such as straight arrows, left turn arrows, stop lines, crosswalks, etc.) are extracted. The colors and meanings of lane markings and road signs (such as the speed value of speed limit signs, etc.) are extracted. Finally, the extracted lane lines, lane markings and other information are converted into data structures that can be used by the autonomous driving system (such as lightweight text data exchange format (JSON), data serialization protocol (Protobuf), etc.), so that the lane markings, lane-level road section data and road signs corresponding to the data structures that can be used by the autonomous driving system can be obtained. In addition, the coordinates in the high-precision map can be converted to the vehicle coordinate system or the world coordinate system for alignment with the vehicle sensor data.

[0078] Step S2053, the lane marking, lane-level road section data and road sign are respectively associated with the position information to obtain the lane marking of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road sign of the current position.

[0079] In the embodiment of the present application, the lane marking, lane-level road section data and road signs obtained from the map information are associated with the position information of the vehicle to be controlled, and the lane marking of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road sign of the current position can be obtained. The position information of the vehicle to be controlled can be matched with the lane line or the center line of the road in the high-precision map, and the lane marking, lane-level road section data and road sign associated with the lane line can be determined from the high-precision map according to the lane line matched with the position information of the vehicle to be controlled, and then the lane marking of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road sign of the current position can be determined. For example, high-precision maps are usually stored in a global coordinate system (such as the World Geodetic System 1984 (WGS84) longitude and latitude or the Universal Transverse Mercator (UTM) coordinate system). The high-precision map data can be first converted to the vehicle's local coordinate system. The vehicle's local coordinate system is a coordinate system established with the vehicle as the center, which is used to describe the relative position and direction of the vehicle's surrounding environment, and then matched with the lane lines or road center lines in the high-precision map according to the vehicle's position and direction.

[0080] Through steps 2051 to 2053, the vehicle collection information and map information are converted into the data structure type required for subsequent scene decisions and behavior decisions, which improves the availability and operability of the data and provides accurate environmental information for the decision-making and control of the vehicle to be controlled.

[0081] Step S206: The server determines the target driving scenario to which the vehicle to be controlled belongs at the current moment based on the lane identification, lane-level road section data and road signs.

[0082] In some embodiments, the above step S206 can be implemented in the following manner: first, based on the lane identification, lane-level road section data and road signs, determine the inclusion relationship between the current position of the vehicle to be controlled and the target lane corresponding to the lane identification; then, in response to the inclusion relationship that the current position is not located in the target lane, determine that the target driving scene is a free navigation scene; in response to the inclusion relationship that the current position is located in the target lane, and the distance between the current position and the lane end of the target lane is greater than a preset distance threshold, determine that the target driving scene is a multi-lane scene; in response to the inclusion relationship that the current position is located in the target lane, and the distance between the current position and the lane end of the target lane is less than or equal to the preset distance threshold, determine that the target driving scene is a free navigation scene; in response to the inclusion relationship that the current position is located in the target lane, and the curvature value of the target lane is greater than the preset curvature threshold, determine that the target driving scene is a free navigation scene.

[0083] In the embodiment of the present application, the position of the target lane and the distance from the current position of the vehicle to be controlled to the end of the lane of the target lane can be determined according to the lane-level road section data, the type of the lane can be determined according to the lane marking, the current traffic driving rules and lane direction information can be determined according to the road signs, and the current position of the vehicle to be controlled and the position of the target lane can be determined. The inclusion relationship between the current position of the vehicle to be controlled and the target lane corresponding to the lane marking can be determined. When the current position of the vehicle to be controlled is not located in the target lane, it indicates that the vehicle to be controlled may be located on the ramp or on the roadside, and the target driving scene can be determined to be the upper lane scene in the free navigation scene. When the current position of the vehicle to be controlled is located in the target lane, and the distance from the current position to the end of the lane of the target lane is greater than the preset distance threshold, it indicates that the vehicle to be controlled may be driving in the target lane, and the target driving scene can be determined to be a multi-lane scene. It should be noted that the preset distance threshold can be set according to actual conditions, and the embodiment of the present application does not limit this. When the current position of the vehicle to be controlled is in the target lane, and the distance between the current position and the lane end of the target lane is less than or equal to the preset distance threshold, it indicates that the vehicle to be controlled may need to stop or enter the ramp, and the target driving scene can be determined to be the down lane scene in the free navigation scene. When the current position of the vehicle to be controlled is in the target lane, and the curvature value of the target lane is greater than the preset curvature threshold, it indicates that the lane where the vehicle to be controlled is currently located is a curve, and the target driving scene can be determined to be a large curvature road section scene in the free navigation scene. It should be noted that the preset curvature threshold can be set according to actual conditions, and the embodiments of the present application are not limited to this.

[0084] The following example illustrates that the vehicle to be controlled can be regarded as a rectangle. When the four vertices of the rectangle are all within the lane, the current position of the vehicle to be controlled can be regarded as being within the target lane; or the vehicle to be controlled can be regarded as a point, and then a lane range with a width smaller than the width of the target lane can be preset with the lane centerline of the target lane. When the point is within the preset lane range, the current position of the vehicle to be controlled can be regarded as being within the target lane. When the current position of the vehicle to be controlled is not within the target lane, if the current position of the vehicle to be controlled is within other lanes, and the lane sign of the current lane has a "straight arrow" or "turn arrow", it indicates that the vehicle to be controlled may be located on the ramp; or if the road sign has a "merge lane" sign, a "forced lane change" sign, and a "slow down" sign, it also indicates that the vehicle to be controlled may be located on the ramp. When the current position of the vehicle to be controlled is not within the target lane, if the current position of the vehicle to be controlled is not within other lanes, it indicates that the vehicle to be controlled may be located on the roadside. Assuming that the preset distance threshold is 20 meters, when the current position of the vehicle to be controlled is in the target lane, and the distance from the current position to the end of the lane of the target lane is 50 meters, which is greater than the preset distance threshold, it indicates that the vehicle to be controlled is driving in the target lane, and the target driving scene can be determined to be a multi-lane scene. When the current position of the vehicle to be controlled is in the target lane, and the distance from the current position to the end of the lane of the target lane is 10 meters, which is less than the preset distance threshold, if the lane marking of the current lane shows a "zebra crossing", it indicates that the vehicle to be controlled may need to stop; or if the road sign shows an "exit" sign, a "forced lane change" sign, and a "slow down" sign, it also indicates that the vehicle to be controlled may need to enter the ramp. When the current position of the vehicle to be controlled is in the target lane, the curvature value of the target lane can be calculated by the lane-level section data. Specifically, the geometric data of the lane-level section of the target lane can be obtained first, including the starting coordinates, end coordinates and length of the section, and then the straight-line distance of the section is calculated according to the starting coordinates and end coordinates of the section. Finally, the straight-line distance is divided by the length of the section to obtain the curvature value. If the curvature value of the target lane is greater than the preset curvature threshold, it indicates that the lane where the vehicle to be controlled is currently located is a curve. If the road signs show "slow down" signs or "watch out for curves" signs, it can also indicate that the lane where the vehicle to be controlled is currently located is a curve.

[0085] Through the above method, the relationship between the vehicle to be controlled and the lane can be determined through lane markings, lane-level road section data and road signs, and then the driving scene of the vehicle to be controlled can be divided in detail based on the relationship between the vehicle to be controlled and the lane, and the specific driving scene of the vehicle to be controlled at the current moment can be determined, so as to divide the driving scene of the vehicle to be controlled in a complex traffic environment, and then the vehicle's driving trajectory can be planned in a refined manner later, thereby improving the adaptability of the vehicle to be controlled in complex driving scenes.

[0086] Step S207: The server determines the planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data.

[0087] In some embodiments, when the target driving scene includes a multi-lane scene, see Figure 4 , Figure 4 It is shown that step S207 can be implemented by the following steps S2071A to S2073A:

[0088] Step S2071A, determining the first behavior node corresponding to the vehicle to be controlled at the current moment from the preset behavior tree corresponding to the multi-lane scenario.

[0089] In the embodiment of the present application, the preset behavior tree is a pre-set behavior tree, and the behavior tree is a graphical tool for describing the decision-making process. The preset behavior tree can be used to describe the decision-making process of the vehicle to be controlled to travel on the road. The preset behavior tree may include a repeat node (repeatedly execute the sub-node a specified number of times, or until a certain sub-node condition is met), a selection node (execute the sub-nodes according to the arrangement order of the selection nodes in the preset behavior tree (for example, from left to right) until one of the sub-nodes returns a success state, then the selection node returns success; or all sub-nodes return a failure state, then the selection node returns failure), a sequence node (execute the sub-nodes according to the arrangement order of the sequence nodes in the preset behavior tree (for example, from left to right) until one of the sub-nodes returns a failure state, then the sequence node returns a failure state; or all sub-nodes return a success state, then the sequence node will also return success), and a priority selection node (no longer execute each sub-node according to the arrangement order of the selection nodes in the preset behavior tree, but execute according to the priority of the sub-node, and the priority of the sub-node can be customized). See. Figure 5 , Figure 5 is a schematic diagram of a preset behavior tree provided in an embodiment of the present application, and the first behavior node may be Figure 5 When the target driving scenario is a multi-lane scenario, the preset behavior tree corresponding to the multi-lane scenario can be Figure 5The behavior tree shown in the figure can be executed in order from left to right. For example, the selection node 501 can be executed in order from left to right. If the sub-node under the selection node 501 has not been executed at the current moment, the first behavior node corresponding to the vehicle to be controlled at the current moment can be determined to be the sequence node 502; if the sub-node under the selection node 501 has been executed at the current moment, such as a sub-node returns a failure state, the first behavior node corresponding to the vehicle to be controlled at the current moment can be determined to be the next node of the failure node. The first sub-node in the sub-node under the sequence node is usually a judgment node (that is, a rhombus box). If the judgment result is yes, then the next sub-node under the sequence node will be entered. If the judgment result is no, the current sequence node will be exited. For example, the sub-node under the selection node 501 has been executed, and the sequence node 502 returns a failure state, then the first behavior node corresponding to the vehicle to be controlled at the current moment can be determined to be the sequence node 503. For another example, there are two sequence nodes under the road keeping selection node (execution order from left to right). If one of the two sequence nodes is successfully executed, the road keeping selection node is successfully executed (that is, the execution of road keeping can be realized in two scenarios: one is that there is a car in front, and the vehicle to be controlled follows the vehicle in front; the other is that there is no car in front, and the vehicle to be controlled drives forward at a constant speed). It should be noted that the child nodes under the priority selection node are no longer executed in the order of arrangement in the behavior tree, but in the order of priority of each child node. The priority of each child node is determined according to the actual situation, and the embodiment of the present application does not limit this.

[0090] Step S2072A, based on the lane-level road section data and the first behavior node, determine the behavior type of the vehicle to be controlled at the next moment.

[0091] In the embodiment of the present application, each behavior node corresponds to a different behavior type, such as parking, forced lane change, single lane, overtaking on the left, overtaking on the right, decelerating, cruising speed, etc.

[0092] In some embodiments, the above step S2072A can be implemented in the following manner: first, determine whether the lane-level road section data meets the judgment condition of the first behavior node to obtain the node judgment result; then, based on the node judgment result, determine the second behavior node where the vehicle to be controlled is located at the current moment from the preset behavior tree; finally, determine the behavior type of the vehicle to be controlled at the next moment based on the second behavior node.

[0093] In the embodiment of the present application, each behavior node corresponds to a different judgment condition, and a judgment can be made based on the lane-level road section data to determine whether the lane-level road section data satisfies the judgment condition of the first behavior node, and a node judgment result is obtained. If the node judgment result indicates that the lane-level road section data satisfies the judgment condition of the first behavior node, the behavior type of the first behavior node is determined as the behavior type of the vehicle to be controlled at the next moment. If the node judgment result indicates that the lane-level road section data does not meet the judgment condition of the first behavior node, the next behavior node of the first behavior node in the preset behavior tree is determined as the second behavior node where the vehicle to be controlled is located at the current moment. Then the behavior type corresponding to the second behavior node is determined as the behavior type of the vehicle to be controlled at the next moment.

[0094] An example is given below. For example, the first behavior node is a sequential node 502. It is determined whether the vehicle to be controlled has reached the destination according to the lane-level road section data. If the node judgment result indicates that the vehicle to be controlled has reached the destination, the behavior type of the vehicle to be controlled at the next moment is parking. If the node judgment result indicates that the vehicle to be controlled has not reached the destination, the second behavior node is determined to be a sequential node 503 according to the execution order of the preset behavior tree. The sequential node 503 is executed. It is determined whether the vehicle to be controlled has received a forced lane change instruction and the target lane is safe according to the lane-level road section data. If the node judgment result indicates that the vehicle to be controlled has received a forced lane change instruction and the target lane is safe, the behavior type of the vehicle to be controlled at the next moment is pre-forced lane change and forced lane change.

[0095] Through the above method, the behavior nodes can be judged one by one, and the driving scenes of the vehicles to be controlled in complex traffic environments can be divided, so as to improve the accuracy of the planned trajectory of the subsequent vehicles to be controlled and improve the adaptability of the vehicles to be controlled in complex driving scenes.

[0096] Step S2073A, determining the planned trajectory of the vehicle to be controlled based on the behavior type at the next moment.

[0097] In the embodiment of the present application, according to the behavior type at the next moment, the trajectory planning module can be used to determine the boundary of the area where the vehicle to be controlled can be driven, and then determine the planned trajectory of the vehicle to be controlled. According to the lane-level road section data, the lane line position, shoulder, road boundary and other information are obtained, and the trajectory that the vehicle can drive is determined in combination with the behavior type of the vehicle to be controlled at the next moment.

[0098] Through steps 2071A to 2073A, the decision logic of the vehicle to be controlled is made clearer and easier to understand by the preset behavior tree. The preset behavior tree can be quickly executed in a real-time environment, which can improve the real-time performance of the decision of the vehicle to be controlled.

[0099] In some embodiments, when the target driving scenario includes a free navigation scenario, see Figure 6 , Figure 6 It is shown that step S207 can be implemented by the following steps S2071B to S2073B:

[0100] Step S2071B, determining the free navigation section of the vehicle to be controlled in the free navigation scenario and the end point of the free navigation section according to the lane identification, lane-level section data and road signs.

[0101] In an embodiment of the present application, according to the current position of the vehicle to be controlled, as well as the lane marking, lane-level section data and road signs, the free navigation section and the end point of the free navigation section in the free navigation scenario can be determined. For example, when the current position of the vehicle to be controlled is in the lane, and the distance from the current position to the end point of the lane of the target lane is 10 meters less than the preset distance threshold, the lane marking appears as a "zebra crossing" and the road sign appears as a "slow down" sign. At this time, the target driving scene is the parking scene in the free navigation scenario. If there is no obstacle between the vehicle to be controlled and the "zebra crossing", it can be determined that the free navigation section of the vehicle to be controlled in the free navigation scenario is the section between the vehicle to be controlled and the "zebra crossing" in the lane where the vehicle to be controlled is currently located, and the end point of the free navigation section is the "zebra crossing" corresponding to the lane where the vehicle to be controlled is currently located. If there is an obstacle between the vehicle to be controlled and the "zebra crossing", the free navigation section of the vehicle to be controlled in the free navigation scenario can be determined as the lane through which the vehicle to be controlled passes to bypass the obstacle, as well as the section between the vehicle to be controlled and the "zebra crossing" in the lane where the vehicle to be controlled is currently located after the lane change, and the end point of the free navigation section is the "zebra crossing" corresponding to the lane after the lane change.

[0102] Step S2072B, determining the section planning trajectory of the vehicle to be controlled in the free navigation section according to the end point of the free navigation section.

[0103] In the embodiment of the present application, the route planning trajectory of the vehicle to be controlled in the free navigation section can be determined according to the current position of the vehicle to be controlled and the end point of the free navigation section. For example, if there is no obstacle between the vehicle to be controlled and the "zebra crossing", the route planning trajectory of the vehicle to be controlled in the free navigation section can be the section between the current position of the vehicle to be controlled and the "zebra crossing"; if there is an obstacle between the vehicle to be controlled and the "zebra crossing", the route planning trajectory of the vehicle to be controlled in the free navigation section can be the trajectory of the vehicle to be controlled bypassing the obstacle, and the section between the vehicle to be controlled and the "zebra crossing" in the lane where the vehicle to be controlled is currently located after changing lanes.

[0104] Step S2073B, adding the planned trajectory of the road section to the planned trajectory of the vehicle to be controlled.

[0105] In the embodiment of the present application, the route planning trajectory can be added to the original planning trajectory of the vehicle to be controlled. For example, the original planning trajectory of the vehicle to be controlled is to drive straight from location A to location B at a constant speed, but there is a "zebra crossing" and a traffic light between the two locations. If the route planning trajectory is to stop at the "zebra crossing", the planning trajectory of stopping at the "zebra crossing" is added to the original planning trajectory.

[0106] Through steps 2071B to 2073B, different driving trajectories can be planned for different driving scenarios, thereby finely planning the vehicle's driving trajectory and improving the adaptability of the vehicle to be controlled in complex driving scenarios.

[0107] Step S208: the server controls the vehicle to be controlled to travel according to the planned trajectory.

[0108] It should be noted that step S208 is the same as the above-mentioned step S105, and the implementation details of step S208 are not repeated in this embodiment of the application.

[0109] In the embodiment of the present application, first, the terminal receives the operation input by the user, and then the terminal sends the task instruction corresponding to the operation to the server. When the task instruction instructs the vehicle to be controlled to drive in the lane, the server will obtain the vehicle collection information and map information of the vehicle to be controlled. In this way, the task objectives of the vehicle to be controlled can be finely divided, which is convenient for the vehicle to be controlled to make quick decisions and plans under multiple task objectives. Then, the vehicle collection information and map information are converted respectively to obtain the lane identification of the current position of the vehicle to be controlled, the lane-level road section data of the current position, and the road sign of the current position. Then, the specific driving scene of the vehicle to be controlled at the current moment is determined, and the complex traffic environment can be finely divided. Then, trajectory planning is performed for different driving scenarios, which improves the accuracy of the planned trajectory of the vehicle to be controlled, improves the adaptability of the vehicle to be controlled in complex driving scenarios, and finally, controls the vehicle to be controlled to drive according to the planned trajectory.

[0110] The following is an explanation of an exemplary application of an embodiment of the present application in a practical application scenario.

[0111] See also Figure 7 , Figure 7 Schematic diagram of the implementation process of the hierarchical decision-making method for autonomous driving provided in the embodiment of the present application. The hierarchical decision-making framework for autonomous driving mainly includes: task decision 701, scenario decision 702 and behavior decision 703. First, in the initialization stage of the autonomous driving vehicle, the task layer (i.e., task decision) determines the current high-level task goal according to the received task instructions, such as lane driving task, free navigation task, demonstration task, teaching task, etc.

[0112] Then, the original data (i.e., vehicle acquisition information and map information) such as sensors, positioning, and high-precision maps of the autonomous driving vehicle (i.e., the vehicle to be controlled) are received, and the original data is converted into the data structure type required for subsequent scene decisions and behavior decisions. For example, the original high-precision map information is converted into routing map information to obtain the lane-level section data structure (i.e., lane-level section data), forced lane change signs (i.e., road signs), upper and lower lane signs (i.e., lane markings), and vehicle (i.e., vehicle to be controlled) and lane belonging relationship detection information.

[0113] Then, after the original data processing is completed, the scene type (i.e., driving scene) to which the vehicle to be controlled belongs at the current moment is further confirmed based on the upper and lower lane signs, vehicle status, and vehicle lane information. Different scene types will use different types of planners (multi-lane navigation scenes (i.e., multi-lane scenes) use lane planners; upper lane scenes (i.e., parking), lower lane scenes (i.e., parking on the side of the road), parking scenes, and scenes with large curvature use free navigation planners). The state machine switching process for confirming the scene type is as follows: Figure 8 As shown, Figure 8It is a schematic diagram of scene switching provided by the embodiment of the present application. Taking a complete road driving condition as an example, the initial scene type of the autonomous driving vehicle is set to "idle". When the task end point is issued and the global path planning is completed, the scene type is switched to the "stop and start" scene type. Under the "stop and start" scene type, if the vehicle position is completely in the lane, the "multi-lane navigation" scene type can be entered. When the task end point is reached, the scene type is switched to "pull over". When the pull over process is completed, it enters "idle". When the vehicle is driving on a multi-lane road section (for example, a three-lane scene, a road is composed of several sections of a certain length (lane)), if the vehicle position has just entered a large curvature section (the upstream of the decision will judge whether this section is a large curvature section based on the comparison result of the average curvature or maximum curvature of the section with the preset curvature threshold), then switch from the "multi-lane navigation" scene to the "large curvature section" scene; then when the vehicle finishes driving on the large curvature section, it will switch from the "large curvature section" scene to the "multi-lane navigation" scene. For example, in a U-turn scenario, the vehicle will go through the process of "multi-lane navigation" --> "high curvature road section" --> "multi-lane navigation". There are two situations in which the "pullover parking" scenario switches to the "idle" scenario, one is the end of free navigation, and the other is that the task end point is in the lane. The first: the vehicle's pullover parking process ends and enters the idle state; the second: if the task end point is in the lane, there will be no pullover parking process, but during the state switching process, one frame of the "pullover parking" scene can be retained. That is, if the vehicle's task end point is in the lane, the vehicle will go through the process of "multi-lane navigation" --> "pullover parking" --> "idle", but the "pullover parking" scene has only one frame instead of one process. When the vehicle is in the lane and is in the lane, the vehicle itself is in the lane, so the "stop and start" also has only one frame instead of one process. "The vehicle runs out of the global planning result" means that the vehicle runs out of the planned path set (it can be understood that when using map software for navigation, the map software will plan out paths represented by red or green. These paths represented by red or green are the planned paths, and the planned paths are at the road segment level). If the vehicle runs out of the red or green lane for some reason, the vehicle will immediately enter the standby (IDLE) state, and the upstream decision-making will recalculate a red and green road segment set, but before the road segment set is recalculated, the vehicle has already run out of the lane set, so the vehicle will enter the "idle" state. If the upstream decision-making calculates a new lane set (red and green road segment set) in a short time, then the vehicle will go through the process of "idle"-->"stop and start" (only exists for one frame)-->"multi-lane navigation".

[0114] When using the free navigation planner for the upper and lower lanes, first determine the free navigation endpoint (i.e., the end point of the free navigation section) (if the car is on the lane, then the scene state machine switches to "idle" -> "stop and start" -> "multi-lane navigation". "Stop and start" will only exist for one frame, and then enter "multi-lane navigation". If the car is not on the lane, or the car is parked outside the lane on the side of the road, the free navigation endpoint can be confirmed. During the process of getting on the lane, when the vehicle enters the range of the lane (for example, the vehicle is regarded as a rectangle, and all four points of the rectangle have entered the lane), then enter "multi-lane navigation" from "stop and start"). Taking the free navigation process of the upper lane as an example, the vehicle's driving target point is determined based on the relationship between the current vehicle position and the center point of the lane. Assume that the current position of the vehicle is (x v ,y v ), and project the position onto the center line of the nearest lane. The center line of the lane can be represented as a set of points C = {(x i ,y i )|i=1,2,…,n}, where each point (x i ,y i ) represents the discrete coordinate points on the lane centerline. The projection point (x p ,y p ) is the point on the lane centerline closest to the vehicle's current position. The coordinates of the projection point can be determined by minimizing the Euclidean distance, see formula (1):

[0115]

[0116] Among them, (x v ,y v ) is the current position coordinate point of the vehicle, (x i ,y i ) are discrete coordinate points on the lane centerline, and argmin is the minimization operation.

[0117] After obtaining the coordinates of the projection point, offset several lane center points along the lane center line to obtain the target point (the offset value can be customized according to different scenarios or road conditions).

[0118] After determining the specific scenario type, the specific decision intention is output. Taking the multi-lane scenario as an example, the vehicle behavior decision can be executed using the behavior tree method, such as Figure 5 As shown in the figure. The behavior tree of the multi-lane scenario executes nodes such as judging the task end point, forced lane change, active lane change, and road keeping in sequence. When the requirements of the corresponding node are met, the node is entered to perform the corresponding action. Among them, if all the child nodes of the selection node fail to execute, failure is returned; if all the child nodes of the sequence node are executed successfully, success is returned.

[0119] The priority of left and right lane changes in multi-lane navigation requires projecting the obstacle onto the center line of the lane ahead and calculating the lateral position of the obstacle relative to the vehicle itself to determine whether the vehicle is changing lanes to the left or right.

[0120] First, the positions of the obstacle and the vehicle are converted from the Cartesian coordinate system (x, y) to the Frenet coordinate system, which contains the longitudinal coordinate s and the lateral offset l. obs ,l obs ) can be expressed by formula (2) and formula (3):

[0121]

[0122] Among them, (x obs ,y obs ) is the coordinate point of the obstacle in the Cartesian coordinate system, (x i ,y i ) are discrete coordinate points on the lane centerline in the Cartesian coordinate system, and argmin is a minimization operation.

[0123]

[0124] Among them, (x obs ,y obs ) is the coordinate point of the obstacle in the Cartesian coordinate system, (x i ,y i ) are discrete coordinate points on the lane centerline in the Cartesian coordinate system.

[0125] Frenet coordinates of the vehicle (s veh ,l veh ) can be expressed by formula (4) and formula (5):

[0126]

[0127] Among them, (x veh ,y veh ) is the coordinate point of the vehicle in the Cartesian coordinate system, (x i ,y i ) are discrete coordinate points on the lane centerline in the Cartesian coordinate system, and argmin is a minimization operation.

[0128]

[0129] Among them, (x veh ,y veh ) is the coordinate point of the vehicle in the Cartesian coordinate system, (x i ,y i) are discrete coordinate points on the lane centerline in the Cartesian coordinate system.

[0130] The sign of the lateral offset l is determined by the lateral judgment formula (6):

[0131] sign(l)=cos(θ i )·(yy i )-sin(θ i )·(xx i ) (6)

[0132] Where (x i ,y i ,θ i ) represents the discrete coordinate point on the lane centerline and the orientation angle of the point in the Cartesian coordinate system. The positive or negative value of sign(l) can indicate the direction of the point (x, y) relative to the lane center. When sign(l)>0, it indicates that the point (x, y) is on the left side of the lane center; when sign(l)<0, it indicates that the point (x, y) is on the right side of the lane center.

[0133] For the nearest obstacle that meets the longitudinal condition (the obstacle is located in front of the vehicle), determine the lateral offset l of the obstacle obs Lateral offset from vehicle obs If l obs ≤l veh , the obstacle is on the right side of the vehicle, and changing lanes to the left has a higher priority; otherwise l obs > evh , the obstacle is on the left side of the vehicle, and changing lanes to the right has a higher priority.

[0134] Define the priority judgment formula (7):

[0135]

[0136] Among them, P left =1 means that changing lane to the left has a higher priority.

[0137] Finally, the vehicle's specific behavioral intentions (forced lane change, obstacle avoidance within the lane, left lane change, right lane change, road keeping, deceleration and parking, and parking at the end of the mission) are output to the downstream trajectory planning module to determine the boundaries of the drivable area.

[0138] In addition, the embodiment of the present application also provides an autonomous driving hierarchical decision system to support the actual implementation of the above method. The specific architecture is as follows: Fig. 9As shown, it includes the following modules: an environment perception unit 901, used to obtain external environment information; a data processing unit 902, used to process raw data (perception, positioning, map); a multi-layer decision unit 903, used to receive upstream data to output decision intentions; a trajectory planning unit 904, used to output a reference trajectory according to the decision intention.

[0139] The embodiments of the present application can efficiently deal with decision-making problems in different complex traffic environments by constructing a multi-layer decision-making framework including a task layer, a scenario layer, and a behavior layer, realize adaptive task switching and refined behavior decision-making, and ultimately improve the adaptability of autonomous driving vehicles in complex scenarios. The embodiments of the present application also reduce the complexity of the decision-making system by dividing the decision into three levels: tasks, scenarios, and behaviors, and independently designing and coordinating the execution between each layer. The modular design makes the decision algorithm easy to upgrade and maintain, and is also convenient for rapid deployment and expansion in different application scenarios. In addition, the embodiments of the present application can effectively reduce the delay from decision-making to planning through the refined execution of the behavior layer, thereby ensuring the stability of the autonomous driving vehicle under driving conditions.

[0140] It is understandable that in the embodiments of the present application, related data such as location information are involved. When the embodiments of the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards.

[0141] Based on the vehicle control method described in the above embodiment, Fig.10 A structural block diagram of a vehicle control device 100 provided in an embodiment of the present application is shown. The vehicle control device may be a device in an electronic device (e.g., a server). The vehicle control device may be implemented in software, which may be software in the form of programs and plug-ins, etc., including the following software modules: an information acquisition module 101, an information conversion module 102, a scene determination module 103, a trajectory determination module 104, and a control module 105. These modules are logical, and therefore may be arbitrarily combined or further split according to the functions implemented.

[0142] Among them, the information acquisition module 101 is used to obtain the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current position of the vehicle to be controlled; the information conversion module 102 is used to convert the vehicle collection information and the map information to obtain the lane identification of the current position, the lane-level road section data of the current position and the road sign of the current position; the scene determination module 103 is used to determine the target driving scene to which the vehicle to be controlled belongs at the current moment based on the lane identification, the lane-level road section data and the road sign; the trajectory determination module 104 is used to determine the planned trajectory of the vehicle to be controlled based on the target driving scene and the lane-level road section data; the control module 105 is used to control the vehicle to be controlled to travel according to the planned trajectory.

[0143] In some embodiments, the target driving scenario includes a multi-lane scenario; the trajectory determination module 104 is further used to determine the first behavior node corresponding to the vehicle to be controlled at the current moment from the preset behavior tree corresponding to the multi-lane scenario; based on the lane-level road section data and the first behavior node, determine the behavior type of the vehicle to be controlled at the next moment; based on the behavior type at the next moment, determine the planned trajectory of the vehicle to be controlled.

[0144] In some embodiments, the trajectory determination module 104 is also used to determine whether the lane-level road section data meets the judgment condition of the first behavior node to obtain a node judgment result; based on the node judgment result, determine the second behavior node where the vehicle to be controlled is located at the current moment from a preset behavior tree; and determine the behavior type of the vehicle to be controlled at the next moment based on the second behavior node.

[0145] In some embodiments, the target driving scenario includes a free navigation scenario; the trajectory determination module 104 is further used to determine the free navigation segment of the vehicle to be controlled in the free navigation scenario and the end point of the free navigation segment according to lane markings, lane-level segment data and road signs; determine the segment planning trajectory of the vehicle to be controlled in the free navigation segment according to the end point of the free navigation segment; and add the segment planning trajectory to the planned trajectory of the vehicle to be controlled.

[0146] In some embodiments, the scene determination module 103 is also used to determine the inclusion relationship between the current position of the vehicle to be controlled and the target lane corresponding to the lane identification according to the lane identification, lane-level road section data and road signs; in response to the inclusion relationship that the current position is not located in the target lane, the target driving scene is determined to be a free navigation scene; in response to the inclusion relationship that the current position is located in the target lane, and the distance between the current position and the lane end of the target lane is greater than a preset distance threshold, the target driving scene is determined to be a multi-lane scene; in response to the inclusion relationship that the current position is located in the target lane, and the distance between the current position and the lane end of the target lane is less than or equal to the preset distance threshold, the target driving scene is determined to be a free navigation scene; in response to the inclusion relationship that the current position is located in the target lane, and the curvature value of the target lane is greater than the preset curvature threshold, the target driving scene is determined to be a free navigation scene.

[0147] In some embodiments, the information conversion module 102 is also used to convert the vehicle collected information to obtain the position information of the current position of the vehicle to be controlled; to convert the map information to obtain lane markings, lane-level road section data and road signs; to associate the lane markings, lane-level road section data and road signs with the position information respectively, to obtain the lane markings of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road signs of the current position.

[0148] In some embodiments, the information acquisition module 101 is also used to receive N task instructions for the vehicle to be controlled, where N is an integer greater than or equal to 1; in response to any task instruction instructing the vehicle to be controlled to drive in a lane, the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current location of the vehicle to be controlled are obtained.

[0149] It should be noted that the description of the device of the embodiment of the present application is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment, so it is not repeated. For technical details not disclosed in the embodiment of the device, please refer to the description of the method embodiment of the present application for understanding.

[0150] The present application also provides an electronic device, see Fig.11 , Fig.11 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Fig.11 As shown, the electronic device 130 includes: at least one processor 131 ( Fig.11 Only one is shown in the figure), a memory 132, and executable instructions 133 stored in the memory 132 and executable on at least one processor 131, when the processor 131 executes the executable instructions 133, the steps in any of the above-mentioned vehicle control method embodiments are implemented.

[0151] The electronic device may include but is not limited to a processor 131 and a memory 132. Those skilled in the art may understand that: Fig.11 This is merely an example of the electronic device 130 and does not constitute a limitation on the electronic device 130 . The electronic device 130 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0152] The processor 131 may be a central processing unit (CPU), or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPG), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0153] In some embodiments, the memory 132 may be an internal storage unit of the electronic device 130, such as a hard disk or memory of the electronic device 130. In other embodiments, the memory 132 may also be an external storage device of the electronic device 130, such as a plug-in hard disk, a smart memory card (SMC, Smart Media Card), a secure digital (SD, Secure Digital) card, a flash card, etc. equipped on the electronic device 130. Further, the memory 132 may also include both an internal storage unit of the electronic device 130 and an external storage device. The memory 132 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as program codes of a computer program. The memory 132 may also be used to temporarily store data that has been output or is to be output.

[0154] The embodiment of the present application provides a computer program product, which includes a computer program or a computer executable instruction, and the computer program or the computer executable instruction is stored in a computer readable storage medium. The processor of the electronic device reads the computer executable instruction from the computer readable storage medium, and the processor executes the computer executable instruction, so that the electronic device executes the vehicle control method described in the embodiment of the present application.

[0155] The present application embodiment provides a computer-readable storage medium, in which computer-executable instructions or computer programs are stored. When the computer-executable instructions or computer programs are executed by a processor, the processor will be caused to execute the vehicle control method provided by the present application embodiment, for example, Figure 1 A vehicle control method is shown.

[0156] In some embodiments, the computer-readable storage medium may be a memory such as RAM, ROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or may be various devices including one or any combination of the above memories.

[0157] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0158] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0159] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0160] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. A vehicle control method, characterized in that: The method comprises: Obtaining vehicle collection information of the vehicle to be controlled at the current moment and map information of the current location of the vehicle to be controlled; Performing information conversion on the vehicle collection information and the map information to obtain the lane identification of the current location, the lane-level road section data of the current location, and the road sign of the current location; Determining a target driving scenario to which the vehicle to be controlled belongs at a current moment based on the lane identification, the lane-level road section data and the road sign; Determining a planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data; Control the vehicle to be controlled to travel according to the planned trajectory.

2. The method according to claim 1, characterized in that The target driving scenario includes a multi-lane scenario; and determining the planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data includes: Determining, from a preset behavior tree corresponding to the multi-lane scenario, a first behavior node corresponding to the vehicle to be controlled at the current moment; Determining a behavior type of the to-be-controlled vehicle at a next moment based on the lane-level road section data and the first behavior node; Based on the behavior type at the next moment, a planned trajectory of the vehicle to be controlled is determined.

3. The method according to claim 2, characterized in that The determining, based on the lane-level road section data and the first behavior node, a behavior type of the vehicle to be controlled at a next moment includes: Determine whether the lane-level road section data meets the judgment condition of the first behavior node, and obtain a node judgment result; Based on the node judgment result, determining the second behavior node where the vehicle to be controlled is located at the current moment from the preset behavior tree; A behavior type of the vehicle to be controlled at a next moment is determined based on the second behavior node.

4. The method according to claim 1, characterized in that: The target driving scenario includes a free navigation scenario; and determining the planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data includes: Determine a free navigation section of the vehicle to be controlled in the free navigation scenario and an end point of the free navigation section according to the lane identification, the lane-level section data and the road sign; Determining a planned trajectory of the vehicle to be controlled on the free navigation section according to an end point of the free navigation section; The planned trajectory of the road section is added to the planned trajectory of the vehicle to be controlled.

5. The method according to any one of claims 1 to 4, characterized in that: The determining, based on the lane identification, the lane-level road section data and the road sign, a target driving scene to which the vehicle to be controlled belongs at the current moment includes: Determining, according to the lane identification, the lane-level road section data and the road sign, an inclusion relationship between the current position of the vehicle to be controlled and the target lane corresponding to the lane identification; In response to the inclusion relationship being that the current position is not located in the target lane, determining that the target driving scenario is a free navigation scenario; In response to the inclusion relationship being that the current position is located in the target lane, and the distance between the current position and a lane end point of the target lane is greater than a preset distance threshold, determining that the target driving scene is a multi-lane scene; In response to the inclusion relationship being that the current position is located in the target lane, and the distance between the current position and a lane end point of the target lane is less than or equal to the preset distance threshold, determining that the target driving scenario is the free navigation scenario; In response to the inclusion relationship being that the current position is located in the target lane and the curvature value of the target lane is greater than a preset curvature threshold, it is determined that the target driving scenario is the free navigation scenario.

6. The method according to any one of claims 1 to 4, characterized in that: The converting the vehicle collected information and the map information to obtain the lane identification of the current location, the lane-level road section data of the current location and the road sign of the current location includes: Performing information conversion on the vehicle collected information to obtain the position information of the current position of the vehicle to be controlled; Performing information conversion on the map information to obtain lane identification, lane-level road section data and road signs; The lane identification, the lane-level road section data and the road sign are respectively associated with the position information to obtain the lane identification of the current position of the vehicle to be controlled, the lane-level road section data of the current position and the road sign of the current position.

7. The method according to any one of claims 1 to 4, characterized in that: The obtaining of the vehicle collection information of the vehicle to be controlled at the current moment and the map information of the current location of the vehicle to be controlled includes: Receiving N task instructions for a vehicle to be controlled, where N is an integer greater than or equal to 1; In response to any of the task instructions instructing the vehicle to be controlled to drive in a lane, vehicle collection information of the vehicle to be controlled at a current moment and map information of the current position of the vehicle to be controlled are obtained.

8. A vehicle control device, characterized in that: The device comprises: An information acquisition module is used to acquire vehicle collection information of the vehicle to be controlled at the current moment and map information of the current location of the vehicle to be controlled; An information conversion module, used to convert the vehicle collected information and the map information to obtain the lane identification of the current location, the lane-level road section data of the current location and the road sign of the current location; A scene determination module, used to determine the target driving scene to which the vehicle to be controlled belongs at the current moment based on the lane identification, the lane-level road section data and the road sign; A trajectory determination module, used to determine the planned trajectory of the vehicle to be controlled based on the target driving scenario and the lane-level road section data; A control module is used to control the vehicle to be controlled to travel according to the planned trajectory.

9. An electronic device, characterized in that: The electronic device comprises: A memory for storing computer executable instructions or computer programs; A processor is used to implement the vehicle control method according to any one of claims 1 to 7 when executing the computer executable instructions or computer programs stored in the memory.

10. A computer-readable storage medium storing computer-executable instructions or a computer program, characterized in that: When the computer executable instructions or computer program are executed by a processor, the vehicle control method according to any one of claims 1 to 7 is implemented.

Images