Method and apparatus for controlling automated driving of vehicle

Abstract

The invention provides a method for controlling the automatic driving of a vehicle, and the method comprises the following steps: S1, determining at least one most possible path of the vehicle in real time under the condition that the vehicle is driven in an automatic driving mode and a navigation function is not enabled; s2, based on the at least one most probable path of the vehicle, evaluating the traffic restriction risk of vehicle driving; and S3, if the evaluation result shows that the traffic control risk exists in the vehicle driving, controlling the vehicle to exit the automatic driving mode and / or triggering the takeover control. The present application also proposes an apparatus for controlling automatic driving of a vehicle, a vehicle and a computer program product. According to the invention, the traffic control risk assessment is deeply embedded into the automatic driving decision chain, so that the problem of risk missed judgment caused by decoupling of the automatic driving function module and the navigation module is effectively avoided, and the safety and compliance of automatic driving are improved.

Description

Technical Field

[0001] The present application relates to a method for controlling the autonomous driving of a vehicle, and also relates to a device for controlling the autonomous driving of a vehicle, a vehicle, and a computer program product. Background Art

[0002] With the continuous development of autonomous driving technology, although vehicles equipped with advanced driver assistance systems can effectively reduce the burden on drivers, they still face challenges when dealing with traffic management measures in different regions. Currently, the assessment and prompt of traffic restriction risks mainly rely on navigation systems, but the effectiveness of this solution is limited. For those autonomous driving systems that do not require the activation of navigation functions (active route guidance) as a prerequisite, it is easy to lead to missed judgments of traffic restriction risks and lack effective avoidance measures.

[0003] Therefore, the existing autonomous driving solutions still need to be further optimized. Summary of the Invention

[0004] The present application relates to a method for controlling the autonomous driving of a vehicle, a device for controlling the autonomous driving of a vehicle, a vehicle, and a computer program product, so as to solve at least some problems in the prior art.

[0005] According to a first aspect of the present application, there is provided a method for controlling the autonomous driving of a vehicle, the method including the following steps:

[0006] Step S1, when the vehicle is traveling in autonomous driving mode and the navigation function is not enabled, determine at least one most likely path of the vehicle in real time;

[0007] Step S2, based on at least one most likely path of the vehicle, evaluate the traffic restriction risk of the vehicle; and

[0008] Step S3, if the evaluation result indicates that there is a traffic restriction risk during the vehicle's travel, control the vehicle to exit the autonomous driving mode and / or trigger takeover control.

[0009] The present application particularly includes the following technical concepts: The present application breaks through the limitations of the risk prompt of traditional navigation functions. Through the evaluation of traffic restriction risks based on the most likely driving path of the vehicle, the depth of traffic restriction risk assessment is embedded in the autonomous driving decision-making chain, effectively avoiding the problem of missed risk judgments caused by the decoupling of the autonomous driving function module and the navigation module. In addition, with the help of the system-level degradation strategy, it provides a strong guarantee for the safe and compliant driving of the vehicle and the standardized operation of autonomous driving.

[0010] In an exemplary embodiment, the step S2 includes: determining a virtual destination for the vehicle's travel based on at least one most likely path; and determining that there is a traffic restriction risk for the vehicle's travel at least when the virtual destination is within a traffic restricted area.

[0011] Thus, by setting a virtual destination, it is possible to more accurately determine the traffic restriction risk when the navigation function is not enabled.

[0012] In an exemplary embodiment, the virtual destination is determined in the following manner: determining the end point of at least one most likely path as the virtual destination; or, determining the extended point of the end point of at least one most likely path as the virtual destination, wherein the position of the extended point is determined according to the vehicle's real-time speed, the road type where the vehicle is located, the road environmental conditions, and / or the vehicle's historical travel data.

[0013] Thus, by reasonably setting the virtual destination, the vehicle can accurately predict potential traffic restriction risks in advance, thereby providing the user with sufficient time and space to take avoidance measures. This effectively avoids the dilemma of being unable to avoid violations due to the late exit of the autonomous driving mode.

[0014] In an exemplary embodiment, the step S2 includes: performing a spatial intersection check on at least one most likely path and a traffic restricted area, and determining that there is a traffic restriction risk for the vehicle's travel at least when there is a spatial intersection between the at least one most likely path and the traffic restricted area.

[0015] Thus, by checking the spatial intersection of the path and the restricted area, potential risks can be more comprehensively identified to avoid missed judgments.

[0016] In an exemplary embodiment, the method further includes: in the case of multiple available most likely paths, selecting the most likely path for which no traffic restriction risk is evaluated and applying it to the vehicle's autonomous driving; and / or, determining that there is a traffic restriction risk for the vehicle's travel in step S2 only when it is evaluated that there is a traffic restriction risk based on all available most likely paths.

[0017] Thus, in a multi-path scenario, a risk-free path can be preferentially selected or the risk can be comprehensively determined, optimizing the autonomous driving decision-making and reducing unnecessary manual intervention.

[0018] In an exemplary embodiment, the method further includes: when the vehicle is traveling on a highway, at a position where there is at least one ramp exit remaining from the highway end point, performing a traffic restriction risk assessment in advance based on the highway end point. If it is determined based on the highway end point that there is a traffic restriction risk for the vehicle's travel, then regardless of the evaluation result based on at least one most likely path in step S2, directly controlling the vehicle to exit the autonomous driving mode and / or triggering takeover control.

[0019] Therefore, in the highway scenario, the risk is forced to be evaluated in advance based on the destination to avoid the situation where the vehicle cannot avoid traffic restrictions anyway due to missing the exit.

[0020] In an exemplary embodiment, in step S1, the at least one most likely path is determined based on the driving state of the vehicle and the road network information, and the at least one most likely path includes the predicted driving trajectory and / or the predicted road to be entered by the vehicle within a future period of time.

[0021] Therefore, through multi-dimensional data integration, it is ensured that the path prediction fits the actual vehicle state and road environment, improving the decision-making rationality of the system in complex traffic environments.

[0022] In an exemplary embodiment, the method further includes: dynamically updating the at least one most likely path over time, and evaluating the traffic restriction risk of the vehicle driving after each update or after multiple updates.

[0023] Therefore, through dynamic path update and continuous risk assessment, the limitation of traditional navigation relying only on initial static assessment is solved, ensuring the real-time and accuracy of risk warning.

[0024] In an exemplary embodiment, the method further includes: taking the entry situation of the vehicle's license plate information as a condition for activating the vehicle's autonomous driving mode, where only when the vehicle's license plate information is entered, is it allowed to activate the vehicle's autonomous driving mode; and / or, obtaining the vehicle's license plate information and additionally evaluating the traffic restriction risk of the vehicle driving based on the license plate information in step S2.

[0025] By combining the license plate information with the traffic restriction rules, the determination mechanism of traffic restriction risk can be further improved. At the same time, blocking the autonomous driving function of vehicles that have not completed the registration of license plate information can forcibly ensure the compliance of autonomous driving behaviors from the source, thereby effectively improving the overall safety and acceptance of the autonomous driving function.

[0026] In an exemplary embodiment, the traffic restriction risk includes: tail number restriction risk, non-new energy vehicle restriction risk, emission standard restriction risk, foreign license plate restriction risk, truck restriction risk, non-priority vehicle restriction risk, and / or traffic control restriction risk.

[0027] Therefore, by covering different traffic restriction types, multi-dimensional violation risk prevention and control can be achieved, improving the safety and compliance of autonomous driving.

[0028] In one embodiment, in step S2, at least one most likely path is provided to the navigation module, and the navigation module, by means of its pre-set traffic restriction judgment function, performs a traffic restriction risk assessment based on the at least one most likely path.

[0029] Thereby, the reuse of existing computing resources is achieved, and redundant algorithm deployment is avoided. The overall computing load of the system is reduced through modular interaction design.

[0030] In an exemplary embodiment, the method further includes: if the navigation function of the vehicle is enabled, obtaining a navigation route to the user-set navigation destination; evaluating the traffic restriction risk of the vehicle's travel based on the navigation route; if it is evaluated based on the navigation route that there is a traffic restriction risk in the vehicle's travel, and when there is no avoidance navigation route and no avoidance operation of the user is detected, controlling the vehicle to exit the autonomous driving mode and / or triggering takeover control.

[0031] Thereby, even if the navigation function is enabled, in the autonomous driving mode, the traffic restriction risk prompt is no longer limited to the passive method of "only prompt without intervention". By introducing an autonomous driving exit or takeover mechanism, not only can the driver's stronger attention be attracted, but also the driving safety and compliance are improved.

[0032] In an exemplary embodiment, in step S3: once the evaluation result in step S2 indicates that there is a traffic restriction risk in the vehicle's travel, immediately control the vehicle to exit the autonomous driving mode and / or trigger takeover control; or, in the case that the evaluation result in step S2 indicates that there is a traffic restriction risk in the vehicle's travel, and when the distance from the vehicle to the geographical boundary of the traffic restriction area is less than a threshold, control the vehicle to exit the autonomous driving mode and / or trigger takeover control.

[0033] Thereby, it is possible to flexibly decide whether to immediately exit or delay exiting the autonomous driving mode according to the urgency of the risk, so as to ensure driving safety while taking into account the continuity of the autonomous driving function.

[0034] According to a second aspect of the present application, there is provided an apparatus for controlling the autonomous driving of a vehicle, the apparatus including a processor and a memory, the memory storing computer program instructions, and when the computer program instructions are executed by the processor, the processor is capable of executing the method according to the first aspect of the present application.

[0035] According to a third aspect of the present application, there is provided a vehicle, the vehicle including the apparatus according to the second aspect of the present application.

[0036] According to a fourth aspect of the present application, there is provided a computer program product comprising computer program instructions, wherein when the computer program instructions are executed by a processor, the processor is enabled to execute the method according to the first aspect of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Hereinafter, the present application will be described in more detail by referring to the accompanying drawings, and the principles, features and advantages of the present application can be better understood. The accompanying drawings include:

[0038] Figure 1 A block diagram of a vehicle showing an exemplary embodiment according to the present application, the vehicle including means for controlling autonomous driving;

[0039] Figure 2 A flowchart of a method for controlling autonomous driving of a vehicle showing an exemplary embodiment according to the present application;

[0040] Figure 3 A flowchart of a method for controlling autonomous driving of a vehicle showing another exemplary embodiment according to the present application;

[0041] Figure 4 A flowchart of a method for controlling autonomous driving of a vehicle showing another exemplary embodiment according to the present application;

[0042] Figure 5 A schematic diagram showing the evaluation of traffic restriction risks based on the end points of at least one most likely path in an exemplary application scenario;

[0043] Figure 6a and 6b A schematic diagram showing the evaluation of traffic restriction risks based on the end point extension points of at least one most likely path in an exemplary application scenario;

[0044] Figure 7 A schematic diagram showing the evaluation of traffic restriction risks in an exemplary application scenario when there are multiple most likely paths available;

[0045] Figure 8 A schematic diagram showing the evaluation of traffic restriction risks in a highway scenario; and

[0046] Figure 9a and 9b A schematic diagram showing the display of a takeover prompt on a display unit of a vehicle. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] To make the technical problems, technical solutions, and beneficial technical effects to be solved by this application more clearly understood, the following will further elaborate on this application in conjunction with the accompanying drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and not to limit the protection scope of this application.

[0048] Figure 1 A block diagram of a vehicle 1 according to an exemplary embodiment of the present application is shown. The vehicle 1 includes a device 10 for controlling autonomous driving.

[0049] The vehicle 1, for example, has an autonomous driving function at L3 level or above. Figure 1 In the illustrated embodiment, the device 10 for controlling autonomous driving can be connected to an autonomous driving function module 20 and a navigation module 30, for example. The autonomous driving function module 20 executes autonomous driving tasks based on the path planning information provided by the device 10. The navigation module 30 is used to plan a navigation route according to the navigation destination input or specified by the user.

[0050] When the navigation function of the vehicle 1 is not enabled, the device 10 can calculate the most likely path based on the real-time state of the vehicle and environmental perception data; transmit the most likely path to the autonomous driving function module 20 in real time to achieve autonomous driving guidance without a preset navigation destination. The navigation module 30 can use its built-in traffic restriction algorithm to evaluate the traffic restriction risks of the most likely path and the conventional navigation route, and transmit the evaluation results back to the device 10 for its use in the exit control of the autonomous driving function. In addition, the device 10 can also obtain traffic restriction rules from the navigation module 30 or external data sources and undertake the risk assessment work by itself.

[0051] The device 10 includes a processor and a memory (not specifically shown for simplicity). Computer program instructions are stored in the memory, and these instructions can be stored in computer-readable storage media such as hard disks, memories, and flash cards. The processor can be a central processing unit (CPU), a microcontroller unit (MCU), a graphics processing unit (GPU), a neural network processing unit (NPU), a digital signal processor (DSP), or other general-purpose processors. When the processor executes the computer program instructions in the memory, it can implement a method for controlling the autonomous driving of the vehicle.

[0052] To determine the drivable path and assess the traffic restriction risk, the device 10 can also be connected to multiple sensors 11, 12, 13 in a wired or wireless manner through an in-vehicle network (such as Ethernet, in-vehicle Ethernet, CAN bus, FlexRay, MOST, HSVL, etc.). With the on-vehicle camera 11 (such as a front-view camera), it is possible to capture the images of the vehicle's surrounding environment in real time, thereby identifying the geometric structure information and lane marking information of the road. With the map module and GNSS module 12, road map data and the vehicle's real-time position information can be obtained. With the vehicle's status sensor 13, vehicle status information such as speed, acceleration, distance to specific environmental objects, etc. can be collected.

[0053] The device 10 can also be connected to the communication interface 14 to receive external information through the communication network, such as the traffic restriction rule information issued by the road supervision platform, or road condition information and weather information, etc.

[0054] The device 10 can also be connected to various input / output modules 15 to evaluate the traffic restriction risk or restrict the activation of the autonomous driving mode in combination with the license plate information and / or vehicle model information input by the user. When necessary, the device 10 can also prompt the user of the traffic restriction risk through the input / output module 15, or output takeover prompts, etc. The input / output module 15 includes, for example, a dashboard display, a central control screen, a head-up display (HUD) and / or an augmented reality head-up display (AR-HUD), etc., and can also include a steering wheel vibration device, a seat vibration device, a speaker and / or an optical output device, etc.

[0055] It should be understood that Figure 1 the number and type of various sensors or actuators connected to the device 10 shown are only examples, and this application is not intended to limit this. In practical applications, other types or numbers of sensors and actuators can be adopted in the vehicle 1 to meet specific requirements and conditions.

[0056] It should also be understood that without departing from the core concept of this application, the connection relationship and function division of each module can be appropriately adjusted according to actual needs, and these adjustments all fall within the protection scope of this application. For example, the device 10 is not limited to Figure 1 the connection relationship shown, and it can also be at least partially constructed integrally with the autonomous driving function module 20 and / or the navigation module 30.

[0057] Figure 2 The flowchart of a method for controlling the autonomous driving of a vehicle according to an exemplary embodiment of this application is shown. This method includes steps S1 to S3, and can be executed, for example, by Figure 1 the device 10 for controlling autonomous driving shown.

[0058] In step S1, when the vehicle is traveling in the autonomous driving mode and the navigation function is not enabled, at least one most likely path of the vehicle is determined in real time.

[0059] "The navigation function is not enabled" can cover various situations, for example: the user does not actively activate the navigation function; for another example, although the user initially activates the navigation function and sets the navigation destination, the vehicle has successfully reached the navigation destination, so the navigation function returns to the state where it is not effectively enabled; for another example, the user connects the mobile phone through CarPlay and sets the navigation destination in the mobile phone navigation App, but the in-vehicle navigation function fails to be effectively enabled due to reasons such as synchronization failure.

[0060] In this case, the autonomous driving mode of the vehicle can be called autonomous driving along the road or autonomous driving without a navigation destination. In this mode, the vehicle does not rely on a pre-planned fixed route, but dynamically makes decisions by real-time sensing of the surrounding environment (such as roads, obstacles, traffic signs, etc.) and plans the most likely path in real time.

[0061] In the simplest case, the "most likely path" refers to the path (MPP, Most Probable Path) that the vehicle is most likely to continue to travel in the traditional sense. The calculation of the most likely path is based on the driving state (motion and position state) of the vehicle and road information. Such path planning focuses on the range of several hundred meters to several kilometers in front of the vehicle, and the path length and update interval are determined by the autonomous driving system framework, mostly preset values. Its core purpose is to provide real-time road information to assist in driving decisions, rather than planning a complete route from the starting point to the end point.

[0062] In another case, at least one most likely path can also be fine-tuned based on the traffic restriction risk screening requirements on the basis of the traditional MPP, so it can also be understood as a "customized most likely path". For example, it can be dynamically updated as the vehicle travels, and the update interval is generally from dozens of milliseconds to several seconds. At the same time, the update interval and path length can be dynamically adjusted according to the vehicle's real-time speed, position and road type: the faster the vehicle speed and the higher the road grade, the shorter the update interval and the longer the path length. The path length can be measured either by distance (such as kilometers) or by time (such as seconds). Exemplarily, when the path length is measured by time, it can be set to 500 seconds on the highway, 300 seconds on the urban expressway, 200 seconds on the urban arterial road, and 100 seconds on the urban branch road.

[0063] The data structure and calculation method of the most likely path can be determined by the framework of vehicle navigation applications, advanced driver assistance systems (ADAS), or autonomous driving systems (ADS). For example, the most likely path may cover the predicted driving trajectory and / or the predicted road to be entered by the vehicle within a certain period in the future. There are various calculation methods, including but not limited to heuristic algorithms (such as A* algorithm, Dijkstra algorithm), machine learning (such as deep learning, reinforcement learning), or probabilistic graphical models (such as Markov decision process, Bayesian network), etc. When there are multiple candidate paths, the path with the highest probability can be selected through scoring as the most likely path.

[0064] Under the ADASIS (Advanced Driver Assistance System Interface Specification) framework, the most likely path is defined as the path that the vehicle is most likely to continue driving on (MPP, Most Probable Path), which usually exists as the "main path", and some alternative paths (sub - path) are also provided. In some cases, the most likely path may not only include the main path but also may additionally include alternative paths.

[0065] Under the ADASIS framework, the data structure of the most likely path includes the following elements:

[0066] ● Path ID: Uniquely identifies a path and is used to represent the current driving path of the ego - vehicle.

[0067] ● Offset: The current position relative to the starting point of this path under this path.

[0068] ● Speed: The current speed of the ego - vehicle

[0069] ● Relative Heading: The heading angle of the ego - vehicle relative to this path

[0070] · Confidence: Represents the confidence level of the current position.

[0071] In other cases, the most likely path can adopt other data formats, such as a sequence of discrete points, polynomial trajectory, grid map, probability distribution, or Frenet coordinate system, etc. This application does not impose special restrictions on the data structure and calculation method of the most likely path. If the most likely path is represented in the form of trajectory positions, it generally needs to be mapped to a digital map for spatial matching with traffic restricted areas; if it is represented by road elements (such as path ID, lane ID, path connection relationship, etc.), this kind of mapping is generally not required.

[0072] In step S2, based on at least one most likely path of the vehicle, evaluate the traffic restriction risk of the vehicle's driving.

[0073] Traffic restriction risks include, for example: license plate tail number restriction risks, non-new energy vehicle restriction risks, emission standard restriction risks, out-of-town license plate restriction risks, traffic control restriction risks, truck restriction risks (e.g., trucks are prohibited from passing through specific areas or roads), non-priority vehicle restriction risks (such as priority vehicles like ambulances and police cars are not affected by restrictions, while other general vehicles face restriction risks) and / or bus lane restriction risks. The evaluation process can be completed with the help of preset restriction rules, which are usually stored in the navigation module 30 of vehicle 1 in the form of structured data.

[0074] For example, Figure 1 After the device 10 calculates the real-time most likely path, it can return it to the navigation module 30, and the navigation module 30 uses its built-in algorithm (or with slight modification) to evaluate traffic restriction risks. Of course, the device 10 can also obtain restriction rules from the navigation module 30 or externally (such as from government traffic management departments, map data providers or real-time traffic information platforms through the Internet) and complete the traffic restriction risk assessment by itself. The restriction rules usually include the following content:

[0075] - Restriction time periods (such as morning and evening rush hours; specific dates; specific days of the week, etc.);

[0076] - Restriction areas (all ground roads within the area; a certain type / a certain road within the area);

[0077] - Vehicle restriction conditions (such as license plate type, license plate tail number, vehicle type (such as diesel vehicles, trucks), emission standards, etc.);

[0078] - Additional conditions (such as temporary restrictions, special area restrictions).

[0079] Traffic restriction risks can be evaluated by spatially matching the most likely path with traffic restriction areas. The traffic restriction area refers to the geographical area where driving restrictions are imposed on specific vehicles as stipulated in the restriction rules, such as the license plate tail number restriction area, the area restricting specific license plate types (such as non-new energy vehicles, out-of-town license plates), and special areas such as control areas.

[0080] When evaluating traffic restriction risks, for example, it can first be determined whether the vehicle license plate information belongs to the restricted driving category in the restriction rules (such as the license plate tail number for restriction, non-new energy vehicle license plate, etc.). If so, the most likely path of the vehicle can be further spatially matched with the corresponding restriction area. If the restriction area is a traffic control area (regardless of license plate type), the most likely path can be directly matched with the area without verifying the license plate information.

[0081] This assessment is not done once, because the most likely path changes dynamically over time. For example, the risk assessment can be performed immediately after each path update, or it can be performed once after every N (e.g., 5) path updates, so the risk assessment is performed multiple times during the vehicle's travel.

[0082] In one embodiment, a virtual destination of the vehicle can be determined based on at least one most likely path, and then at least when the virtual destination is located in a traffic restriction area, it is determined that the vehicle is at risk of traffic restriction. For example, the end point of the most likely path can be directly set as the virtual destination. For another example, the extension point of the end point of the most likely path can also be used as the virtual destination.

[0083] In one embodiment, the method for determining the virtual destination can be selected according to the real-time speed of the vehicle and / or the driving scene in which it is located. For example, when the vehicle speed is fast, the virtual destination can be determined as the extension point of the most likely path, thereby reserving sufficient evasive reaction time for the driver in advance; when the vehicle speed is slow, the virtual destination can be determined as the end point of the most likely path to avoid inaccuracies caused by premature risk assessment. In complex urban road sections with many uncertainties, the end point of the most likely path can be directly used as the virtual destination to avoid misjudgment; when driving on highways with less traffic, the end point extension point can be used as the virtual destination to reserve reaction time in advance. If the end point of the most likely path is a clear road node (such as an intersection, toll station, parking lot entrance, etc.), the end point can be directly used as the virtual destination; if the end point is the middle point of the road, the end point can be appropriately extended to the nearest road node as the virtual destination.

[0084] In one embodiment, the location of the terminal extension point can also be determined based on the real-time speed of the vehicle, the type of road the vehicle is on, road environmental conditions (such as traffic conditions, weather, road construction, etc.) and / or the historical driving data of the vehicle. For example, on a highway, if the vehicle speed is fast and the road environment is good, the terminal extension point can be set farther away to avoid potential risks in advance; while in congested urban roads, the vehicle speed is slow and the uncertainty is high, and the extension point distance can be appropriately shortened to avoid misjudgment due to premature evaluation. At the same time, combined with the historical driving data of the vehicle, a more likely virtual destination can also be inferred. For example, if the driver often turns right at a certain intersection to go to a residential area, the navigation module can use the entrance of the residential area as a virtual destination.

[0085] In one embodiment, a spatial intersection check may be performed on at least one most likely path and the restricted traffic area, and at least when at least one most likely path and the restricted traffic area have a spatial intersection, it is determined that the vehicle is at risk of being restricted. For example, a geo-fence algorithm (such as a ray intersection method or a polygon intersection algorithm) may be used to determine whether the most likely path passes through the restricted traffic area.

[0086] When there are multiple most likely paths, for example, the most likely path without traffic restriction risk can be preferentially selected and applied to vehicle autonomous driving. Specifically, multiple candidate paths (such as straight, left-turn, right-turn paths, etc.) can be generated based on the current driving state of the vehicle, road network information, and environmental perception data, denoted as MPP1, MPP2, ..., MPPn, and confidence levels are assigned to each path according to different cost metrics. After traversing all paths, the set of paths without traffic restriction risk {Safe_MPPs} can be filtered out. If there are multiple risk-free paths (|Safe_MPPs| ≥ 2), the path with the highest confidence level can be selected as the most likely path and applied to vehicle autonomous driving; if there is a unique risk-free path (|Safe_MPPs| = 1), it can be directly applied to vehicle autonomous driving; if the number of risk-free paths is zero (|Safe_MPPs| = 0), it can be determined that there is a traffic restriction risk currently.

[0087] In step S3, if the evaluation result indicates that there is a traffic restriction risk during vehicle driving, the vehicle is controlled to exit the autonomous driving mode and / or trigger takeover control.

[0088] The ways to trigger takeover control include: sending a takeover request prompt to the driver through the in-vehicle infotainment system. The prompt methods are diverse. For example, text or graphic information is displayed on the in-vehicle infotainment system interface, the steering wheel lights flash, the steering wheel vibrates, or a sound prompt is issued, etc. If the driver responds to the takeover request, the autonomous driving mode can be exited and switched to the manual driving mode. If the driver does not respond to the takeover request within the specified time, the prompt level will be automatically increased, such as increasing the prompt volume, flashing frequency, or displaying more prominent warning information to further attract the driver's attention. If the driver still does not respond for a long time, the vehicle will execute the minimum risk strategy, such as automatically decelerating and safely stopping to ensure the safety of the vehicle and passengers.

[0089] In one embodiment, once the evaluation result in step S2 indicates that there is a traffic restriction risk during vehicle driving, the vehicle is immediately controlled to exit the autonomous driving mode and trigger takeover control. In another embodiment, after the traffic restriction risk is evaluated in step S2, the downgrading measure is not triggered immediately. Instead, when the distance of the vehicle from the geographical boundary of the traffic restriction area is less than the set threshold, the vehicle is controlled to exit the autonomous driving mode and / or trigger takeover control.

[0090] For example, in scenarios with adverse weather (such as heavy rain, heavy snow, thick fog) or a single road structure (such as the main road of an expressway, a one-way road, etc., where there are fewer avoidance and detour options), due to low visibility and limited road choices, the vehicle's perception and decision-making capabilities are restricted. Once the traffic restriction risk is evaluated, the autonomous driving mode can be immediately exited and / or takeover control can be triggered to ensure that the driver takes over the vehicle in a timely manner and avoid violations or dangers. On the contrary, in scenarios with good weather and rich road choices (such as multiple parallel roads and multi-intersection roads in the city), even if the traffic restriction risk is evaluated, the autonomous driving mode can be exited and / or takeover control can be triggered when the vehicle travels to a distance less than a predetermined distance threshold from the boundary of the restricted area to ensure the continuity of the autonomous driving function and enhance the driving experience.

[0091] Figure 3 The flowchart of a method for controlling the autonomous driving of a vehicle according to another exemplary embodiment of the present application is shown. The method includes steps S1, S2', and S3 and optional steps S01 to S04. Among them, the implementation processes of steps S1 and S3 are the same as the corresponding steps in Figure 2 and will not be elaborated here. Only the differences from the Figure 3 method shown in Figure 2 will be emphasized below.

[0092] In optional step S01, when an activation instruction for the vehicle's autonomous driving mode (issued by the user, for example) is received, check whether the license plate information of the vehicle has been entered. This can be done by checking whether there is an entered license plate record locally in the vehicle or by prompting the user through the in-vehicle device or voice interaction to enter the license plate information.

[0093] If the license plate information has not been entered, directly proceed to step S03 to prohibit the activation of the autonomous driving mode. If the license plate information has been entered, proceed from step S01 to step S04 to activate the autonomous driving mode.

[0094] In addition, if it is confirmed that the license plate information has been entered, it is also possible to proceed from step S01 to step S02 to verify the entered license plate information. For example, the license plate number entered by the user can be compared with the license plate number recorded in the user's account in the cloud server for consistency. If the comparison result is inconsistent, proceed to step S03 to prohibit the activation of the autonomous driving mode; if it is consistent, proceed to step S04 to activate the autonomous driving mode.

[0095] After activating the autonomous driving mode in step S04, at least one most likely path can be calculated in subsequent step S1, and then the traffic restriction risk of the vehicle's travel can be evaluated based on the entered license plate information and at least one most likely path in step S2'.

[0096] For example, the current date and the last digit of the license plate can be combined first to determine whether the vehicle is on a date with license plate tail number restriction. If so, the spatial comparison between at least one most likely route and the traffic restriction area is combined to determine whether there is a traffic restriction risk. In addition, the license plate type can be determined according to the license plate information (such as whether it is a new energy vehicle, whether it is an out-of-town license plate, etc.), and whether it is necessary to further combine the most likely route for risk assessment is determined according to the restriction rules. If necessary, the scope of the traffic restriction area based on which the risk assessment is carried out can also be determined according to the license plate information, so as to more accurately carry out the restriction risk assessment.

[0097] In an embodiment not shown, additional conditions of the vehicle can also be obtained, such as the vehicle type (whether it is a truck), the vehicle emission information (whether it is a high-emission vehicle type), and the vehicle size or weight (large vehicles may be restricted from passing due to road height limits, width limits, or weight limits), and the traffic restriction risk of the vehicle's travel is evaluated based on these additional conditions in step S2'.

[0098] Figure 4 The flowchart of a method for controlling the automatic driving of a vehicle according to another exemplary embodiment of the present application is shown. The method includes steps S1 - S3 and optional steps S100 - S700. Among them, the implementation processes of steps S1 - S3 are the same as Figure 2 the corresponding steps in, which will not be elaborated here. Only the following will focus on introducing Figure 4 the differences from Figure 2 the method shown.

[0099] In step S100, when the vehicle is in the automatic driving mode, it is checked whether the navigation function is enabled. If the navigation function is not enabled, it directly enters step S1 to determine the most likely route of the vehicle in real time, and in subsequent steps S2 and S3, it is determined whether there is a traffic restriction risk based on this most likely route, and corresponding countermeasures are taken.

[0100] If it is determined in step S100 that the navigation function is enabled, it enters step S200 to obtain the navigation route from the current location to the user-set navigation destination. Then, in step S300, the traffic restriction risk is evaluated based on this navigation route. If there is no risk, the evaluation ends in step S700, and the vehicle is guided to travel according to the navigation route.

[0101] If the evaluation result shows that there is a traffic restriction risk, it enters step S400 to check whether there is a navigation route that can avoid the risk. If there is an avoidance route, it automatically switches to this route and maintains the automatic driving mode, and then the evaluation ends in step S700.

[0102] If no avoidance route is found in step S400, step S500 is entered to detect whether the user has performed an avoidance operation, such as actively changing the navigation destination, parking, or taking over vehicle control, etc. If there is neither an avoidance route nor a detected user operation, step S600 is entered to control the vehicle to exit the autonomous driving mode and / or trigger takeover control to ensure safe driving.

[0103] It should be noted that although Figure 4 only shows that the autonomous driving mode is exited and / or takeover control is triggered when there is neither an avoidance route nor an active avoidance operation by the user, according to actual requirements, the autonomous driving mode can also be exited immediately or takeover control can be triggered when it is determined in step S400 that there is no avoidance route.

[0104] In addition, although Figure 4 and Figure 3 show different implementation manners respectively, they can also be implemented in combination with each other according to actual requirements.

[0105] Figure 5 shows a schematic diagram for evaluating traffic restriction risks based on the end point of at least one most likely path in an exemplary application scenario.

[0106] In Figure 5 the shown scenario, vehicle 1 travels along road 50 in autonomous driving mode without the navigation function enabled and is about to reach a road intersection including a left branch road 53 and a right branch road 52. As Figure 5 shown, the most likely path 200 calculated in real time for vehicle 1 points to the left branch road 53. With the help of the geographic information system (GIS) geofencing technology and the end point coordinate information, it can be clearly determined that the end point 201 of this most likely path 200 is located within the traffic restriction area 300. Accordingly, it can be determined that there is a traffic restriction risk for the vehicle to travel.

[0107] Figure 6a and 6b shows a schematic diagram for evaluating traffic restriction risks based on the end point extension point of at least one most likely path in an exemplary application scenario.

[0108] In Figure 6a the shown scenario, vehicle 1 also travels along road 50 in autonomous driving mode without the navigation function enabled and is about to reach a road intersection including two branch roads 53 and 52. The most likely path 200 calculated in real time for vehicle 1 points to the left branch road 53, and the end point extension point 202 of this most likely path 200 is further determined. As Figure 5Differently, the end point 201 of the most likely path 200 does not fall within the traffic restricted area 300, but the end point extension 202 is located within the restricted area. In this embodiment, when evaluating the traffic restriction risk, it is not directly based on the spatial comparison between the end point 201 of the most likely path 200 and the traffic restricted area 300, but is evaluated based on the end point extension 202. This can determine the traffic restriction risk earlier before the vehicle 1 actually reaches the intersection, thus providing sufficient time for the driver to take over the vehicle and select a detour path (such as the right fork 52).

[0109] In Figure 6b the scenario shown, different from Figure 6a that, the end point extension 202 of the most likely path 200 extends further forward relative to the end point 201. This may be because the vehicle 1 has a faster driving speed or is on a higher-class road in this scenario. For example, in Figure 6a where the vehicle 1 is in a congested section and drives slowly, the end point extension 202 is relatively close; while in Figure 6b where the vehicle 1 is in an unobstructed section and maintains a high vehicle speed, the end point extension 202 extends further. Therefore, in this embodiment, when evaluating the traffic restriction risk based on the end point extension 202 of the most likely path 200, the risk can be further estimated in advance and the takeover control and / or exiting the autonomous driving mode can be triggered earlier.

[0110] Figure 7 shows a schematic diagram for evaluating the traffic restriction risk in an exemplary application scenario when there are multiple most likely paths available.

[0111] In Figure 7 the scenario shown, the vehicle 1 is driving in autonomous driving mode along the road 50 without enabling the navigation function and is about to reach a road intersection including two forks 53, 52. Different from Figure 5 and Figure 6a , 6b shown scenarios, in this embodiment, two most likely paths 200, 210 are calculated simultaneously: the first most likely path 200 points to the left fork 53, and the second most likely path 210 points to the right fork 52. By respectively performing risk assessment strategies on them (such as respectively checking whether their end points 201, 211 fall within the traffic restricted area 300), it can be determined that there is a traffic restriction risk in driving along the first most likely path 200, while there is no risk in driving along the second most likely path 210. Therefore, the second most likely path 210 can be selected as the autonomous driving guidance path for the vehicle 1.

[0112] Figure 8 shows a schematic diagram for evaluating the traffic restriction risk in a highway scenario.

[0113] In this scenario, vehicle 1 is traveling on highway 80, and the end point 85 of the highway is within the traffic restriction area 300. There are multiple ramp exits 81, 82, and 83 on highway 80, and vehicle 1 can leave highway 80 through these exits 81, 82, and 83.

[0114] Figure 8 Two exemplary position points P1 and P2 are marked. The first position point P1 is between the last ramp exit 81 and the end point 85 of the highway; there are also two ramp exits 81 and 82 between the second position point P2 and the end point 85 of the highway.

[0115] If it is judged that there is a traffic restriction risk based on the most likely path calculated in real time only at the first position point P1, it will be too late. At this time, even if takeover control is triggered or the autonomous driving mode is exited, vehicle 1 will no longer be able to leave highway 80 through the ramp exit and cannot effectively avoid the risk of traffic violations. In addition, if the minimum safety strategy is triggered at the first position point P1 because the driver does not respond to the takeover prompt, vehicle 1 may only be temporarily parked in the middle of the highway or on the emergency deceleration lane and cannot quickly enter the safe parking state.

[0116] On the contrary, if the risk is pre-judged at the second position point P2, the driver still has enough time and opportunity to leave highway 80 through ramp exit 81 or 82, so as to choose other routes to avoid the traffic restriction or find a safe parking position in time.

[0117] Therefore, for example, it can be stipulated that at a position where there is at least one ramp exit from the end point 85 of the highway, a traffic restriction risk assessment is carried out in advance based on the end point 85 of the highway. If it is judged based on the end point 85 that there is a traffic restriction risk for vehicle travel, regardless of the evaluation result based on the most likely path, vehicle 1 will be directly controlled to exit the autonomous driving mode and / or trigger takeover control.

[0118] Figure 9a and 9b shows a schematic diagram of displaying a takeover prompt on the display unit 9 of vehicle 1.

[0119] In Figure 9a Vehicle 1 is traveling in the autonomous driving mode and the navigation function is not enabled. Its display unit 9 displays the calculated most likely path 200 in real time. Since the end point 201 of the most likely path 200 is within the traffic restriction area 300, it is determined that there is a traffic restriction risk for vehicle travel. Therefore, in the upper prompt box 91 of the display unit 9, the driver is prompted to take over vehicle control with the steering wheel icon 92.

[0120] In Figure 9bIn this case, vehicle 1 is traveling in an autonomous driving mode and the navigation function is enabled. Three navigation routes 61, 62, and 63 are displayed on the display unit 9, all of which extend from the current position of vehicle 1 to the user-set navigation destination 90. Among them, the main recommended route 61 (boldly displayed) is the best route calculated based on factors such as travel time, fuel consumption, and distance, while the other two are alternative routes 62 and 63. Since the user-set navigation destination 90 is within the traffic restriction area 300 (such as the license plate number restriction area), none of the three navigation routes 61, 62, and 63 can avoid the risk of restriction. If vehicle 1 travels along these navigation routes, it will violate traffic regulations. At this time, a prompt box 91 pops up on the display unit 9, suggesting that the driver select another destination. If the driver does not respond, for example, a takeover prompt icon 92 will be further displayed to require the driver to take over vehicle control.

Claims

1. A method for controlling the automatic driving of a vehicle (1), the method comprising the following steps: Step S1, determining at least one most likely path (200) of the vehicle (1) in real time when the vehicle (1) is traveling in an automatic driving mode and the navigation function is not enabled; Step S2, based on at least one most likely path (200) of the vehicle (1), evaluating the traffic restriction risk of the vehicle; as well as Step S3: If the evaluation result indicates that there is a risk of traffic restriction when the vehicle is driving, the vehicle (1) is controlled to exit the automatic driving mode and / or take over control is triggered.

2. The method according to claim 1, wherein: The step S2 comprises: determining a virtual destination for vehicle travel based on the at least one most likely path (200); At least when the virtual destination is located in the traffic restriction area (300), it is determined that there is a risk of traffic restriction when the vehicle is traveling.

3. The method according to claim 2, wherein: The virtual destination is determined by: determining an end point (201) of at least one most likely path (200) as a virtual destination; or An end extension point (202) of at least one most likely path (200) is determined as a virtual destination, wherein the position of the end extension point (202) is determined based on the real-time speed of the vehicle (1), the type of road on which the vehicle (1) is located, road environmental conditions and / or historical driving data of the vehicle (1).

4. The method according to any one of claims 1 to 3, wherein: The step S2 comprises: performing a spatial cross-check between at least one most likely path (200) and the traffic restricted area (300), At least when the at least one most likely path (200) spatially intersects with the traffic restriction area (300), it is determined that there is a risk of traffic restriction in the vehicle's travel.

5. The method according to any one of claims 1 to 4, wherein: The method further comprises: In the case where there are multiple available most likely paths (200, 210), selecting the most likely path (200) in which no traffic restriction risk is assessed for application to the autonomous driving of the vehicle (1); and / or Only when it is assessed that there is a risk of traffic restriction based on all available most likely paths ( 200 , 210 ), is it determined in step S2 that there is a risk of traffic restriction in the vehicle driving.

6. The method according to any one of claims 1 to 5, wherein: The method further comprises: When the vehicle (1) is traveling on a highway (80), at a position where there is still at least one ramp exit (81) away from the end point (85) of the highway, a traffic restriction risk assessment is performed in advance based on the end point (85) of the highway. If it is determined based on the end point (85) of the highway that there is a risk of traffic restriction in the vehicle's driving, then regardless of the assessment result based on at least one most likely path (200) in step S2, the vehicle (1) is directly controlled to exit the automatic driving mode and / or take over control is triggered.

7. The method according to any one of claims 1 to 6, wherein: In step S1, the at least one most likely path (200) is determined based on the driving state of the vehicle (1) and the road network information, wherein the at least one most likely path (200) includes an estimated driving trajectory and / or an estimated road to be entered by the vehicle (1) within a period of time in the future; and / or The method further comprises: dynamically updating the at least one most likely path (200) as time changes, and evaluating the traffic restriction risk of the vehicle after each update or after multiple updates.

8. The method according to any one of claims 1 to 7, wherein: The method further comprises: Using the entry of the license plate information of the vehicle (1) as a condition for activating the automatic driving mode of the vehicle (1), wherein the automatic driving mode of the vehicle (1) is activated only when the license plate information of the vehicle (1) is entered; and / or The license plate information of the vehicle (1) is obtained, and in step S2, the traffic restriction risk of the vehicle is additionally evaluated based on the license plate information.

9. The method according to any one of claims 1 to 8, wherein: The traffic restriction risks include: license plate restriction risk, non-new energy vehicle restriction risk, emission standard restriction risk, out-of-town license plate restriction risk, truck restriction risk, non-priority vehicle restriction risk and / or traffic control restriction risk.

10. The method according to any one of claims 1 to 9, wherein: In step S2, at least one most likely path (200) is provided to a navigation module, and the navigation module uses its preset traffic restriction judgment function to perform a traffic restriction risk assessment based on the at least one most likely path (200).

11. The method according to any one of claims 1 to 10, wherein: The method further comprises: If the navigation function of the vehicle (1) is enabled, a navigation route to a navigation destination set by a user is obtained; Evaluate the traffic restriction risk of the vehicle based on the navigation route; If it is assessed based on the navigation route that there is a risk of traffic restrictions on the vehicle, in the absence of an evasive navigation route and no evasive operation by the user is detected, the vehicle (1) is controlled to exit the automatic driving mode and / or take over control is triggered.

12. The method according to any one of claims 1 to 11, wherein: In step S3: Once the evaluation result in step S2 indicates that there is a risk of traffic restriction in the vehicle's driving, the vehicle is immediately controlled (1) to exit the automatic driving mode and / or to trigger takeover control; or When the evaluation result in step S2 indicates that there is a risk of traffic restriction when the vehicle is traveling, when the distance between the vehicle (1) and the geographical boundary of the traffic restriction area (300) is less than a threshold, the vehicle (1) is controlled to exit the automatic driving mode and / or trigger takeover control.

13. A device (10) for controlling the automatic driving of a vehicle (1), the device (10) comprising a processor and a memory, the memory storing computer program instructions, and when the computer program instructions are executed by the processor, the processor is capable of executing the method according to any one of claims 1 to 12.

14. A vehicle (1) comprising a device (10) according to claim 13.

15. A computer program product comprising computer program instructions, wherein: The computer program instructions, when executed by a processor, enable the processor to perform a method according to any one of claims 1 to 12.

Images