Method and device for assisting driving control based on camera occlusion detection

Abstract

This invention provides a method and apparatus for assisted driving control based on camera occlusion detection. The method is applied to vehicles equipped with a forward-facing camera and assisted driving functions, and includes: detecting whether there is equivalent occlusion in the image of the vehicle's front captured by the forward-facing camera in assisted driving mode; upon initial detection of occlusion, querying a current safe time threshold from a preset set of safe time thresholds based on the occluded area and the current road feature type, and continuously detecting the equivalent occlusion while maintaining the assisted driving function; if the duration of the equivalent occlusion reaches the current safe time threshold, then downgrading the assisted driving function and / or issuing a takeover request to the driver.

Description

Technical Field

[0001] This application relates to the field of vehicle driver assistance, and more particularly to a driver assistance control method and device based on camera occlusion detection. Background Technology

[0002] With the rapid development of driver assistance technology, more and more vehicles with driver assistance functions are on the road. These vehicles are often equipped with forward-facing cameras to provide visual information for the vehicle to achieve driver assistance functions.

[0003] However, in some situations, the forward-view camera may be physically blocked by foreign objects such as fallen leaves and bird droppings, and may also be affected by optical interference such as the high beams of oncoming vehicles and backlighting from tunnel exits. All of these will weaken (or even eliminate) the visual perception capability of the forward-view camera, posing safety hazards to assisted driving.

[0004] In response, relevant regulations have incorporated the detection and mitigation of forward-view camera occlusion into the Operational Design Domain (ODD) for driver assistance functions. This means that the functional safety requirements of the ODD for vehicle driver assistance functions explicitly define the certification criteria for forward-view camera occlusion mitigation solutions. However, the industry currently lacks effective solutions for detecting and mitigating forward-view camera occlusion. Summary of the Invention

[0005] In view of this, this application provides an assisted driving control method and device based on camera occlusion detection. By detecting the equivalent occlusion of the forward-view camera and determining the current road feature type, the method queries the corresponding current safe time threshold from a preset set of safe time thresholds, and controls the vehicle to downgrade assisted driving and / or request manual takeover after the continuous occlusion duration reaches the threshold. This effectively addresses abnormal driving scenarios where the forward-view camera is occluded, thereby improving the safety of vehicle assisted driving.

[0006] Specifically, this application is achieved through the following technical solution:

[0007] According to a first aspect of this application, a driver assistance control method based on camera occlusion detection is provided, applied to a vehicle equipped with a forward-facing camera and having driver assistance functions, the method comprising:

[0008] When the vehicle's driver assistance function is activated, the system detects whether there is an equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-facing camera. The equivalent occlusion includes physical occlusion and / or optical interference.

[0009] Upon initial detection of equivalent occlusion in the image ahead of the vehicle, the system queries a preset safe time threshold from a set of safe time thresholds based on the occluded area of ​​the image ahead and the current road feature type of the vehicle's current position. While maintaining the assisted driving function, the system continuously detects the equivalent occlusion. Each safe time threshold in the set of safe time thresholds is predetermined based on the vehicle's longest safe detection time under different driving scenarios. The longest safe detection time under any driving scenario characterizes the latest response time required for the vehicle to avoid colliding with the vehicle ahead when the preceding vehicle brakes at a preset deceleration in that driving scenario.

[0010] If the duration of the equivalent occlusion reaches the current safe time threshold, the driver assistance function will be downgraded and / or a takeover request will be issued to the driver.

[0011] According to a second aspect of this application, a driver assistance control device based on camera occlusion detection is provided, applied to a vehicle equipped with a forward-facing camera and having driver assistance functions, the device comprising:

[0012] The camera occlusion detection unit is used to detect whether there is equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-view camera when the vehicle's assisted driving function is activated. The equivalent occlusion includes physical occlusion and / or optical interference.

[0013] The time threshold determination unit is used to, upon first detecting an equivalent occlusion in the image in front of the vehicle, query the current safe time threshold from a preset safe time threshold set based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position, and continuously detect the equivalent occlusion while maintaining the assisted driving function; wherein, each safe time threshold in the safe time threshold set is predetermined based on the longest safe detection time of the vehicle under different driving scenarios, and the longest safe detection time under any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario;

[0014] The driver assistance control unit is configured to downgrade the driver assistance function and / or issue a takeover request to the driver if the duration of the equivalent occlusion reaches the current safe time threshold.

[0015] According to a third aspect of this application, a vehicle is provided, the vehicle being equipped with an image acquisition device and having assisted driving functions, the vehicle comprising:

[0016] Processor, memory for storing processor-executable instructions, and various sensors;

[0017] The processor implements the method described in the first aspect above by running the executable instructions.

[0018] According to a fourth aspect of this application, a computer program and / or instructions are provided, which, when executed by a processor, implement the steps of the method as described in the first aspect above.

[0019] The technical solution provided in this application can include at least the following beneficial effects:

[0020] Through the above embodiments, when the vehicle activates the assisted driving function and detects for the first time an equivalent occlusion (physical obstruction and / or optical interference, etc.) in the image ahead of the vehicle, it can query the current safe time threshold from a preset set of safe time thresholds based on the corresponding obstructed area and the current road feature type. Each safe time threshold in this set is predetermined based on the vehicle's longest safe detection time under different driving scenarios. The longest safe detection time under any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario. After determining the current safe time threshold, the vehicle continues to detect the equivalent occlusion while maintaining the assisted driving function until the duration of the equivalent occlusion reaches the aforementioned current safe time threshold, at which point the assisted driving function is downgraded and / or a takeover request is issued to the driver.

[0021] It is understandable that each safety time threshold in the set of safety time thresholds is predetermined based on the longest safety detection time of the vehicle in different driving scenarios. The longest safety detection time in any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario. Moreover, the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position can comprehensively characterize the current driving scenario of the vehicle. Therefore, the current safety time threshold queried from this set based on the occluded area and the current road feature type can meet the shortest time requirement of not colliding with the vehicle in front in the current driving scenario. In other words, in the current driving scenario, even if there is a vehicle in front braking at a preset deceleration, the vehicle can still maintain its current speed for a period of time. It will only degrade or request manual intervention when the duration of the obstruction reaches the current safe time threshold. This ensures that the vehicle does not rear-end the vehicle in front, thereby minimizing the forced intervention and interference with the driver assistance functions while ensuring driving safety in the abnormal situation of "equivalent obstruction of the forward-view camera". This effectively addresses the above-mentioned abnormal situation and meets the requirements of relevant regulations. Attached Figure Description

[0022] Figure 1This is a schematic diagram of the hardware structure of an assisted driving system shown in an embodiment of this application.

[0023] Figure 2 This is a real vehicle front image with equivalent occlusion shown in an embodiment of this application.

[0024] Figure 3 This is a flowchart illustrating an exemplary embodiment of an assisted driving control method based on camera occlusion detection.

[0025] Figure 4 This is a schematic diagram illustrating the occlusion response effect in a following vehicle scenario, according to an exemplary embodiment.

[0026] Figure 5 This is a schematic diagram illustrating a stable feature scenario and a variable feature scenario according to an exemplary embodiment.

[0027] Figure 6 This is a schematic diagram illustrating the region division of a vehicle's frontal image according to an exemplary embodiment.

[0028] Figure 7 This is an exemplary embodiment illustrating the relationship between deceleration and time in a follow-up deceleration process.

[0029] Figure 8 This is a schematic diagram illustrating the field-of-view relationship between a front-view camera and a surround-view camera, as shown in an exemplary embodiment.

[0030] Figure 9 This is a schematic structural diagram of a vehicle shown in an exemplary embodiment.

[0031] Figure 10 This is a block diagram illustrating an assisted driving control device based on camera occlusion detection, as shown in an exemplary embodiment. Detailed Implementation

[0032] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0033] It should be noted that the steps of the corresponding methods in other embodiments are not necessarily performed in the order shown and described in this application. In some other embodiments, the methods may include more or fewer steps than those described in this application. Furthermore, a single step described in this application may be broken down into multiple steps in other embodiments; and multiple steps described in this application may be combined into a single step in other embodiments.

[0034] With the rapid development of driver assistance technology, more and more vehicles with driver assistance functions are on the road. These vehicles are often equipped with forward-facing cameras to provide visual information for the vehicle to achieve driver assistance functions.

[0035] However, in some situations, the forward-view camera may be physically blocked by foreign objects such as fallen leaves and bird droppings, and may also be affected by optical interference such as the high beams of oncoming vehicles and backlighting from tunnel exits. All of these will weaken (or even eliminate) the visual perception capability of the forward-view camera, posing safety hazards to assisted driving.

[0036] In response, relevant regulations have incorporated the detection and mitigation of forward-view camera occlusion into the Operational Design Domain (ODD) for driver assistance functions. This means that the functional safety requirements of the ODD for vehicle driver assistance functions explicitly define the certification criteria for forward-view camera occlusion mitigation solutions. However, the industry currently lacks effective solutions for detecting and mitigating forward-view camera occlusion.

[0037] In order to meet the above requirements of relevant regulations, this application proposes an assisted driving control scheme based on camera occlusion detection, which will be described in detail below with reference to the accompanying drawings and related embodiments.

[0038] The assisted driving control scheme described in this application mainly involves the assisted driving system detecting whether there is an equivalent occlusion in the image of the front of the vehicle captured by the forward-view camera. Then, if there is an equivalent occlusion, it queries and matches the current safe time threshold of the current driving scenario from the preset safe time threshold set based on the occluded area and the current road feature type. The system then controls the vehicle to downgrade the assisted driving function or request manual takeover when the duration of the occlusion reaches the time threshold.

[0039] Figure 1 This is a schematic diagram of the hardware architecture of a driver assistance system shown in an embodiment of this application. Figure 1As shown, from a hardware perspective, the system can include only vehicle 11, or it can include both vehicle 11 and server 13. From a software perspective, the driver assistance system can include an occlusion detection module and a driver assistance control module. If the driver assistance system includes only vehicle 11, both the occlusion detection module and the driver assistance control module are deployed locally on vehicle 11, such as in the domain controller of vehicle 11. For example, the occlusion detection module and the driver assistance control module can be deployed in the domain controller (such as the intelligent driving domain controller ADCU) of the driver assistance domain of vehicle 11. It is understood that the driver assistance system in this case is an in-vehicle system, and the system can even run offline. If the driver assistance system includes both vehicle 11 and server 13, either the occlusion detection module or the driver assistance control module can be deployed in server 13, while the other module can be deployed locally on the vehicle. For example, the occlusion detection module can be deployed in server 13 and the driver assistance control module can be deployed in the intelligent driving domain controller of the vehicle, which will not be elaborated further.

[0040] In addition to domain control, vehicle 11 may also be equipped with sensors for collecting data on the external environment. This application embodiment does not limit the number or installation location of these sensors. For example, the sensors include a forward-facing camera 115, which can be a monocular camera or the main forward-facing channel in a multi-camera system (such as a binocular camera), where the image perception results provided are used as the primary visual decision-making basis by the driver assistance system. It may also include multiple surround-view cameras (such as cameras 111, 112, and 113), a lidar 114, a millimeter-wave radar, ultrasonic radar, etc., which will not be elaborated further. Of course, the vehicle may also be equipped with at least one in-vehicle sensor (such as an in-vehicle camera, in-vehicle microphone, in-vehicle biosensor, in-vehicle odor sensor, etc.) in suitable locations to achieve corresponding in-vehicle functions. This specification embodiment does not limit the specific model, size, or installation location of various sensors.

[0041] Furthermore, when the driver assistance system includes server 13 or there is a need to interact with server 13, the vehicle can establish a network connection with the remote server 13 via a wireless communication module to exchange data. For example, if server 13 maintains a set of safe time thresholds set by designers / maintenance personnel, this set can be sent to the vehicle via OTA (Over-the-Air) and stored locally so that the current safe time threshold can be queried from the set when there is equivalent occlusion. As another example, if server 13 locally maintains high-precision map information, road speed limit information, weather information, etc., the vehicle 11 can initiate a request to server 13 for specific information when there is a data acquisition need, and receive the corresponding information returned by server 13.

[0042] The server 13 can be a physical server containing an independent host, or it can be a virtual server, cloud server, etc., hosted in a host cluster. Furthermore, this application embodiment does not limit the number, type, or specific interaction method between the servers 13 and the vehicle. As for the network 10 that enables interaction between the vehicle 11 and the server 13, it can be based on the communication methods supported by the corresponding devices, specifically selecting an appropriate type of wireless network for communication; this application does not impose any limitations on this.

[0043] Furthermore, the vehicle described in this application (such as vehicle 11) can be a pickup truck, sedan, SUV (Sport Utility Vehicle), campervan, or truck, in terms of its functional form; and it can be a gasoline-powered vehicle or a new energy vehicle (such as a hybrid vehicle, pure electric vehicle, hydrogen fuel cell vehicle, or methanol fuel cell vehicle). This invention does not limit the specific form of the vehicle. Additionally, the occupant 12 in the cabin can be a driver sitting in the driver's seat, and the vehicle may also include at least one passenger sitting in other positions. This application also does not limit the number of occupants or their seating positions.

[0044] The driver assistance functions described in this application are used to assist the driver in driving the vehicle, such as Figure 1The vehicle 11 shown provides driver assistance functions to driver 12 based on data collected by various onboard sensors, which will not be described in detail here. These driver assistance functions can be Adaptive Cruise Control (ACC) or Intelligence Cruise Control (ICC), etc. Of course, this function can be implemented using L2, L3, or L4 level assisted / autonomous driving technologies, and this application does not impose any limitations on this. Furthermore, these driver assistance functions are also referred to as intelligent driving functions in some scenarios, and the driver assistance system is also referred to as an advanced driver assistance system.

[0045] However, it should be noted that the driver assistance functions implemented by the driver assistance system described in this application shall comply with the relevant laws, regulations and standards of the corresponding country and region (such as the place of sale and / or place of use of the vehicle), and provide relevant personnel (such as drivers) with corresponding operation interfaces so that they can choose to authorize or refuse to use it.

[0046] It should be noted that from the perspective of functional safety of driver assistance systems, both physical occlusion and optical interference result in the same image performance: the system cannot reliably detect key targets such as lane lines, vehicles, and pedestrians, leading to a degraded perception capability. Therefore, a safety response mechanism needs to be activated to effectively address the issue. This is precisely why this application unifies the two terms as "equivalent occlusion," mainly to emphasize that the two are "equivalent to occlusion," rather than being limited to "blocking" in a physical sense.

[0047] Figure 2 This is a flowchart illustrating an exemplary embodiment of a driver assistance control method based on camera occlusion detection. This method is applied to vehicles equipped with a forward-facing camera and driver assistance functions, such as those used in… Figure 1 Vehicle 11 is shown. (As shown) Figure 2 As shown, the method includes steps 202-206.

[0048] Step 202: When the vehicle's assisted driving function is activated, detect whether there is an equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-view camera. The equivalent occlusion includes physical occlusion and / or optical interference.

[0049] During vehicle operation, the driver assistance function can be activated by default or by a passenger (such as the driver or front passenger). Once activated, the vehicle is in driver assistance mode. This driver assistance function can include multiple levels, and the implementation logic of different levels may differ. For example, the cruising speed (i.e., the speed at which the vehicle automatically maintains a constant speed) at different levels can be positively correlated with the level. For instance, at higher levels, the vehicle can cruise at a relatively higher speed, while at lower levels, it can cruise at a relatively lower speed. Further details are omitted here.

[0050] It should be noted that during the driving process after the assisted driving function is activated, the vehicle may accelerate, decelerate, maintain a constant speed (i.e., the speed is not zero), or stop (i.e., the speed is zero, such as when waiting at a red light). Regardless of the speed, the forward-facing camera can capture real-time images of the area in front of the vehicle after the assisted driving function is activated (e.g., taking photos at preset intervals or continuously recording video at a preset sampling rate), and these images are processed by the assisted driving control module according to this scheme. Therefore, at any given moment when an image of the area in front of the vehicle is captured, the vehicle's speed may be zero or not.

[0051] It should be noted that the forward-facing camera can acquire optical signals from all visible objects within its field of view according to its mounting position to obtain an image, which is the vehicle front image described in this manual. Therefore, "front" as used in this manual refers to the area within the field of view of the forward-facing camera, and is not limited to directly in front of the vehicle's centerline; it can be any angle and position within ±75° of the centerline.

[0052] As mentioned earlier, the vehicle's driver assistance system described in this specification primarily relies on the image of the vehicle's front view captured by a forward-facing camera to make driver assistance decisions. When the vehicle's driver assistance function is activated, if the forward-facing camera experiences physical obstruction and / or optical interference, the image of the vehicle's front view captured by it will also experience corresponding equivalent obstruction. For example, physical obstruction refers to the forward-facing camera's field of view being completely or partially blocked by an object; in other words, a physical substance is attached to or obstructs the camera lens surface or the optical path, causing some or all of the image area to fail to form a normal image. Specifically, physical obstruction can be categorized into several types, including obstruction by attached objects, structural obstruction, and obstruction by external objects. Obstruction by attached objects can include dirt, dust, snow, oil, insect remains, stickers, tape, bird droppings, etc., adhering to the camera's outer surface. This type of obstruction can cause localized blurring, discoloration, or completely black / white blocks in the image of the vehicle's front view. Structural obstruction typically includes windshield wipers accidentally entering the field of view, distorted car logo reflectors, or obstruction by added equipment. This type of obstruction will display the outline of the obstructing object at a fixed position in the image of the vehicle's front view. External object occlusion can be caused by tree branches scratching, water dripping from the tunnel ceiling, or mud splashed by a vehicle in front. This type of occlusion is a sudden, localized coverage that usually changes over time. The physical occlusions mentioned above are usually in relatively fixed positions (such as being attached to the lens) and do not change with the scene content (even if the road changes, the occluded area usually always exists), which is a typical "sensor failure" mode.

[0053] The optical interference refers to the interference caused by optical phenomena such as illumination, reflection, and scattering that prevent the forward-facing camera from normally acquiring environmental images, resulting in a severe deterioration in image quality and the inability to extract effective perceptual information. Examples include strong glare (such as direct sunlight, oncoming headlights, or reflections from water / glass, typically manifesting as localized overexposure, halos, streaks, and dynamic light spots in the image in front of the vehicle), lens flare / ghosting (such as multiple reflections of strong light sources within the lens, manifesting as unrealistic light spots or rainbow patterns in the image in front of the vehicle), fog / water vapor condensation (such as fogging caused by large temperature differences inside and outside the lens, manifesting as overall blurring and a sharp drop in contrast in the image in front of the vehicle), the adhesion of transparent / semi-transparent objects such as rain / snow / water droplets (primarily optical effects, such as raindrops forming a lens effect on the lens surface, manifesting as distortion, localized magnification / blurring, and dynamic changes in the image in front of the vehicle), and low illumination / darkness without supplemental lighting (such as no streetlights at night or no lights on when entering a tunnel, manifesting as extremely low signal-to-noise ratio and loss of detail in the image in front of the vehicle). The aforementioned optical interference may change dynamically with the environment (such as the movement of the sun and changes in the position of glare), and usually does not have a fixed shape. Some of it can be mitigated by image enhancement algorithms, but in severe cases it still needs to be regarded as "equivalent occlusion".

[0054] For example, such as Figure 3The image shown is a real image of the front of the vehicle captured by the front-view camera under actual physical obstruction (part of the left side of the front-view camera is covered by a sticker) and optical interference (being illuminated by the strong light of the high beams of an oncoming vehicle).

[0055] The reason this solution applies a certain time threshold when the forward-facing camera is effectively occluded is to prevent the vehicle from colliding with objects in its direction of travel (such as other vehicles, pedestrians, or curbs) after its visual perception has decreased (due to the aforementioned equivalent occlusion), thus avoiding traffic accidents. Figure 4 As shown, assume that car A (the vehicle) is in front of car B (the vehicle in front) traveling in the same lane, and car A has activated ACC (Adaptive Cruise Control) with a following distance of S0. If car B brakes at a preset deceleration (e.g., 6 m / s^2) at time T0, car A can usually continue driving at its current speed for a period of time without immediately braking due to the existing following distance. If car A brakes at its maximum deceleration and decelerates to the same speed as car B after passing S2 (e.g., both vehicles stop), let S2 be the distance traveled relative to car B during this deceleration time (i.e., the relative distance between the two vehicles shortens by S2). Then, distance S1 = S0 - S2 is the maximum distance that car A can continue to travel at a constant speed after car B brakes. The corresponding time T1 - T1 is the maximum duration that car A can continue to travel at a constant speed after car B brakes. This duration is also called the maximum safe detection time for car A, which should not be less than the Fault Tolerant Time Interval (FTTI) specified during the vehicle design phase. In layman's terms, the maximum safe detection time refers to the safe duration for which the driver assistance system can drive autonomously (e.g., "blindly driving" at the current speed) after the camera is obstructed and malfunctions. After this time, it must downgrade or require manual intervention; otherwise, a collision with the vehicle in front is highly likely. The purpose of this solution is to determine the maximum safe detection time (i.e., the current safe time threshold) when the vehicle's forward-facing camera is effectively obstructed, and to control the vehicle to maintain driver assistance within this time, downgrading driver assistance or requesting manual intervention at the end of this time. This minimizes forced intervention and interference with driver assistance functions while ensuring safety.

[0056] In one embodiment, the vehicle can detect whether there is equivalent occlusion in the image in front of the vehicle in various ways. For example, during vehicle movement, if the camera is not obstructed, the content of the image it captures is usually not fixed (it changes); while after the camera is obstructed, the image usually contains content with fixed boundaries (i.e., the obstructed area usually maintains its boundaries). To address this, while the vehicle's current speed is not zero, the images in front of the vehicle captured in real time by the forward-facing camera can be continuously acquired, and the content in each image can be identified. Then, frame-by-frame comparison is performed: if a predetermined number of consecutive images in front of the vehicle contain the same content, it is determined that there is equivalent occlusion at the location corresponding to that content; conversely, if there is no predetermined number of images in front of the vehicle containing the same content, it is determined that there is no equivalent occlusion in the image in front of the vehicle. The preset number can be determined based on the sampling frame rate of the images in front of the vehicle. For example, at a frame rate of 24 FPS, the preset number can be 6. With this number, if a fixed piece of image content is detected to exist continuously for 0.25 seconds, then it is determined that these 6 images in front of the vehicle contain equivalent occlusion. In this case, the sampling time of the first image in front of the vehicle can be determined as the start time of the occlusion, or the sampling time of the sixth image in front of the vehicle can be determined as the time when the equivalent occlusion is first detected. Further details are omitted. Any two images in front of the vehicle containing the same image content means that the size, position, brightness, color, texture, and other parameters of the image content are the same or substantially the same in both images (i.e., the error / difference is within an acceptable range). This method determines the existence of equivalent occlusion when the image content of consecutive frames is the same, and is suitable for dynamic scenes where the vehicle speed is not zero.

[0057] For example, an appropriate method can be selected to detect equivalent occlusion based on the current road feature type. The current road feature type can be a stable feature scene or a variable feature scene. A stable feature scene refers to a road scene with a fixed road environment structure, unchanging visual features over a long period, and easy to model and predict, such as highways (clear lane lines, fixed guardrails / medians), urban expressways, internal roads of enclosed parks, and long straight tunnels (with consistent internal structures). A variable feature scene, on the other hand, refers to a road scene with a complex road environment, frequently changing visual content, and difficulty in establishing stable expectations, such as ordinary urban streets (with significant changes in roadside parking, pedestrians, temporary construction, and billboards), rural roads (without fixed lane lines, tree occlusion, and varying road surface materials), roads in rainy or snowy weather, and transition areas such as tunnel entrances / exits, bridges, and interchange ramps.

[0058] like Figure 5As shown, the area near the road intersection is a variable feature area. The variable feature road segments located within the variable feature area are segments BC and BE, and the road feature scenarios of these road segments are variable feature scenarios. Along the driving direction, the road segments located outside the variable feature area are stable feature road segments, such as segments AB, CD, and EF, and the road feature scenarios of these road segments are stable feature scenarios.

[0059] Specifically, the current road feature type of the current location can be determined based on the road feature information or type labeling results of the current road, where the type labeling results are pre-created based on the road feature information. It is understood that determining the current road feature type based on the road feature information may require the vehicle to run relatively complex data processing logic locally; while determining the current road feature type based on the type labeling results only requires locating the current location on the map and querying the corresponding labeling results, which is faster and has less latency.

[0060] The road characteristic information may include at least one of the following forms. For example, it may include road policy information of the current road, which is used to characterize policies related to the current road, such as traffic regulations, traffic restrictions, road maintenance / repair information, etc. This information may affect the type, number, and time of occurrence of traffic participants on the current road. For example, if the current road is a highway, then there is a high probability that there will be no risk sources such as pedestrians or non-motorized vehicles on that road; if the current road is a section of an urban elevated road with restrictions on motorcycles, then there is a high probability that there will be no risk sources such as trucks, pedestrians, or motorcycles on that road, which will not be elaborated further.

[0061] For example, the road feature information may also include current road status information, which describes the road conditions at the current moment. This information can be obtained by the vehicle from the cloud or sensed by its own onboard environmental perception devices. Examples include current traffic flow, traffic light information, and lane conditions (such as whether there is an emergency lane or a non-motorized vehicle lane), etc., which will not be elaborated further.

[0062] For example, the road feature information may also include historical road feature information of the current road, which characterizes the specific circumstances of traffic participants or related objects (such as road equipment or vehicles) that have appeared on the current road. This historical road feature information can be collected by environmental data acquisition equipment installed on the current road. This environmental data acquisition equipment can be installed and managed by public service departments, such as traffic police departments installing cameras, laser / millimeter-wave radar, etc., on both sides and / or above the road. These devices are fixed in location and typically operate stably, collecting high-quality data that helps determine accurate and reliable parking risk probability values.

[0063] And / or, the road feature information can also be collected through crowdsourcing by vehicles traveling on the current road before the current moment. It is understood that vehicles traveling on the current road before the current moment may include the vehicle itself (i.e., the vehicle has previously traveled on this road) or other vehicles. Of course, regardless of which vehicle collects the information, that vehicle should be equipped with corresponding environmental perception devices, such as cameras. It is understood that through crowdsourcing, a large number of vehicles traveling on the current road can upload real-world road safety-related information, such as "where pedestrians, motorcycles, construction cones, and / or abnormal obstacles frequently occur," to the server based on the results collected by their own environmental perception devices. This allows the server to accurately determine and label the characteristic scene types of each road / segment based on this real-world information. The aforementioned server can be a self-built server of the vehicle's brand, allowing multiple vehicles under that brand to upload their collected real-world information, thus helping the brand's vehicles to subsequently achieve reliable assisted driving functions. Of course, the aforementioned servers can also be established by authoritative organizations (such as transportation authorities, vehicle industry alliances, etc.), which helps more vehicles upload real information in a crowdsourcing manner, greatly increasing the number of real road information collection sources, helping to form a scale effect of information and improve the authenticity of information.

[0064] It is understandable that the positions of road markings, guardrails, streetlights, signs, etc., in the stable feature scene are almost unchanged. Therefore, the system can know in advance "what this place should look like under normal circumstances" through high-precision maps or crowdsourced maps. For example, the crowdsourced map can contain the standard forward image captured by the crowdsourced vehicle when it has traveled to this road segment / location before the current moment and the camera is unobstructed. For example, when the current road feature type is a stable feature scene, the standard forward image corresponding to the current position can be obtained and compared with the vehicle forward image captured by the forward-looking camera: if the vehicle forward image is different from the standard forward image corresponding to the current position, it can be determined that there is equivalent occlusion in the vehicle forward image; otherwise, if the two are the same, it can be determined that there is no equivalent occlusion. In this way, once the image seen by the camera does not match the "expectation" (for example, the lower half is suddenly blurred), it can be determined with high confidence that there is equivalent occlusion, rather than considering that the actual road environment has changed.

[0065] The visual characteristics of the same road in a variable feature scenario can be completely different at different times (e.g., it looks very different today and yesterday, or during peak and off-peak hours). On such roads, changes in the content of the image ahead of the vehicle are not necessarily due to equivalent occlusion, but are more likely due to changes in the actual road environment. Therefore, it is difficult to establish an accurate standard image ahead of the vehicle on such roads because there are too many false alarms (normal changes are easily mistaken for malfunctions). Therefore, in the case of a variable feature scenario, the presence of equivalent occlusion in the image ahead of the vehicle can be determined by analyzing its image features. For example, the image features of the image ahead of the vehicle can be extracted and input into a locally deployed occlusion recognition model for automatic and intelligent recognition, and the corresponding recognition results can be obtained; alternatively, the image ahead of the vehicle itself or the extracted image features can be included in an occlusion recognition request and sent to a cloud server, so that the server can use the image or features to call the locally deployed occlusion recognition model to perform occlusion recognition and receive the recognition results returned by the server. Since local resources on the vehicle are limited, a lightweight occlusion recognition model can be deployed, resulting in faster recognition speed and lower latency. The server, on the other hand, has abundant resources and can deploy a full suite of occlusion recognition models, leading to higher accuracy. In scenarios with varying features, this method no longer relies on comparing the image with a standard image to determine the existence of equivalent occlusion. Instead, it focuses on identifying abnormal patterns in the image itself (such as a sudden darkening of the entire image, the appearance of fixed color blocks, or the disappearance of textures).

[0066] Step 204: When an equivalent occlusion is detected for the first time in the image in front of the vehicle, the current safe time threshold is queried from the preset safe time threshold set based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the current position of the vehicle, and the equivalent occlusion is continuously detected while maintaining the assisted driving function.

[0067] The safety time thresholds in the set are predetermined based on the longest safe detection time for the vehicle under different driving scenarios. The longest safe detection time in any driving scenario represents the latest response time required for the vehicle to avoid colliding with the vehicle in front when the preceding vehicle brakes at a preset deceleration (e.g., 6 m / s²). The set of safety time thresholds can be pre-stored locally on the vehicle, such as during the vehicle's production phase; or it can be distributed to the vehicle via OTA (Over-The-Air) updates after the vehicle leaves the factory. Alternatively, this set can be configured in the vehicle's ADAS domain controller or intelligent driving software and stored as a state machine, decision table, or configuration file. The vehicle then needs to query the corresponding current safety time threshold (i.e., the current safety time threshold corresponding to the current driving) from the pre-stored set based on the obscured area and current road feature type. Alternatively, the set of safety time thresholds can be stored on a remote server. In this case, the vehicle needs to submit the obscured area and current road feature type to the server, which then queries the corresponding current safety time threshold from the locally stored set and returns it to the vehicle.

[0068] It is understood that each safe time threshold in the set of safe time thresholds can be pre-calibrated by technicians for various driving scenarios according to certain design rules and safety requirements. Thus, the current safe time threshold determined by the vehicle from this set based on the obscured area and the current road feature type is the pre-calibrated safe time threshold for the current driving scenario, and can be used for assisted driving control in the current driving scenario.

[0069] Typically, the practical implementation of this solution involves three stages: vehicle design, vehicle production, and vehicle operation (driving). In the vehicle design stage (deriving safe time thresholds based on safety theory), designers can calculate various factors such as FTTI (Fullest Safe Detection Time), compensation time required for synchronized braking of following vehicles, sensor latency, control link latency, and human-machine takeover time under different driving scenarios (i.e., different operating conditions) through kinematic modeling, scenario simulation, and functional safety analysis. Then, by combining these factors and selecting the most conservative combination that covers the worst-case scenario, a set of fixed safe time thresholds applicable to various driving scenarios is determined. For example, in stable characteristic scenarios with occlusion in the lower half of the driving area, a warning is issued in 2.8 seconds, and a degradation occurs in 4.2 seconds; in variable characteristic scenarios with occlusion in the core area, a warning is issued in 1.5 seconds, and a degradation occurs in 3.0 seconds, etc. During the vehicle production phase (where strategies and thresholds are written into the driver assistance system), scene classification rules, image partitioning logic, sensor configuration judgments, and corresponding safe time thresholds for each combination are pre-configured in the vehicle's ADAS domain controller or intelligent driving software. These are stored locally in the vehicle as operating parameters of the driver assistance system, such as in the form of a state machine, decision table, or configuration file. It is evident that these safe time thresholds are not calculated in real-time by the vehicle, but rather verified and solidified as operating parameters of the driver assistance system. During the vehicle operation phase (where the current safe time threshold is queried according to the preset strategy and driver assistance control is performed), the specific process can be found in steps 302-306 and related embodiments.

[0070] The following is a brief explanation of the calibration process for safe time thresholds in various driving scenarios.

[0071] In stable feature scenarios, first determine the longest safe detection time ts (actually the duration, the same below) that the camera can detect without a target for different vehicle speeds. Then, combine the maximum acceptable damage value (e.g., the speed difference between the front and rear vehicles reaches 30km / h as S0) to conduct safety strategy design analysis. Among them, the longest safe detection time ts refers to the normal driving time without collision. In strictly representative scenario motion, the formula for the time when the rear vehicle collides with the front vehicle is (1 / 2)*a*(ts^2)=V0*tf. Based on this, we can derive ts^2=2 V0*tf / a. Here, the following distance time (i.e., the time when the front vehicle stops immediately and the vehicle continues to travel at the current constant speed until the collision) tf can be defined as a fixed value of 2 seconds for following distance (if it is less than this value, the driver can be actively reminded that the following is too close and needs to pay attention to driving safety); the initial vehicle speed is V0; the deceleration of the front vehicle is a (which can be 2m / s^2; cases greater than this value are variable feature scenarios and are not classified as stable feature scenarios).

[0072] Assuming the vehicle in front decelerates at a rate of 2 m / s², the relevant parameters are shown in Table 1 below:

[0073] Table 1

[0074]

[0075] As mentioned earlier, the maximum safe detection time refers to the safe duration for which the driver assistance system can drive autonomously (e.g., "blindly driving" at the current speed) after the camera is obstructed and malfunctions. After this time, it must degrade or require manual intervention; otherwise, a collision with the vehicle in front is highly likely. However, the reality is more complex because the vehicle itself may brake prematurely, in which case the time that can be "tolerated" for camera failure will be even longer. Therefore, in order for the vehicle to stop within a safe distance (i.e., to ensure no collision), it needs to "take a little more time" than the vehicle in front to complete synchronized deceleration—this time is the "compensation" time mentioned in the table.

[0076] As shown in Table 1, when the deceleration required for a speed difference of 30 km / h between the front and rear vehicles is the same as the normal deceleration of the vehicle in front, if the vehicle speed is not less than 30 km / h, the initial braking time can be compensated by 4.2 seconds, thereby ensuring that there is no risk of collision.

[0077] Assuming the vehicle in front decelerates at a rate of 3 m / s², the relevant parameters are shown in Table 2 below:

[0078] Table 2

[0079]

[0080] As shown in Table 2, when the deceleration required for a speed difference of 30 km / h between the vehicles in front and behind is the same as the normal deceleration of the vehicle in front, and if the vehicle speed is not less than 30 km / h, the initial braking time can be compensated by 2.8 seconds, thus ensuring the goal of eliminating the risk of a collision. However, in this driving scenario, the human driver will make a controllable intervention during heavy braking. Therefore, a deceleration slightly greater than that of the vehicle in front can still be considered a controllable state, so a intervention time of 2.0 seconds can also achieve safety.

[0081] Assuming the vehicle in front decelerates at a rate of 4 m / s², the relevant parameters are shown in Table 3 below:

[0082] Table 3

[0083]

[0084] As shown in Table 3, when the deceleration required for a speed difference of 30 km / h between the vehicles in front and behind is the same as the normal deceleration of the vehicle in front, and if the vehicle speed is not less than 30 km / h, the initial braking time can be compensated by 2.1 seconds, thus ensuring the goal of eliminating the risk of a collision. However, in this driving scenario, the human driver will make a controllable intervention during heavy braking. Therefore, a deceleration slightly greater than that of the vehicle in front can still be considered a controllable state, so a intervention time of 2.5 seconds can also achieve safety.

[0085] Assuming the vehicle in front decelerates at a rate of 5 m / s², the relevant parameters are shown in Table 4 below:

[0086] Table 4

[0087]

[0088] As shown in Table 4, when the deceleration required for a speed difference of 30 km / h between the front and rear vehicles is the same as the normal deceleration of the vehicle in front, if the vehicle speed is not less than 30 km / h, the initial braking time can be increased by 1.72 seconds, thereby ensuring that there is no risk of collision.

[0089] Assuming the vehicle in front decelerates at a rate of 6 m / s², the relevant parameters are shown in Table 5 below:

[0090] Table 5

[0091]

[0092] As shown in Table 5, when the deceleration required for a speed difference of 30 km / h between the front and rear vehicles is the same as the normal deceleration of the vehicle in front, if the vehicle speed is not less than 30 km / h, the initial braking time can be compensated by 1.4 seconds, thereby ensuring that there is no risk of collision.

[0093] Based on the above calibration results, corresponding safe time thresholds can be set for various driving scenarios (the thresholds for typical scenarios can be found in the following examples), and a corresponding set of safe time thresholds can be constructed.

[0094] In one embodiment, the current road feature type can be determined in advance. For example, when the vehicle's assisted driving function is activated, the current road feature type at the vehicle's current location can be determined in real time. Based on this, when subsequently querying the current safe time threshold (matching the current driving scenario) from the set of safe time thresholds based on the occluded area and the current road feature type, the occluded area of ​​the image in front of the vehicle can be determined according to the detection result of the equivalent occlusion, and then the current safe time threshold can be queried from the preset set of safe time thresholds based on the occluded area and the current road feature type. This method determines the current road feature type in real time before the equivalent occlusion is actually detected, which helps to quickly query the current safe time threshold after the equivalent occlusion is detected, improving the response speed and processing efficiency for occlusion, and avoiding collisions caused by excessively long process times.

[0095] In another embodiment, the current road feature type can be temporarily determined after equivalent occlusion is detected. For example, when querying the current safe time threshold from the set of safe time thresholds based on the occluded area and the current road feature type, the occluded area of ​​the image in front of the vehicle can be determined first based on the detection result of the equivalent occlusion, and the current road feature type of the vehicle's current position can be determined; then, the current safe time threshold can be queried from the set of preset safe time thresholds based on the occluded area and the current road feature type.

[0096] In the foregoing embodiments, when determining the current road feature type of the vehicle's current location, the current road feature type can be determined based on the current road feature information or type marking results. The type marking results are pre-created based on the road feature information. The road feature information includes road policy information, road status information, and / or historical road feature information. The historical road feature information is collected by vehicles traveling on the current road before the current time and / or by environmental data acquisition devices installed on the current road. For details, please refer to the descriptions in the preceding embodiments; they will not be repeated here.

[0097] In one embodiment, as mentioned above, the current road feature type may be a stable feature scenario or a variable feature scenario. The current safe time threshold corresponding to the stable feature scenario can be set to be greater than the current safe time threshold corresponding to the stable feature scenario. It is understood that a larger current safe time threshold indicates a longer period for the vehicle to maintain a constant speed and drive safely. The basis for this setting is that the feature complexity of stable feature scenarios is lower than that of variable feature scenarios, thus this setting conforms to the risk distribution characteristics of actual driving scenarios. For example, in stable feature scenarios (such as highways), the road structure is fixed, traffic flow is orderly, and the behavior of the vehicle in front is predictable. Therefore, even if the forward-facing camera is briefly obstructed, the system can still maintain safe following for a relatively long time through map prior knowledge and radar, thus allowing for a longer tolerance time in this scenario. However, in variable feature scenarios (such as urban intersections), pedestrians / non-motorized vehicles may suddenly appear, lane lines may be blurred, and the vehicle in front may brake suddenly frequently. In such cases, failure of the forward-facing camera means an instantaneous loss of critical perception capabilities, thus requiring rapid degradation or manual takeover. This solution sets a "safe time threshold for stable characteristic scenarios > safe time threshold for variable characteristic scenarios," binding the abstract safe time window with specific road risk distribution patterns. This reflects the assisted driving safety logic of "rapid response in high-risk environments and gradual withdrawal in low-risk environments," balancing driving safety and solution usability.

[0098] Step 206: If the duration of the equivalent occlusion reaches the current safe time threshold, then the driver assistance function is downgraded and / or a takeover request is issued to the driver.

[0099] In one embodiment, downgrading the driver assistance function may include: controlling the vehicle to decelerate (i.e., controlling the vehicle to decelerate to avoid a collision when the driver assistance function is activated); or, directly turning off the driver assistance function (i.e., disabling driver assistance).

[0100] And / or, issuing a takeover request to the driver may include: playing a voice request for takeover (such as "The camera is obstructed, please pay attention to driving safety") by an audio playback device in the cockpit, and / or displaying a screen request for takeover (such as the text "The camera is obstructed, please pay attention to driving safety", an image or animation showing the obstruction effect or takeover prompt, etc.) by a display device in the cockpit.

[0101] Through the above embodiments, when the vehicle activates the assisted driving function and detects for the first time an equivalent occlusion (physical obstruction and / or optical interference, etc.) in the image ahead of the vehicle, it can query the current safe time threshold from a preset set of safe time thresholds based on the corresponding obstructed area and the current road feature type. Each safe time threshold in this set is predetermined based on the vehicle's longest safe detection time under different driving scenarios. The longest safe detection time under any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario. After determining the current safe time threshold, the vehicle continues to detect the equivalent occlusion while maintaining the assisted driving function until the duration of the equivalent occlusion reaches the aforementioned current safe time threshold, at which point the assisted driving function is downgraded and / or a takeover request is issued to the driver.

[0102] It is understandable that each safety time threshold in the set of safety time thresholds is predetermined based on the longest safety detection time of the vehicle in different driving scenarios. The longest safety detection time in any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario. Moreover, the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position can comprehensively characterize the current driving scenario of the vehicle. Therefore, the current safety time threshold queried from this set based on the occluded area and the current road feature type can meet the shortest time requirement of not colliding with the vehicle in front in the current driving scenario. In other words, in the current driving scenario, even if there is a vehicle in front braking at a preset deceleration, the vehicle can still maintain its current speed for a period of time. It will only degrade or request manual intervention when the duration of the obstruction reaches the current safe time threshold. This ensures that the vehicle does not rear-end the vehicle in front, thereby minimizing the forced intervention and interference with the driver assistance functions while ensuring driving safety in the abnormal situation of "equivalent obstruction of the forward-view camera". This effectively addresses the above-mentioned abnormal situation and meets the requirements of relevant regulations.

[0103] In one embodiment, when the current road feature type is a variable feature scenario, the image in front of the vehicle can be divided into a core area located in the center of the image and a non-core area located at the edge of the image. The "core area" and "non-core area" do not refer to the physical area of ​​the camera hardware itself (such as a part of the lens), but rather to logical areas within the image (i.e., video frame) output by the camera, divided according to the importance of driving safety; they are spatial sub-regions in the image captured by the forward-looking camera, belonging to the software processing level. This division is based on perspective projection and driving task requirements, and is not related to the specific situation of equivalent occlusion. The core area can be a "Region of Interest (ROI)" on the image, which typically contains the area where objects related to driving safety are most likely to appear. For example, the core area can be the area inside the inscribed circle / ellipse of the image in front of the vehicle, in which case the non-core area is the area outside the inscribed circle / ellipse; or, the core area can also be the area inside a trapezoid or rectangle located slightly below the center of the image, in which case the area outside the trapezoid or rectangle is the non-core area.

[0104] When the current road feature type is a stable feature scenario, the image in front of the vehicle can be divided into an upper half corresponding to the nearby road surface and a lower half corresponding to the distant field of vision. This division is a logical partition based on the image coordinate system. For example, it can be divided into two regions using the center line of the image height (such as the 540th line of a 1080p image) as the boundary, or dynamically divided according to the actual road area of ​​interest based on the perspective projection model. The upper half mainly includes distant roads, traffic signs, traffic lights, the sky, and the roof of the vehicle in front, and is more used for environmental understanding, navigation assistance, and long-distance perception. The lower half mainly includes the nearby road surface, lane lines, curbs, and nearby obstacles. This area is crucial for lateral control (lane keeping) and longitudinal following. It is evident that anomalies in the lower half have a greater and more urgent impact on driving safety, while anomalies in the upper half have a relatively smaller impact and a longer tolerance time can be set.

[0105] For example, such as Figure 6 As shown in 'a', the boundary between the core area and the non-core area is the inscribed ellipse of the image in front of the vehicle, where the area inside the ellipse is the core area and the area outside the ellipse is the non-core area. Figure 6 As shown in b, the boundary between the upper and lower halves is the centerline of the image in front of the vehicle, with the upper half above the centerline and the lower half below. Alternatively, the upper half can be the portion of the upper half of the image in front of the vehicle located within the core area, and the lower half can be the portion of the lower half of the image in front of the vehicle located within the core area. This approach minimizes interference from low-influence factors outside the core area, improving the accuracy of subsequent decision-making.

[0106] In one embodiment, after determining the current safe time threshold based on the obscured area and the current road feature type, in order to improve the accuracy of the time threshold in the current driving scenario, the current safe time threshold can be comprehensively adjusted using the perception results of other sensors, the friction information of the current road (such as whether there is water accumulation, ice, road surface material, etc.), the vehicle deceleration performance information (such as maximum deceleration, deceleration mechanism delay, etc.), and the subsequent step 206 is executed using the adjusted time threshold.

[0107] In one embodiment, step 206 can be executed in different ways depending on the specific scenario of the current road feature type. For example, if the current road feature type is a stable feature scenario and the occluded area is located in the lower half of the road, if the duration of the equivalent occlusion (i.e., the duration of continuous occlusion) reaches a first duration threshold, a visual limitation warning message can be output to the driver and / or high-risk driving actions of the assisted driving function can be restricted; if the duration of the equivalent occlusion reaches a second duration threshold, the assisted driving function can be downgraded and / or a takeover request can be issued to the driver (in this driving scenario, the current safe time threshold includes the first duration threshold and the second duration threshold). The first duration threshold is less than the second duration threshold, for example, they can be 2.8 seconds and 4.2 seconds respectively.

[0108] For example, if the current road feature type is a stable feature scenario and the occluded area is located in the upper half of the area, and the duration of the equivalent occlusion reaches a third duration threshold, then the driver assistance function is downgraded and / or a takeover request is issued to the driver; wherein, the second duration threshold is less than the third duration threshold, such as the third duration threshold being set to 10 seconds.

[0109] Because the second duration threshold is smaller than the third duration threshold, de-escalation / takeover can be implemented more quickly when anomalies occur in the lower half of the traffic area (where visual loss due to occlusion is relatively more urgent). Conversely, when anomalies occur in the upper half of the traffic area (where visual loss due to occlusion is relatively less urgent), processing can be delayed slightly, achieving targeted handling based on the urgency / severity of the anomaly. Furthermore, since the first duration threshold is smaller than the second duration threshold, in the event of anomalies in the lower half of the traffic area, tiered alerts (first a warning, then de-escalation / takeover) can be used to prepare the driver psychologically, preventing accidents caused by panic after sudden de-escalation or takeover.

[0110] For example, if the current road feature type is a variable feature scenario and the obscured area is located in the core area, if the duration of the equivalent obscuration reaches a fourth duration threshold, a visual limitation warning message can be output to the driver and / or high-risk driving actions of the assisted driving function can be restricted; if the duration of the equivalent obscuration reaches a fifth duration threshold, the assisted driving function can be downgraded and / or a takeover request can be issued to the driver. In other words, when the continuous obscuration reaches the fourth duration threshold, the driver is first warned and / or a high-risk action is displayed; then, when the continuous obscuration further reaches the fifth duration threshold, the assisted driving function is downgraded and / or a manual takeover request is issued. The fourth duration threshold is less than the fifth duration threshold, for example, 1.5 seconds and 3 seconds respectively. If the obscured area is located in the non-core area, the obscuration can be ignored directly without any specific processing.

[0111] This approach allows for "progressive risk control" when the core area is obstructed, effectively avoiding abrupt exits and preventing drivers from being caught off guard. If the assisted driving system immediately exits upon detecting obstruction (e.g., by directly displaying "Please take over"), the driver may make operational errors due to lack of preparation. A fourth time threshold (e.g., 1.5 seconds) first issues a "visual impairment" warning, giving the driver a psychological alert that "the system's perception ability is decreasing, please be careful." Only at the fifth time threshold (e.g., 3.0 seconds) is takeover required, reserving a buffer period of approximately 1.5 seconds, which aligns with human cognitive response patterns and helps improve the success rate of manual takeover. Moreover, early suppression of key functions can reduce the accident rate: restricting high-risk actions (such as lane changes and autonomous starts) during the warning stage can prevent the execution of operations requiring high-precision perception when vision is unreliable (e.g., prohibiting automatic start when pedestrians cannot be clearly seen at intersections can avoid "ghost pedestrian" accidents). By adopting this proactive safety intervention strategy, rather than passively waiting for failure, it ensures that even if manual takeover is ultimately required, the system has significantly reduced the window of dangerous operations in the early stages, effectively improving the safety of both assisted driving and manual takeover.

[0112] In the above embodiments, the visual impairment warning information can be output to the driver via voice or visual means. For example, the visual impairment warning voice can be played by an in-cabin audio playback device, and / or the visual impairment warning image (such as text, pictures, or animations) can be displayed by an in-cabin display device. Furthermore, high-risk driving actions that restrict the execution of the assisted driving functions may include lane changing, overtaking, acceleration, and / or starting.

[0113] The previous embodiments rely on preset (static) current safe time thresholds (such as 2.8s, 4.2s, 3.0s, etc.), applicable to the hypothetical scenario of "the vehicle in front decelerating at a constant or normal speed." However, in reality, the vehicle in front may brake suddenly (e.g., brake suddenly at a deceleration of not less than 6 m / s^2). In this case, even if the obstruction lasts only for a very short time (e.g., 1s), there may not be enough time to brake, resulting in a rear-end collision. To address this, non-visual sensors such as lidar and V2X communication modules additionally equipped on the vehicle can be used to detect whether the vehicle in front is decelerating, and emergency measures can be taken when it is determined that the vehicle in front is decelerating, rather than continuing to drive at a constant speed until the current safe time threshold is reached.

[0114] In one embodiment, if the duration of the equivalent occlusion does not reach the current safety time threshold, in response to the non-visual sensor detecting that the vehicle in front is decelerating, an emergency safety time threshold corresponding to the deceleration of the vehicle in front can be determined first. Then, when the non-visual sensor detects that the duration of the continuous deceleration of the vehicle in front reaches the emergency safety time threshold, the driver assistance function (such as controlling the vehicle to decelerate) can be downgraded and / or a takeover request can be issued to the driver.

[0115] Based on the calibration results corresponding to Tables 1 to 5 above, the correspondence between the deceleration of the preceding vehicle and the starting time point of the following vehicle's braking can be determined, such as... Figure 7 As shown in the diagram, a deceleration of 6 m / s^2 corresponds to 1.4 s. This means that if the vehicle in front brakes at a deceleration of 6 m / s^2, and the vehicle can start and continue to decelerate within 1.4 s, a rear-end collision can be avoided. Otherwise, if the vehicle starts to decelerate after 1.4 s, a rear-end collision is highly likely.

[0116] In another embodiment, if the duration of the equivalent occlusion does not reach the current safe time threshold, the driver assistance function can be immediately downgraded (e.g., immediately control the vehicle to decelerate) and / or a takeover request can be issued to the driver in response to the non-visual sensor detecting that the vehicle in front is decelerating. This is to effectively deal with the abnormal situation where the safe time window is shortened due to the sudden deceleration of the vehicle in front, and to avoid a collision due to braking too late in this situation.

[0117] The above approach can fully utilize the vehicle's sensor redundancy to enhance system resilience, especially in the case of abnormal situations where the forward-facing camera is obstructed: when the forward-facing camera fails due to equivalent obstruction, the perception results of non-visual sensors are used as a safety backup, demonstrating the comprehensive and collaborative safety processing logic of heterogeneous sensors.

[0118] In one embodiment, the vehicle may also be equipped with a surround-view camera that overlaps with the field of view of the forward-facing camera. In this case, the surround-view camera can be used for close-range emergency obstacle avoidance. Normally, the sensing distance of the surround-view camera is shorter than that of the forward-facing camera; however, when the forward-facing camera is obstructed, the surround-view camera can be used to detect obstacles at shorter distances, such as pedestrians and vehicles that suddenly appear (like "ghost peeking out"). Figure 8 As shown, the "front wide-angle camera" is a forward-facing camera, and the surrounding cameras on the front and left / right sides all have overlapping fields of view with the forward-facing camera. In this case, if there is an equivalent occlusion and the vehicle's current speed is not zero (i.e., the vehicle is moving), the vehicle can be controlled to avoid the obstacle in response to the surrounding cameras detecting an obstacle located within the overlapping fields of view. This obstacle can be any form of static or dynamic obstacle, and this specification does not limit this.

[0119] This approach introduces surround-view cameras as a redundancy means of near-field perception when the front-view camera is obstructed, making it particularly suitable for providing critical safety backup in low-speed or complex urban scenarios. This method effectively compensates for near-field perception blind spots when the front-view camera fails due to obstruction: front-view cameras typically cover a range of 5-150 meters, but blind spots often exist in the 0-15 meter range (especially directly in front of the vehicle) due to installation height and tilt angle. Surround-view cameras (such as forward-facing fisheye lenses) can cover the near-field area of ​​0-3 meters or even 0-6 meters. Therefore, when the front-view camera fails due to obstruction, the surround-view camera can be used to detect obstacles that suddenly appear in front of the vehicle (such as children, pets, or traffic cones), improving perception and obstacle avoidance capabilities in the aforementioned blind spots and effectively preventing low-speed collisions.

[0120] Please see Figure 9 , Figure 9 This is an exemplary embodiment illustrating the hardware structure of a vehicle housing a driver assistance control device based on camera occlusion detection. At the hardware level, the device includes a processor 902, an internal bus 904, a network interface 906, memory 908, and non-volatile memory 910, and may also include other hardware required for various operations. One or more embodiments of this application can be implemented in software, for example, the processor 902 reads the corresponding computer program from the non-volatile memory 910 into memory 908 and then runs it. Of course, besides software implementation, one or more embodiments of this application do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution entity of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0121] Please see Figure 10 , Figure 10This is a block diagram illustrating an exemplary embodiment of a driver assistance control device based on camera occlusion detection. This device can be applied to... Figure 9 The vehicle shown embodies the technical solution of this application. The device is applied to a vehicle equipped with a forward-facing camera and possessing driver assistance functions, and the device includes:

[0122] The camera occlusion detection unit 1001 is used to detect whether there is equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-looking camera when the vehicle's assisted driving function is activated. The equivalent occlusion includes physical occlusion and / or optical interference.

[0123] The time threshold determination unit 1002 is used to, when the equivalent occlusion in the image in front of the vehicle is detected for the first time, query the current safe time threshold from a preset safe time threshold set based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the current position of the vehicle, and continuously detect the equivalent occlusion while maintaining the assisted driving function; wherein, each safe time threshold in the safe time threshold set is predetermined based on the longest safe detection time of the vehicle in different driving scenarios, and the longest safe detection time in any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario;

[0124] The driver assistance control unit 1003 is configured to downgrade the driver assistance function and / or issue a takeover request to the driver if the duration of the equivalent occlusion reaches the current safe time threshold.

[0125] Optionally, the camera occlusion detection unit 1001 is specifically used for:

[0126] When the vehicle's current speed is not zero, continuously acquire real-time images of the front of the vehicle captured by the forward-facing camera, and identify the content of each image. If a preset number of consecutive images of the front of the vehicle contain the same content, then determine that there is equivalent occlusion at the location corresponding to that content in the image; or...

[0127] If the current road feature type is a stable feature scenario, and the image in front of the vehicle is different from the standard image in front of the vehicle at the current location, then it is determined that there is equivalent occlusion in the image in front of the vehicle; if the current road feature type is a variable feature scenario, the presence of equivalent occlusion in the image in front of the vehicle is determined by analyzing the image features of the image in front of the vehicle.

[0128] Optional,

[0129] The device further includes a road type determination unit 1004, used to determine the current road feature type of the vehicle's current position in real time when the vehicle's assisted driving function is activated; the time threshold determination unit 1002 is specifically used to determine the occluded area of ​​the image in front of the vehicle based on the detection result of the equivalent occlusion, and query the current safe time threshold from a preset safe time threshold set based on the occluded area and the current road feature type; or;

[0130] The time threshold determination unit 1002 is specifically used to: determine the occluded area of ​​the image in front of the vehicle based on the detection result of the equivalent occlusion, and determine the current road feature type of the current position of the vehicle; and query the current safe time threshold from the preset safe time threshold set based on the occluded area and the current road feature type.

[0131] Optionally, the road type determination unit 1004 is specifically used for:

[0132] The current road feature type of the current location is determined based on the road feature information or type labeling result of the current road, wherein the type labeling result is pre-created based on the road feature information;

[0133] The road feature information includes the current road policy information, road status information, and / or historical road feature information. The historical road feature information is collected by vehicles traveling on the current road before the current time and / or by environmental data collection devices installed on the current road.

[0134] Optionally, the current road feature type is a stable feature scenario or a variable feature scenario, wherein the current safe time threshold corresponding to the stable feature scenario is greater than the current safe time threshold corresponding to the variable feature scenario.

[0135] Optionally, when the current road feature type is a stable feature scenario, the image in front of the vehicle is divided into an upper half corresponding to the nearby road surface and a lower half corresponding to the distant field of vision. The driver assistance control unit 1003 is specifically used for:

[0136] When the current road feature type is a stable feature scenario and the occluded area is located in the lower half of the area, if the duration of the equivalent occlusion reaches a first duration threshold, a visual limitation warning message is output to the driver and / or high-risk driving actions of the assisted driving function are restricted. If the duration of the equivalent occlusion reaches a second duration threshold, the assisted driving function is downgraded and / or a takeover request is issued to the driver.

[0137] If the current road feature type is a stable feature scenario and the occluded area is located in the upper half of the area, and the duration of the equivalent occlusion reaches a third duration threshold, then the driver assistance function is downgraded and / or a takeover request is issued to the driver.

[0138] Wherein, the first duration threshold is less than the second duration threshold, the second duration threshold is less than the third duration threshold, and the high-risk driving actions include lane changing, overtaking, and / or acceleration.

[0139] Optionally, when the current road feature type is a variable feature scenario, the image in front of the vehicle is divided into a core area located in the center of the image and a non-core area located at the edge of the image. The driver assistance control unit 1003 is specifically used for:

[0140] If the current road feature type is a variable feature scenario and the occluded area is located in the core area, if the duration of the equivalent occlusion reaches the fourth duration threshold, a visual limitation warning message will be output to the driver and / or high-risk driving actions of the assisted driving function will be restricted. If the duration of the equivalent occlusion reaches the fifth duration threshold, the assisted driving function will be downgraded and / or a takeover request will be issued to the driver.

[0141] Wherein, the fourth duration threshold is less than the fifth duration threshold, and the high-risk driving actions include lane changing, overtaking, and / or acceleration.

[0142] Optionally, the vehicle is also equipped with a non-visual sensor for detecting the motion of the vehicle ahead, and the device further includes a following braking unit 1005 for:

[0143] If the duration of the equivalent occlusion does not reach the current safe time threshold, in response to the non-visual sensor detecting that the vehicle ahead is slowing down,

[0144] Degrade the driver assistance features and / or issue a takeover request to the driver; or

[0145] Determine an emergency safety time threshold corresponding to the deceleration of the vehicle ahead, and when the non-visual sensor detects that the duration of the continuous deceleration of the vehicle ahead reaches the emergency safety time threshold, downgrade the driver assistance function and / or issue a takeover request to the driver.

[0146] Optionally, the driver assistance control unit 1003 is specifically used for:

[0147] Control the vehicle to decelerate, or disable the driver assistance function; and / or,

[0148] The request for takeover is played via the in-cabin audio playback device, and / or displayed via the in-cabin display device.

[0149] Optionally, the vehicle is also equipped with a surround-view camera that has an overlapping field of view with the forward-looking camera, and the device further includes a near-field obstacle avoidance unit 1006, used for:

[0150] In the presence of the equivalent occlusion and when the vehicle's current speed is not zero, the vehicle is controlled to avoid the obstacle in response to the surround-view camera detecting an obstacle located within the overlapping field of view.

[0151] The specific implementation process of the functions and roles of each unit in the device is detailed in the implementation process of the corresponding steps in the method, and will not be repeated here.

[0152] Accordingly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the assisted driving control method based on camera occlusion detection as described in any of the preceding embodiments.

[0153] Accordingly, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the assisted driving control method based on camera occlusion detection as described in any of the preceding embodiments.

[0154] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0155] The systems, devices, modules, or units described in the embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

[0156] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0157] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0158] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0159] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0160] The description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0161] The terminology used in one or more embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the scope of one or more embodiments of this application. The singular forms “a,” “the,” and “the” used in one or more embodiments of this application and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0162] It should be understood that although the terms first, second, third, etc., may be used to describe various information in one or more embodiments of this application, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of one or more embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0163] The above description is merely a preferred embodiment of one or more embodiments of this application and is not intended to limit the scope of one or more embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of one or more embodiments of this application should be included within the protection scope of one or more embodiments of this application.

[0164] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

Claims

1. A driver assistance control method based on camera occlusion detection, characterized in that, Applied to vehicles equipped with forward-facing cameras and driver assistance functions, wherein the vehicles are also equipped with non-visual sensors for detecting the motion state of vehicles ahead, the method includes: When the vehicle's driver assistance function is activated, the system detects whether there is an equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-facing camera. The equivalent occlusion includes physical occlusion and / or optical interference. Upon initial detection of equivalent occlusion in the image ahead of the vehicle, the system queries a preset safe time threshold from a set of safe time thresholds based on the occluded area of ​​the image ahead and the current road feature type of the vehicle's current position. While maintaining the assisted driving function, the system continuously detects the equivalent occlusion. Each safe time threshold in the set of safe time thresholds is predetermined based on the vehicle's longest safe detection time under different driving scenarios. The longest safe detection time under any driving scenario characterizes the latest response time required for the vehicle to avoid colliding with the vehicle ahead when the preceding vehicle brakes at a preset deceleration in that driving scenario. If the duration of the equivalent occlusion reaches the current safe time threshold, the driver assistance function will be downgraded and / or a takeover request will be issued to the driver. If the duration of the equivalent occlusion does not reach the current safe time threshold, and the non-visual sensor detects that the vehicle in front is decelerating, then the driver assistance function is downgraded and / or a takeover request is issued to the driver; or, an emergency safe time threshold corresponding to the deceleration of the vehicle in front is determined, and when the duration of the continuous deceleration of the vehicle in front reaches the emergency safe time threshold, the driver assistance function is downgraded and / or a takeover request is issued to the driver.

2. The method of claim 1, wherein, The detection of whether there is equivalent occlusion in the real-time image of the vehicle's front captured by the forward-facing camera includes: When the vehicle's current speed is not zero, continuously acquire real-time images of the front of the vehicle captured by the forward-facing camera, and identify the content of each image. If a preset number of consecutive images of the front of the vehicle contain the same content, then determine that there is equivalent occlusion at the location corresponding to that content in the image; or... If the current road feature type is a stable feature scenario, and the image in front of the vehicle is different from the standard image in front of the vehicle at the current location, then it is determined that there is equivalent occlusion in the image in front of the vehicle; if the current road feature type is a variable feature scenario, the presence of equivalent occlusion in the image in front of the vehicle is determined by analyzing the image features of the image in front of the vehicle.

3. The method according to claim 1, characterized in that, The method further includes: when the vehicle's assisted driving function is activated, determining the current road feature type of the vehicle's current position in real time; querying the current safe time threshold from a preset safe time threshold set based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position includes: determining the occluded area of ​​the image in front of the vehicle based on the detection result of the equivalent occlusion, and querying the current safe time threshold from the preset safe time threshold set based on the occluded area and the current road feature type; or; The step of querying the current safe time threshold from a preset set of safe time thresholds based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position includes: determining the occluded area of ​​the image in front of the vehicle based on the detection result of the equivalent occlusion, and determining the current road feature type of the vehicle's current position; and querying the current safe time threshold from the preset set of safe time thresholds based on the occluded area and the current road feature type.

4. The method of claim 3, wherein, The determination of the current road feature type of the vehicle's current location includes: The current road feature type of the current location is determined based on the road feature information or type labeling result of the current road, wherein the type labeling result is pre-created based on the road feature information; The road feature information includes the current road policy information, road status information, and / or historical road feature information. The historical road feature information is collected by vehicles traveling on the current road before the current time and / or by environmental data collection devices installed on the current road.

5. The method of claim 4, wherein, The current road feature type is either a stable feature scenario or a variable feature scenario, wherein the current safe time threshold corresponding to the stable feature scenario is greater than the current safe time threshold corresponding to the variable feature scenario.

6. The method of claim 1, wherein, When the current road feature type is a stable feature scenario, the image in front of the vehicle is divided into an upper half corresponding to the nearby road surface and a lower half corresponding to the distant field of view. If the duration of the equivalent occlusion reaches the current safe time threshold, the driver assistance function is downgraded and / or a takeover request is issued to the driver, including: When the current road feature type is a stable feature scenario and the occluded area is located in the lower half of the area, if the duration of the equivalent occlusion reaches a first duration threshold, a visual limitation warning message is output to the driver and / or high-risk driving actions of the assisted driving function are restricted. If the duration of the equivalent occlusion reaches a second duration threshold, the assisted driving function is downgraded and / or a takeover request is issued to the driver. If the current road feature type is a stable feature scenario and the occluded area is located in the upper half of the area, and the duration of the equivalent occlusion reaches a third duration threshold, then the driver assistance function is downgraded and / or a takeover request is issued to the driver. Wherein, the first duration threshold is less than the second duration threshold, the second duration threshold is less than the third duration threshold, and the high-risk driving actions include lane changing, overtaking, and / or acceleration.

7. The method of claim 1, wherein, In the case where the current road feature type is a variable feature scenario, the image in front of the vehicle is divided into a core area located in the center of the image and a non-core area located at the edge of the image. If the duration of the equivalent occlusion reaches the current safe time threshold, the driver assistance function is downgraded and / or a takeover request is issued to the driver, including: If the current road feature type is a variable feature scenario and the occluded area is located in the core area, if the duration of the equivalent occlusion reaches the fourth duration threshold, a visual limitation warning message will be output to the driver and / or high-risk driving actions of the assisted driving function will be restricted. If the duration of the equivalent occlusion reaches the fifth duration threshold, the assisted driving function will be downgraded and / or a takeover request will be issued to the driver. Wherein, the fourth duration threshold is less than the fifth duration threshold, and the high-risk driving actions include lane changing, overtaking, and / or acceleration.

8. The method according to claim 1, characterized in that, Degrading the driver assistance function includes: controlling the vehicle to decelerate, or disabling the driver assistance function; and / or, The process of issuing a takeover request to the driver includes: playing a voice request for takeover via an audio playback device in the cockpit, and / or displaying a screen requesting takeover via a display device in the cockpit.

9. The method of claim 1, wherein, The vehicle is also equipped with a surround-view camera that has an overlapping field of view with the forward-looking camera, and the method further includes: In the presence of the equivalent occlusion and when the vehicle's current speed is not zero, the vehicle is controlled to avoid the obstacle in response to the surround-view camera detecting an obstacle located within the overlapping field of view.

10. An auxiliary driving control device based on camera occlusion detection, characterized in that, A device applicable to vehicles equipped with forward-facing cameras and driver assistance functions, wherein the vehicle is also equipped with non-visual sensors for detecting the motion of vehicles ahead, the device comprising: The camera occlusion detection unit is used to detect whether there is equivalent occlusion in the real-time image of the front of the vehicle captured by the forward-view camera when the vehicle's assisted driving function is activated. The equivalent occlusion includes physical occlusion and / or optical interference. The time threshold determination unit is used to, upon first detecting an equivalent occlusion in the image in front of the vehicle, query the current safe time threshold from a preset safe time threshold set based on the occluded area of ​​the image in front of the vehicle and the current road feature type of the vehicle's current position, and continuously detect the equivalent occlusion while maintaining the assisted driving function; wherein, each safe time threshold in the safe time threshold set is predetermined based on the longest safe detection time of the vehicle under different driving scenarios, and the longest safe detection time under any driving scenario is used to characterize the latest response time required for the vehicle not to collide with the vehicle in front when the vehicle in front brakes at a preset deceleration in that driving scenario; The driver assistance control unit is configured to downgrade the driver assistance function and / or issue a takeover request to the driver if the duration of the equivalent occlusion reaches the current safe time threshold. The following braking unit is configured to, if the duration of the equivalent occlusion does not reach the current safety time threshold and the non-visual sensor detects that the vehicle in front is decelerating, downgrade the driver assistance function and / or issue a takeover request to the driver; or, determine an emergency safety time threshold corresponding to the deceleration of the vehicle in front, and downgrade the driver assistance function and / or issue a takeover request to the driver when the duration of the continuous deceleration of the vehicle in front reaches the emergency safety time threshold.

11. A vehicle equipped with a front-view camera and having an assist driving function, the vehicle comprising: processor; Memory used to store processor-executable instructions; The processor implements the method as described in any one of claims 1-9 by executing the executable instructions.

12. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1-9.

13. A computer program product comprising computer programs and / or instructions, characterized in that, When the computer program and / or instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1-9.

Images