ETC antenna control method and vehicle

Abstract

The invention discloses an ETC antenna control method and a vehicle, and belongs to the technical field of vehicle control. And the active intervention on the ETC rubbing behavior is realized. A mechanism is started in a high-incidence scene that the target vehicle is located at the entrance of the ETC channel, a dynamic intention is predicted and captured based on the trajectory of the vehicle motion information, and the accuracy of risk judgment is improved. Through multi-source data fusion of integrating a predicted trajectory and a lane relationship, risk assessment fits an actual scene. And finally, when the risk value exceeds the standard, the state of the ETC antenna is directly controlled, illegal vehicles are prevented from overtaking, and normal operation of the toll collection system and rights and interests of legal vehicle owners are guaranteed.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a control method for an ETC antenna and a vehicle within the field of vehicle control technology. Background Technology

[0002] Electronic Toll Collection (ETC) systems achieve automatic toll deduction through dedicated short-range communication between roadside units (RSUs) and on-board units (OBUs), improving traffic efficiency. However, at ETC lane entrances, "ETC fraud" frequently occurs: vehicles without ETC or with malfunctioning OBUs closely follow or suddenly cut in front of vehicles paying tolls, rushing through after the OBU and RSU transaction is successful and the barrier rises. This results in financial losses and traffic delays for legitimate vehicle owners and easily causes lane congestion.

[0003] Currently, the main approach to addressing this issue relies on tollbooth cameras for post-event license plate recognition and retrieval, which has a long processing time.

[0004] Therefore, how to proactively assess and intervene in the risks of "ETC fraud" is a hot research topic. Summary of the Invention

[0005] This application provides a method for controlling an ETC antenna and a vehicle, which can detect potential ETC theft risks and actively control the ETC antenna status to prevent ETC theft, protect the rights and interests of legitimate vehicle owners, and improve the efficiency and security of ETC lanes. The technical solution is as follows: On the one hand, a control method for an ETC antenna is provided, the method comprising: When the target vehicle is at the entrance of the ETC lane, based on the vehicle movement information of suspicious vehicles around the target vehicle, the first predicted driving trajectory of the suspicious vehicle is determined, and the suspicious vehicle is a vehicle that has the risk of using ETC without authorization. Based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle, the risk value of the suspicious vehicle is determined; If the risk value of the suspected vehicle is greater than or equal to the first risk value threshold, the ETC antenna of the target vehicle shall be turned off or put into sleep mode.

[0006] On the one hand, a control device for an ETC antenna is provided, the device comprising: The trajectory determination module is used to determine the first predicted driving trajectory of a suspicious vehicle based on the vehicle movement information of suspicious vehicles around the target vehicle when the target vehicle is at the entrance of the ETC lane. The suspicious vehicle is a vehicle that has the risk of using ETC without authorization. The risk value determination module is used to determine the risk value of the suspicious vehicle based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle. The antenna control module is used to control the ETC antenna of the target vehicle to be turned off or put into sleep mode when the risk value of the suspected vehicle is greater than or equal to a first risk value threshold.

[0007] In one possible implementation, the trajectory determination module is used to determine the lateral and longitudinal movement trajectories of the suspicious vehicle relative to the target vehicle within a preset future time period based on the vehicle motion information of the suspicious vehicle; and to generate the first predicted driving trajectory based on the lateral and longitudinal movement trajectories.

[0008] In one possible implementation, the trajectory determination module is configured to determine the lateral movement trajectory of the suspicious vehicle relative to the target vehicle within the future preset time period based on the current lateral position, current lateral speed, and current heading angle in the vehicle motion information; and to determine the longitudinal movement trajectory of the suspicious vehicle relative to the target vehicle within the future preset time period based on the current longitudinal position, current longitudinal speed, and current heading angle in the vehicle motion information.

[0009] In one possible implementation, the risk value determination module is configured to: determine the predicted minimum lateral distance between the suspicious vehicle and the target vehicle based on the first predicted driving trajectory and the second predicted driving trajectory; determine the predicted lane intrusion time and predicted lane intrusion angle of the suspicious vehicle based on the first predicted driving trajectory and the lane in which the target vehicle is located; and determine the risk value of the suspicious vehicle based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle.

[0010] In one possible implementation, the risk value determination module is used to acquire a plurality of first trajectory points on the first predicted driving trajectory and a plurality of second trajectory points on the second predicted driving trajectory; determine the lateral distance between each of the plurality of first trajectory points and a second trajectory point with a corresponding timestamp among the plurality of second trajectory points; and select the minimum value from the determined plurality of lateral distances as the predicted minimum lateral distance.

[0011] In one possible implementation, the risk value determination module is used to determine the predicted intersection point of the first predicted driving trajectory and the lane boundary line of the lane where the target vehicle is located; determine the time required for the suspicious vehicle to reach the predicted intersection point as the predicted lane intrusion time; and determine the angle between the first predicted driving trajectory and the predicted intersection point and the lane as the predicted lane intrusion angle.

[0012] In one possible implementation, the risk value determination module is used to normalize the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle to obtain a distance risk factor corresponding to the predicted minimum lateral distance, a time risk factor corresponding to the predicted lane intrusion time, and an angle risk factor corresponding to the predicted lane intrusion angle; weightedly fuse the distance risk factor, the time risk factor, and the angle risk factor to obtain a comprehensive risk value; and determine the comprehensive risk value as the risk value of the suspicious vehicle.

[0013] In one possible implementation, the antenna control module is configured to determine whether the first predicted driving trajectory intersects with a virtual connecting line when the risk value of the suspected vehicle is greater than or equal to a risk value threshold. The virtual connecting line is obtained by extending a preset distance from the target vehicle in the direction of its lane. When the first predicted driving trajectory intersects with the virtual connecting line, the module controls the ETC antenna of the target vehicle to be turned off or put into sleep mode.

[0014] In one possible implementation, the device further includes: The alert module is used to trigger a first alert message when the risk value of the suspicious vehicle is greater than or equal to a second risk value threshold and less than a first risk threshold. The first alert message is used to prompt attention to the suspicious vehicle. When the risk value of the suspicious vehicle is less than the second risk value threshold, no alert is triggered.

[0015] In one possible implementation, the control module is further configured to control the ETC antenna of the target vehicle to turn on or off sleep mode and trigger a second prompt message when the risk value of the suspicious vehicle becomes less than the first risk value threshold and continues for a preset duration. The second prompt message is used to prompt the suspicious vehicle to give up trying to use the ETC.

[0016] In one possible implementation, the device further includes: A suspicious vehicle identification module is used to determine the relative position and relative speed between a candidate vehicle and a target vehicle when there are candidate vehicles around a target vehicle; and to determine the candidate vehicle as a suspicious vehicle when the relative position between the candidate vehicle and the target vehicle meets a first preset condition and the relative speed meets a second preset condition.

[0017] In one possible implementation, the device further includes: The candidate vehicle recognition module is used to acquire first visual perception information around the target vehicle; based on the first visual perception information, it determines whether there are candidate vehicles around the target vehicle.

[0018] In one possible implementation, the device further includes: The location recognition module is used to acquire the vehicle position of the target vehicle and / or second visual perception information in front of the target vehicle; based on the vehicle position and / or the second visual perception information, it determines whether the target vehicle is at the entrance of the ETC lane.

[0019] On one hand, a vehicle is provided, the vehicle including one or more processors and one or more memories, the one or more memories storing at least one piece of program code, the program code being loaded and executed by the one or more processors to implement the control method of the ETC antenna.

[0020] On one hand, a computer-readable storage medium is provided, wherein at least one piece of program code is stored in the computer-readable storage medium, the program code being loaded and executed by a processor to implement the control method of the ETC antenna.

[0021] The technical solution provided in this application enables proactive intervention against unauthorized ETC (Electronic Toll Collection) use. The mechanism is activated when the target vehicle is at the entrance of an ETC lane, a high-risk scenario. Based on vehicle movement information, trajectory prediction captures dynamic intent, improving the accuracy of risk assessment. By integrating multi-source data on predicted trajectories and lane relationships, the risk assessment closely reflects the actual scenario. Ultimately, when the risk value exceeds the limit, the ETC antenna status is directly controlled to prevent illegal vehicles from overtaking, ensuring the normal operation of the toll collection system and protecting the rights of legitimate vehicle owners. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the implementation environment of an ETC antenna control method provided in an embodiment of this application; Figure 2 This is a flowchart of a control method for an ETC antenna provided in an embodiment of this application; Figure 3This is a flowchart of another ETC antenna control method provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of a control device for an ETC antenna provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application. Detailed Implementation

[0023] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0024] In the following text, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features reflected. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0025] In order to illustrate the technical solutions provided in the embodiments of this application, some terms involved in the embodiments of this application will be introduced below.

[0026] ETC (Electronic Toll Collection): Electronic toll collection system.

[0027] OBU (On-Board Unit): An on-board electronic tag installed inside the vehicle for communication with the RSU.

[0028] RSU (Road-Side Unit): A roadside unit installed above or to the side of the tollbooth lane for communication with the OBU.

[0029] 5.8GHz DSRC: A dedicated short-range communication protocol for the 5.8GHz frequency band, which is the communication foundation of the ETC system.

[0030] ETC Freeloading: refers to the act of another vehicle using a successful ETC transaction of another vehicle to illegally pass through a toll station.

[0031] The Advanced Driver Assistance System (ADAS Perception System) includes surround-view cameras, millimeter-wave radar, and ultrasonic radar, which are used to perceive the 360-degree environment around the vehicle.

[0032] During the operation of the Electronic Toll Collection (ETC) system, when a target vehicle is located at the entrance area of ​​an ETC lane, there are instances where vehicles without onboard units or with malfunctioning onboard units may cut in front of or closely follow vehicles paying tolls, attempting to pass through after the transaction between the roadside unit and the onboard unit is completed. This phenomenon reduces the passage efficiency of legitimate vehicles, makes it difficult for the toll collection system to accurately identify actual passing vehicles, and, due to its reliance on post-event license plate recognition and collection mechanisms, introduces a significant delay in risk intervention.

[0033] For example, at the entrance of an ETC lane, a vehicle without an onboard unit (OI) experiences a sudden change in its lateral acceleration and longitudinal velocity after the target vehicle enters the transaction area, based on real-time vehicle motion information. The predicted trajectory indicates that the vehicle will cut into the target vehicle's lane. Furthermore, the target vehicle's OI communicates with the roadside unit, triggering the barrier to lift. At this point, the vehicle without an OI takes advantage of this window to rush through, causing the target vehicle to require manual resetting due to traffic interruption, and disrupting lane traffic order.

[0034] If the above problems are not resolved, the continued occurrence of ETC fraud will reduce the accuracy of toll system transactions, increase the complexity of post-event recovery management, and frequent traffic interruptions will reduce the overall capacity of ETC lanes, affecting the operational efficiency of the highway network and harming the interests of car owners.

[0035] The implementation environment of the embodiments of this application is described below. See also... Figure 1 The implementation environment of the ETC antenna control method provided in this application embodiment includes a vehicle controller 101 and an ETC antenna 102.

[0036] The vehicle controller 101 is a terminal installed on the vehicle. The vehicle controller 101 can acquire and process relevant information. In this embodiment, the vehicle controller 101 can acquire and process vehicle motion information. The vehicle controller 101 is communicatively connected to the ETC antenna 102. The vehicle controller 101 can control whether the ETC antenna 102 can perform external communication, that is, whether it can communicate with the roadside unit of the toll station.

[0037] In this embodiment of the application, the ETC antenna can be implemented as a 5.8GHz DSRC antenna for the OBU.

[0038] After introducing the implementation environment of the embodiments of this application, the application scenarios of the technical solutions provided by the embodiments of this application will be introduced below.

[0039] The technical solution provided in this application can be applied to vehicles equipped with ETC and driver assistance systems. By adopting the technical solution provided in this application, vehicles that use ETC without authorization can be identified and actively defended against, thus protecting the interests of vehicle owners and improving traffic efficiency.

[0040] After introducing the implementation environment and application scenarios of the embodiments of this application, the technical solutions provided by the embodiments of this application are described below. (See also...) Figure 2 Taking the vehicle controller as the executing entity as an example, the method includes the following steps.

[0041] 201. When the target vehicle is at the entrance of the ETC lane, the vehicle controller determines the first predicted driving trajectory of the suspicious vehicle based on the vehicle movement information of the suspicious vehicles around the target vehicle. The suspicious vehicle is a vehicle that has the risk of using ETC without authorization.

[0042] Suspicious vehicles refer to those posing a risk of unauthorized ETC (Electronic Toll Collection) use. They can be identified based on the relative position and speed of the vehicle and the target vehicle, or by analyzing the rate of change in the distance between them using onboard radar. The first predicted driving trajectory is the trajectory obtained after predicting the trajectory of the suspicious vehicle, used for a dynamic quantitative assessment of the suspicious vehicle's future movement trend.

[0043] 202. The vehicle controller determines the risk value of the suspicious vehicle based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle.

[0044] The target vehicle's lane is its current lane, representing its actual location. The second predicted driving trajectory refers to the predicted path of the target vehicle within a preset future time period, based on its current movement and lane. The risk value is used to comprehensively measure the degree of risk of suspected ETC fraud; a high risk value indicates a greater risk of suspicious vehicles using ETC, while a low risk value indicates a lower risk.

[0045] 203. If the risk value of the suspicious vehicle is greater than or equal to the first risk value threshold, the vehicle controller controls the ETC antenna of the target vehicle to be turned off or put into sleep mode.

[0046] Specifically, when the risk value exceeds a preset threshold, the ETC antenna can be controlled by software commands to trigger it into a low-power sleep state, or by hardware circuitry to cut off the antenna power supply and interrupt communication. This enables immediate blocking of illegal transactions. Therefore, the technical solution provided in this application, by predicting suspicious vehicle behavior in real time and dynamically assessing risk, initiates a targeted intervention mechanism when the target vehicle is at the ETC lane entrance, avoiding the processing delays associated with post-event license plate recognition and achieving proactive prevention of ETC-piracy behavior.

[0047] In steps 201-203 above, the vehicle controller collects vehicle movement information of suspicious vehicles around the target vehicle through the onboard sensor network. Suspicious vehicles refer to those that, after initial screening, pose a risk of unauthorized ETC use. Based on the acquired vehicle movement information, the vehicle controller predicts the first predicted driving trajectory of the suspicious vehicle in the future. This trajectory reflects the potential lateral and longitudinal movement trends of the vehicle, thereby detecting abnormal entry intentions. The vehicle controller combines the first predicted driving trajectory, the lane where the target vehicle is located, and the target vehicle's own second predicted driving trajectory to comprehensively assess the risk value of the suspicious vehicle. This risk value is generated through multi-dimensional data fusion, comprehensively quantifying the degree of vehicle approach and the urgency of the behavior. When the risk value reaches or exceeds a preset first risk value threshold, a control command is automatically triggered, causing the target vehicle's ETC antenna to enter a closed or dormant state, thereby interrupting the communication link with the roadside unit and blocking the illegal transaction process at its source.

[0048] The technical solution provided in this application enables proactive intervention against unauthorized ETC (Electronic Toll Collection) use. The mechanism is activated when the target vehicle is at the entrance of an ETC lane, a high-risk scenario. Based on vehicle movement information, trajectory prediction captures dynamic intent, improving the accuracy of risk assessment. By integrating multi-source data on predicted trajectories and lane relationships, the risk assessment closely reflects the actual scenario. Ultimately, when the risk value exceeds the limit, the ETC antenna status is directly controlled to prevent illegal vehicles from overtaking, ensuring the normal operation of the toll collection system and protecting the rights of legitimate vehicle owners.

[0049] It should be noted that steps 201-203 above are a simplified explanation of the ETC antenna control method provided in the embodiments of this application. The control method for the ETC antenna provided in the embodiments of this application will be described in more detail below with some examples. See [link to relevant documentation]. Figure 3 Taking the vehicle controller as the executing entity as an example, the method includes the following steps.

[0050] 301. When the target vehicle is at the entrance of the ETC lane, the vehicle controller determines whether there are candidate vehicles around the target vehicle.

[0051] The target vehicle refers to a vehicle equipped with an On-Board Unit (OBU) and employing the technical solution provided in this application's embodiments; it is the subject responsible for environmental perception, risk decision-making, and defensive actions. The ETC lane entrance refers to the specific area within a toll station defined by the lane designated for ETC vehicles, defined between its physical starting point and the starting point of the effective communication range of the Roadside Unit (RSU). Candidate vehicles refer to potential vehicles around the target vehicle that may pose a risk of unauthorized ETC use. Identifying candidate vehicles helps limit the processing scope and avoid invalid computations in vehicle-free environments.

[0052] In one possible implementation, the vehicle controller acquires first visual perception information about the vicinity of the target vehicle. Based on this first visual perception information, the vehicle controller determines whether there are candidate vehicles around the target vehicle.

[0053] The first visual perception information refers to real-time visual data of the surrounding environment of the target vehicle. This data can be collected using the vehicle's onboard surround-view camera system or captured by roadside visual sensor arrays at highway intersections. The first visual perception information can be implemented as continuous image frames or video streams. It is used to dynamically acquire the spatial distribution and motion status of traffic participants around the target vehicle. Determining whether candidate vehicles exist around the target vehicle involves identifying and locating vehicle targets based on the first visual perception information, used to filter out neighboring vehicles as candidates for subsequent risk assessment. In some embodiments, the first visual perception information can be collected using the vehicle's onboard radar; in this case, it can be implemented as a point cloud. Of course, the first visual perception information can also be multimodal visual perception information, that is, a composite information of images and point clouds; this application does not limit this aspect.

[0054] To illustrate the above implementation methods more clearly, several examples are provided below.

[0055] Example 1: The first visual perception information is a first video stream. The vehicle controller acquires the first video stream around the target vehicle through the onboard camera. The vehicle controller inputs the first video stream into a first target detection model, processes the first video stream through the first target detection model, and obtains a video stream recognition result. This video stream recognition result is used to indicate whether there are candidate vehicles in the first video stream.

[0056] The first object detection model is a model with vehicle detection capabilities. This model is obtained through multiple rounds of supervised training by sampling multiple sample images and the corresponding annotation results for each sample image. The annotation results are used to indicate whether a vehicle exists in the corresponding sample image. For example, the first object detection model is a YOLO (You Only Look Once) series model. This application embodiment does not limit the structure and training method of the first object detection model.

[0057] For example, the vehicle controller acquires a first video stream of the area surrounding the target vehicle using an onboard camera. The vehicle controller inputs this first video stream into a first target detection model, which extracts features from each video frame in the first video stream to obtain the video frame features for each frame. Based on these video frame features, the vehicle controller performs target detection using the first target detection model, obtaining the recognition results for each video frame. If the recognition results for M consecutive video frames (where M is a positive integer) indicate the presence of a vehicle, the vehicle controller determines that there are candidate vehicles surrounding the target vehicle. Otherwise, the vehicle controller determines that there are no candidate vehicles surrounding the target vehicle.

[0058] Wherein, M is set by technicians according to the actual situation, and this application embodiment does not limit it. The process of feature extraction of video frames by the first object detection model can use convolution or encoding based on an attention mechanism, and this application embodiment does not limit it.

[0059] Example 2: The first visual perception information is a first point cloud set. The vehicle controller acquires the first point cloud set around the target vehicle through the vehicle-mounted radar. The vehicle controller inputs the first point cloud set into a second target detection model, which processes the first point cloud set to obtain a point cloud set recognition result. This point cloud set recognition result is used to indicate whether there are candidate vehicles in the first point cloud set.

[0060] The second object detection model is a model with vehicle detection capabilities. This model is obtained through multiple rounds of supervised training, sampling multiple point clouds and their corresponding annotations. The annotations indicate whether a vehicle exists in the corresponding point cloud. For example, the second object detection model may be a PointNet series model. This application does not limit the structure or training method of the second object detection model.

[0061] For example, the vehicle controller acquires a first set of point clouds around the target vehicle using an onboard camera. The vehicle controller inputs this first set of point clouds into a second target detection model, which extracts features from each frame of the first set of point clouds to obtain the point cloud features of each frame. The vehicle controller then uses this second target detection model to perform target detection based on the point cloud features of each frame, obtaining the recognition results for each frame of point clouds. If the recognition results for M consecutive point clouds across multiple frames all indicate the presence of a vehicle, the vehicle controller determines that there are candidate vehicles around the target vehicle. Otherwise, the vehicle controller determines that there are no candidate vehicles around the target vehicle.

[0062] The second target detection model can use three-dimensional convolution or other feature extraction methods to extract features from the point cloud, and this application embodiment does not limit this.

[0063] Example 3: The first visual perception information includes the first video stream and the first point cloud set. The vehicle controller performs time synchronization and multimodal fusion on the first video stream and the first point cloud set to obtain a multimodal information stream.

[0064] The vehicle controller inputs the multimodal information stream into the third target detection model, which processes the multimodal information stream to obtain the video stream recognition result. This video stream recognition result is used to indicate whether there are candidate vehicles in the multimodal information stream.

[0065] The third object detection model is a model with vehicle detection capabilities. This model is obtained through multiple rounds of supervised training, sampling multimodal information from multiple samples and the corresponding annotations for each sample's multimodal information. The annotations indicate whether a vehicle exists in the corresponding sample's multimodal information. For example, the third object detection model may be a YOLO series model. This application does not limit the structure or training method of the third object detection model.

[0066] For example, the vehicle controller acquires a multimodal information stream around the target vehicle using an onboard camera. This multimodal information stream is then input into a third object detection model, which extracts features from each multimodal information segment to obtain the multimodal information features of each segment. Based on these features, the vehicle controller performs object detection using the third object detection model, obtaining the recognition results for each multimodal information segment. If the recognition results for M consecutive multimodal information segments indicate the presence of a vehicle, the vehicle controller determines that a candidate vehicle exists around the target vehicle. Otherwise, the vehicle controller determines that no candidate vehicle exists around the target vehicle.

[0067] The third object detection model can use convolution or encoding based on an attention mechanism to extract features from multimodal information, and this application does not limit this.

[0068] It should be noted that the vehicle controller can select any of the above methods to determine whether there are candidate vehicles around the target vehicle based on the hardware configuration of the target vehicle, and this application embodiment does not limit this.

[0069] Optionally, prior to step 301, the vehicle controller may also determine whether the target vehicle is at the entrance of the ETC lane by performing the following steps.

[0070] In one possible implementation, the vehicle controller acquires the vehicle position of the target vehicle and / or second visual perception information in front of the target vehicle. Based on the vehicle position and / or the second visual perception information, the vehicle controller determines whether the target vehicle is at the entrance of the ETC lane.

[0071] Among these, vehicle location refers to the coordinates of the target vehicle in geographic space, providing macro-area coverage to quickly confirm whether the vehicle is approaching the toll station's geographical area. Secondary visual perception information refers to image or video data of the environment in front of the target vehicle, used to obtain physical details specific to the ETC lane, such as roadside unit markings, lane markings, or barrier structures. Determining whether the target vehicle is at the ETC lane entrance means judging whether the vehicle is located in the starting area of ​​the ETC toll lane, ensuring that risk assessment is only initiated in actual entrance scenarios, avoiding waste of computational resources.

[0072] Under this implementation method, it is possible to avoid misjudgment of the entrance status due to environmental interference to a certain extent, and ensure that the risk assessment method provided in this application embodiment is activated only in real ETC lane entrance scenarios, thereby reducing the probability of system false triggering and improving the reliability of ETC-grabbing behavior detection.

[0073] To illustrate the above implementation methods more clearly, several examples are provided below.

[0074] Example 1: The vehicle controller obtains the target vehicle's location through the target vehicle's positioning system. Based on this vehicle location, the vehicle controller determines whether the target vehicle is at the entrance of the ETC lane.

[0075] For example, the vehicle controller obtains the target vehicle's location through the target vehicle's positioning system. If the target vehicle's location indicates that it is in the entrance area of ​​any toll station, the vehicle controller determines that the target vehicle is at the entrance of an ETC lane. If the target vehicle's location indicates that it is not in the entrance area of ​​any toll station, the vehicle controller determines that the target vehicle is not at the entrance of an ETC lane.

[0076] Due to limitations in the accuracy of the positioning system, when determining that the target vehicle is in the entrance area of ​​a toll station, it can be directly determined that the target vehicle is at the entrance of an ETC lane. This low-precision determination method allows the technical solution provided in this application embodiment to be activated when needed. In some embodiments, the positioning system is a GPS positioning system or a BeiDou positioning system, and this application embodiment does not limit this.

[0077] Example 2: The vehicle controller acquires second visual information about the front of the target vehicle through the vehicle's onboard camera. The vehicle controller then uses a target recognition model to identify this second visual information to determine whether the target vehicle is at the entrance of the ETC lane.

[0078] For example, the second visual information is a second video stream. The vehicle controller acquires this second video stream in front of the target vehicle through the onboard camera. The vehicle controller inputs this second video stream into a target recognition model, which extracts features from the second video stream to obtain its video stream features. The vehicle controller then uses this target recognition model to perform fully connected processing and normalization on the video stream features to obtain a classification value for the second video stream. If the classification value is greater than or equal to a classification threshold, the vehicle controller determines that the target vehicle is at the entrance of an ETC lane. If the classification value is less than the classification threshold, the vehicle controller determines that the target vehicle is not at the entrance of an ETC lane.

[0079] The classification threshold is set by technicians according to the actual situation, and this application embodiment does not limit it. The target recognition model is a binary classification model, which can be used to classify video streams, that is, to determine whether the video stream is a video stream captured at the entrance of the ETC lane. The target recognition model can be any structure of a video stream-based binary classification model, and this application embodiment does not limit it.

[0080] Example 3: The vehicle controller acquires the vehicle position of the target vehicle and second visual perception information in front of the target vehicle. Based on the vehicle position and the second visual perception information, the vehicle controller determines whether the target vehicle is at the entrance of the ETC lane.

[0081] For example, the vehicle controller obtains the target vehicle's location through its positioning system and acquires second visual information about the front of the target vehicle through its onboard camera. If the target vehicle's location indicates that it is in the entrance area of ​​any toll station, the vehicle controller inputs this second video stream into a target recognition model. The model extracts features from the second video stream to obtain its video stream features. The vehicle controller then uses this target recognition model to perform fully connected processing and normalization on the video stream features to obtain a classification value for the second video stream. If the classification value is greater than or equal to a classification threshold, the vehicle controller determines that the target vehicle is at the entrance of an ETC lane. If the classification value is less than the classification threshold, the vehicle controller determines that the target vehicle is not at the entrance of an ETC lane. If the target vehicle's location indicates that it is not in the entrance area of ​​any toll station, the vehicle controller determines that the target vehicle is not at the entrance of an ETC lane.

[0082] Under the above implementation method, the comprehensive ETC lane entrance judgment is achieved by combining vehicle location and second visual information, and the judgment accuracy is relatively high.

[0083] 302. When there are candidate vehicles around the target vehicle, the vehicle controller determines the relative position and relative speed between the candidate vehicle and the target vehicle.

[0084] In practical applications, there may be one or more candidate vehicles. Since the processing methods for multiple candidate vehicles and those for a single candidate vehicle belong to the same inventive concept, for ease of explanation, the following description will use one candidate vehicle as an example. Relative position refers to the spatial distance between the candidate vehicle and the target vehicle in the lateral and longitudinal directions, used to reflect whether the vehicles are in a typical ETC-piloting scenario. In this embodiment, relative position includes relative lane position and relative vehicle position. The relative lane position indicates whether the candidate vehicle is in an adjacent lane to the target vehicle, and the relative vehicle position indicates its location relative to the target vehicle. Relative speed refers to the relative speed between the candidate vehicle and the target vehicle, used to capture abnormal dynamic behavior characteristics.

[0085] In one possible implementation, when candidate vehicles exist around the target vehicle, the vehicle controller determines the relative position between the candidate vehicle and the target vehicle based on the onboard camera or radar of the candidate vehicle and first visual perception information. The vehicle controller then measures the speed of the candidate vehicle using the onboard radar of the target vehicle to obtain the relative speed between the candidate vehicle and the target vehicle.

[0086] The vehicle-mounted camera or radar, installed behind the target vehicle, has a specific field of view corresponding to at least one orientation of the target vehicle. By utilizing the vehicle-mounted camera or radar that identifies the candidate vehicle, preliminary positioning of the candidate vehicle can be achieved. For example, if the field of view of the vehicle-mounted camera that identifies the candidate vehicle covers the front of the target vehicle, the vehicle controller determines the relative vehicle position between the candidate vehicle and the target vehicle as front. If the field of view of the vehicle-mounted camera that identifies the candidate vehicle covers the left front of the target vehicle, the vehicle controller determines the relative vehicle position between the candidate vehicle and the target vehicle as left front.

[0087] The above implementation reduces the probability of misidentifying legitimate following behavior as suspicious, thus mitigating the risk of erroneously triggering the ETC antenna to shut down. Furthermore, by performing subsequent calculations only on candidate vehicles that meet both conditions, the real-time response capability and overall operational efficiency of risk assessment are improved.

[0088] For example, when candidate vehicles are present around the target vehicle, the vehicle controller determines the relative vehicle position between the candidate vehicle and the target vehicle based on the onboard camera or radar that identifies the candidate vehicle. When candidate vehicles are present around the target vehicle, the vehicle controller determines the relative lane position between the candidate vehicle and the target vehicle based on first visual perception information. The vehicle controller then uses the target vehicle's onboard radar to measure the speed of the candidate vehicle, obtaining the relative speed between the two vehicles.

[0089] In addition to determining the position of candidate vehicles solely based on the field of view of the onboard camera and radar, a comprehensive judgment can be made by combining the position of the candidate vehicle in the first visual perception information, thereby improving the accuracy of phase position determination. Determining the relative lane position using the first visual perception information includes identifying whether the candidate vehicle is located in a lane adjacent to the target vehicle; relative lane position includes being in an adjacent lane and not being in an adjacent lane. Using onboard radar for speed measurement is an inherent function of onboard radar and will not be elaborated upon here.

[0090] 303. The vehicle controller determines whether the relative position between the candidate vehicle and the target vehicle meets the first preset condition and whether the relative speed meets the second preset condition.

[0091] The first preset condition refers to the relative position meeting a specific threshold range, which can be set as a software-configurable threshold value. Its purpose is to filter out vehicles located in a threat zone. In this embodiment, the threat zone refers to a vehicle located in an adjacent lane and to the side or front of the target vehicle, excluding vehicles not located in an adjacent lane, and vehicles located in an adjacent lane but directly in front of, behind, or to the side of the target vehicle. Accordingly, the first condition means the vehicle's position is within the threat zone. The second preset condition refers to the vehicle speed meeting a specific threshold range, which can be dynamically adjusted based on the target vehicle's speed. Its purpose is to identify low-speed or unstable driving patterns. In this embodiment, the second preset condition means the relative vehicle speed is greater than or equal to a relative vehicle speed threshold. This relative vehicle speed threshold is set by a technician according to actual conditions, and this embodiment does not limit it.

[0092] In one possible implementation, if the relative position between the candidate vehicle and the target vehicle indicates that the candidate vehicle is in a threat area in the adjacent lane of the target vehicle, the vehicle controller determines that the relative position meets the first preset condition. Otherwise, the vehicle controller determines that the relative position does not meet the first preset condition. If the relative speed between the candidate vehicle and the target vehicle is greater than or equal to a relative speed threshold, the vehicle controller determines that the relative speed meets the second preset condition. Otherwise, the vehicle controller determines that the relative speed does not meet the second preset condition.

[0093] In some embodiments, the threat area includes a left-side threat area and a right-side threat area. The left-side threat area comprises a fan-shaped region extending 45° from the left side of the target vehicle to its left front, and correspondingly, the right-side threat area comprises a fan-shaped region extending 45° from the right side of the target vehicle to its right front. Of course, the aforementioned 45° is merely an example, and those skilled in the art can adjust the angle according to actual circumstances; this application does not limit this.

[0094] 304. If the relative position between the candidate vehicle and the target vehicle meets the first preset condition and the relative speed meets the second preset condition, the vehicle controller determines that the candidate vehicle is a suspicious vehicle, and the suspicious vehicle is a vehicle with the risk of using ETC fraudulently.

[0095] 305. The vehicle controller determines the first predicted driving trajectory of the suspicious vehicle based on the vehicle movement information of the suspicious vehicles around the target vehicle.

[0096] The first predicted driving trajectory is the driving trajectory obtained after predicting the trajectory of the suspicious vehicle, which is used for dynamic quantitative assessment of the future movement trend of the suspicious vehicle.

[0097] In one possible implementation, the vehicle controller determines the lateral and longitudinal movement trajectories of the suspected vehicle relative to the target vehicle within a preset future time period based on the vehicle motion information of the suspected vehicle. The vehicle controller then generates the first predicted driving trajectory based on the lateral and longitudinal movement trajectories.

[0098] The lateral movement trajectory refers to the positional change trajectory of a suspicious vehicle in the direction perpendicular to the lane. It reflects the lateral dynamic characteristics of the suspicious vehicle, avoiding the generality of overall trajectory prediction and thus capturing lateral entry intentions. The longitudinal movement trajectory refers to the positional change trajectory of a suspicious vehicle along the lane direction, used to identify the acceleration or deceleration behavior of the suspicious vehicle, providing dynamic evidence for risk assessment. Generating the first predicted driving trajectory specifically involves synthesizing the lateral and longitudinal movement trajectories using coordinates, for example, through Cartesian coordinate transformation. The purpose is to integrate the dynamic information of both dimensions, improving the continuity and completeness of the predicted trajectory and avoiding the one-sidedness of single-dimensional prediction.

[0099] In the above implementation, the trajectory of a suspicious vehicle is decomposed into two independent dimensions, lateral and longitudinal, for modeling. That is, the lateral and longitudinal trajectories are calculated separately based on the vehicle's motion information, and then combined to generate a first predicted driving trajectory. This decomposition modeling method can specifically address the dynamic characteristics of suspicious vehicles in different directions. The lateral trajectory focuses on lateral cutting-in risks, while the longitudinal trajectory focuses on following behavior. Dimensional separation avoids ambiguity in the overall prediction, thereby improving the accuracy of trajectory prediction. The generated first predicted driving trajectory provides a highly accurate input for subsequent risk value calculations, improving the reliability of risk assessment.

[0100] The above implementation methods improve the accuracy of suspicious vehicle trajectory prediction and reduce prediction deviation, which in turn improves the reliability of subsequent risk value calculation and avoids, to some extent, the miscontrol or omission of ETC antennas.

[0101] To provide a clearer explanation of the above embodiments, the following description is divided into several parts.

[0102] Part 1: The vehicle controller determines the lateral and longitudinal movement trajectories of the suspicious vehicle relative to the target vehicle within a preset time period based on the vehicle motion information of the suspicious vehicle.

[0103] In one possible implementation, the vehicle controller determines the lateral trajectory of the suspicious vehicle relative to the target vehicle within a predetermined future time period based on the current lateral position, current lateral velocity, and current heading angle in the vehicle motion information. The vehicle controller also determines the longitudinal trajectory of the suspicious vehicle relative to the target vehicle within the predetermined future time period based on the current longitudinal position, current longitudinal velocity, and current heading angle in the vehicle motion information.

[0104] The current lateral position refers to the instantaneous spatial coordinate reference point of the suspicious vehicle in the lateral direction, providing an initial spatial reference for lateral trajectory prediction. The current lateral velocity refers to the instantaneous velocity component of the suspicious vehicle in the lateral direction, dynamically reflecting the trend of lateral movement. The current heading angle is the angle between the suspicious vehicle's direction of travel and the lane reference coordinate system, capturing the vehicle's steering intention and directional change trend. The current longitudinal position refers to the instantaneous spatial coordinate reference point of the suspicious vehicle in the longitudinal direction, establishing the initial reference for the longitudinal trajectory. The current longitudinal velocity refers to the instantaneous velocity component of the suspicious vehicle in the longitudinal direction, characterizing the continuous state of the vehicle's forward or deceleration.

[0105] The principle of the above implementation method is as follows: by using the current lateral position as a spatial reference point, combining the current lateral velocity to reflect the dynamic trend of lateral movement, and introducing the current heading angle to capture the vehicle's turning intention, these three factors work together to construct a predictive model of the lateral movement trajectory, enabling the trajectory to simulate the curvature changes when a suspicious vehicle suddenly changes lanes. Simultaneously, the current longitudinal position ensures the initial accuracy of the longitudinal trajectory, the current longitudinal velocity characterizes the vehicle's continuous movement within the lane, and the current heading angle supplements the directional component in oblique driving scenarios, collectively forming a comprehensive representation of longitudinal movement. This parameter combination mechanism addresses the instantaneous and directional sensitivity of vehicle movement at ETC lane entrances. By integrating the multi-dimensional dynamic characteristics of position, velocity, and heading angle, it overcomes the insufficient accuracy problem caused by single parameters in traditional trajectory prediction, thus providing a basis for subsequent risk value calculation.

[0106] For example, the current lateral and longitudinal positions are obtained through clustering of point clouds collected by the vehicle-mounted radar; the current lateral and longitudinal velocities are determined based on the Doppler frequency shift principle; and the current heading angle is obtained through the tangent direction of the trajectory fitted from continuous position points. The vehicle controller uses a polynomial fitting algorithm to generate a lateral motion trajectory using the current lateral position, current lateral velocity, and current heading angle as input parameters, and simultaneously generates a longitudinal motion trajectory using the current longitudinal position, current longitudinal velocity, and current heading angle as input parameters.

[0107] For example, the vehicle controller can generate the lateral motion trajectory using the following formula (1) and the longitudinal motion trajectory using the following formula (2).

[0108] X pred (t) = X current + V·cos(θ)·t (1) Y pred (t) = Y current + V·sin(θ)·t (2) Among them, X current Y is the x-coordinate of the suspicious vehicle in the world coordinate system. current X is the ordinate of the suspicious vehicle in the world coordinate system. pred (t) is the predicted abscissa of the suspicious vehicle in the world coordinate system, a function of time t, which can represent the lateral movement trajectory. pred (t) is the predicted ordinate of the suspicious vehicle in the world coordinate system, which is a function of time t and can represent the longitudinal trajectory. V is the speed of the suspicious vehicle in the world coordinate system, θ is the heading angle of the suspicious vehicle, and t is the time window for trajectory prediction (the aforementioned future time period), which is usually 0.5s to 2.0s.

[0109] The second part is that the vehicle controller generates the first predicted driving trajectory based on the lateral motion trajectory and the longitudinal motion trajectory.

[0110] In one possible implementation, the vehicle controller fuses the lateral motion trajectory and the longitudinal motion trajectory to obtain the first predicted driving trajectory.

[0111] Trajectory fusion can be achieved through coordinate synthesis in a Cartesian coordinate system.

[0112] 306. The vehicle controller determines the risk value of the suspicious vehicle based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle.

[0113] The target vehicle's lane is its current lane, representing its actual location. The second predicted driving trajectory refers to the predicted path of the target vehicle within a preset future time period, based on its current movement and lane. The risk value is used to comprehensively measure the degree of risk of suspected ETC fraud; a high risk value indicates a greater risk of suspicious vehicles using ETC, while a low risk value indicates a lower risk.

[0114] In one possible implementation, the vehicle controller determines the predicted minimum lateral distance between the suspicious vehicle and the target vehicle based on the first predicted driving trajectory and the second predicted driving trajectory. The vehicle controller then determines the predicted lane intrusion time and predicted lane intrusion angle of the suspicious vehicle based on the first predicted driving trajectory and the lane in which the target vehicle is located. Finally, the vehicle controller determines the risk value of the suspicious vehicle based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle.

[0115] Among these parameters, the predicted minimum lateral distance refers to the minimum lateral distance that the suspicious vehicle and the target vehicle may reach along their predicted trajectories, quantifying the spatial proximity of the two vehicles in the lateral direction. The predicted lane intrusion time refers to the expected time when the suspicious vehicle will enter the lane occupied by the target vehicle, used to assess the time urgency of the intrusion. The predicted lane intrusion angle is the geometric angle between the predicted trajectory of the suspicious vehicle and the lane boundary line at the intersection, reflecting the steepness of the suspicious vehicle's entry. Determining the risk value based on the predicted minimum lateral distance, predicted lane intrusion time, and predicted lane intrusion angle means integrating these parameters into a comprehensive indicator that provides a multi-dimensional quantitative risk assessment.

[0116] The principle of the above implementation is as follows: First, the predicted minimum lateral distance is calculated based on the first and second predicted driving trajectories to capture the minimum spatial interval between the two vehicles in the lateral direction. Then, the potential lane intrusion area is located based on the boundary relationship between the first predicted driving trajectory and the lane where the target vehicle is located, and the predicted lane intrusion time and angle are derived accordingly, thus incorporating temporal dynamics and geometric features into the risk assessment. Finally, the predicted minimum lateral distance, predicted lane intrusion time, and predicted lane intrusion angle are comprehensively integrated to generate a risk value. These steps are executed sequentially and are interconnected, ensuring that the risk assessment not only relies on static distance comparisons but also incorporates the temporal urgency and steepness of the intrusion behavior, enabling the vehicle controller to distinguish between normal lane changes and high-risk ETC-grabbing behavior, thereby improving the accuracy of risk judgment.

[0117] The above implementation methods can more accurately identify ETC-piracy behavior, reduce the probability of misjudging normal lane changes as risky behavior, reduce the missed detection of high-risk ETC-piracy behavior, and improve the safety and efficiency of ETC lane passage.

[0118] To provide a clearer explanation of the above embodiments, the following description is divided into several parts.

[0119] The first part involves the vehicle controller determining the minimum predicted lateral distance between the suspicious vehicle and the target vehicle based on the first predicted driving trajectory and the second predicted driving trajectory.

[0120] In one possible implementation, the vehicle controller acquires a plurality of first trajectory points on the first predicted driving trajectory and a plurality of second trajectory points on the second predicted driving trajectory. The vehicle controller determines the lateral distance between each of the plurality of first trajectory points and a second trajectory point with a corresponding timestamp among the plurality of second trajectory points. The vehicle controller selects the minimum value from the determined plurality of lateral distances as the predicted minimum lateral distance.

[0121] The first trajectory point refers to a discrete sampling point on the predicted trajectory of the suspicious vehicle, which can convert a continuous trajectory into a quantifiable time-series data unit. The second trajectory point refers to a discrete sampling point on the predicted trajectory of the target vehicle, which can provide a motion reference benchmark for the target vehicle. The corresponding timestamp means that the first and second trajectory points have the same time identifier. This can be achieved using a system clock-synchronized timestamp or a virtual timestamp derived from a motion model, ensuring that the position comparison is strictly limited to the same moment. The lateral distance refers to the distance between the two vehicles in the direction perpendicular to the lane within the vehicle's motion plane, quantifying the degree of lateral approach between the two vehicles. The predicted minimum lateral distance is the minimum value among all time-synchronized lateral distances, focusing on the critical point in the dynamic approach process between the two vehicles. The smaller the predicted minimum lateral distance, the more likely the suspicious vehicle is to get infinitely close to the target vehicle, and the higher the risk.

[0122] The principle of the above implementation is as follows: First, multiple trajectory points on the first and second predicted driving trajectories are acquired. These points represent the vehicle's position at different times. Second, based on a timestamp alignment mechanism, each first trajectory point is matched with a corresponding second trajectory point at a different timestamp, and the lateral distance between them is calculated, thereby eliminating time misalignment caused by differences in vehicle speed. Finally, by iterating through all time-synchronized lateral distance values, the minimum value is selected as the predicted minimum lateral distance. This value directly reflects the closest lateral distance between the two vehicles during dynamic driving. Through this time-synchronized trajectory point matching logic, it is ensured that the lateral distance calculation strictly corresponds to the positional relationship at the same time, avoiding the problem of inflated or understated distances caused by time misalignment, and enabling the predicted minimum lateral distance to accurately represent the dynamic relative position.

[0123] The above implementation method can obtain a more accurate prediction of the minimum lateral distance, thus providing a reliable basis for risk value judgment. This avoids misjudgment of ETC risk due to inaccurate distance calculation, reduces the phenomenon of subsequent ETC antennas being mistakenly turned off or not turned off in time, and ensures the passage efficiency of legitimate vehicles and the reliability of the toll collection system.

[0124] The second part involves the vehicle controller determining the predicted lane intrusion time and predicted lane intrusion angle of the suspicious vehicle based on the first predicted driving trajectory and the lane where the target vehicle is located.

[0125] In one possible implementation, the vehicle controller determines the predicted intersection point between the first predicted driving trajectory and the lane boundary line of the lane where the target vehicle is located. The vehicle controller determines the time required for the suspicious vehicle to reach the predicted intersection point as the predicted lane intrusion time. The vehicle controller determines the angle between the first predicted driving trajectory and the lane at the predicted intersection point as the predicted lane intrusion angle.

[0126] The predicted intersection point refers to the geometric location where the first predicted driving trajectory intersects with the lane boundary line of the target vehicle's lane. It is used to pinpoint the critical position where a suspicious vehicle might cut into the target lane, avoiding positioning errors caused by trajectory estimation deviations. The predicted lane intrusion time refers to the time required for the suspicious vehicle to move from its current position to the predicted intersection point. It is used to correlate the vehicle's real-time movement characteristics with the urgency of the risk; the smaller the predicted lane intrusion time, the shorter the predicted lane intrusion time, indicating imminent danger and a shorter reaction time. The predicted lane intrusion angle is the geometric angle between the first predicted driving trajectory at the predicted intersection point and the lane direction. It is used to enhance the ability to distinguish between slight deviations and sudden cut-in behaviors. The larger the predicted lane intrusion angle, the less likely the other vehicle is driving parallel, but rather has a clear lateral cutting motive, and the more obvious the intention.

[0127] The principle of the above implementation method is as follows: First, the predicted intersection point of the first predicted driving trajectory and the lane boundary line is determined to capture the critical position where a suspicious vehicle may cut into the target lane. Then, based on the motion state of the suspicious vehicle, the time required to reach this intersection point is calculated as the predicted lane intrusion time, directly reflecting the dynamic urgency of the risk. Simultaneously, the geometric angle between the first predicted driving trajectory and the lane is determined at the intersection point as the predicted lane intrusion angle, characterizing the instantaneous directional characteristics of the cutting behavior. These parameters together constitute the input basis for risk assessment, thus forming a complete lane intrusion parameter quantification mechanism.

[0128] For example, the vehicle controller uses a geometric algorithm to determine the intersection of the first predicted driving trajectory and the lane boundary line of the target vehicle's lane as the predicted intersection point. The vehicle controller divides the lateral distance between the suspicious vehicle and the lane by the lateral speed of the suspicious vehicle to obtain the time to reach the predicted intersection point, and uses this time as the predicted lane intrusion time. At the predicted intersection point, the vehicle controller determines the predicted lane intrusion angle based on the lateral and longitudinal speeds of the suspicious vehicle; alternatively, the vehicle controller extracts the tangent direction of the first predicted driving trajectory and the lane direction vector, determines the angle between the tangent direction and the lane direction vector, and uses this angle as the predicted lane intrusion angle.

[0129] For example, the predicted lane intrusion time can be determined by the following formula (3), and the predicted lane intrusion angle can be determined by the following formula (4).

[0130] TTC lat = D lat curren / |V lat | (3) θ cut in = arctan(V lat / V lon (4) Among them, TTC lat D represents the predicted lane intrusion time. lat curren V represents the lateral distance between a suspicious vehicle and the lane. lat θ represents the lateral velocity of the suspicious vehicle. cut in V represents the predicted lane intrusion angle. lon Indicates the longitudinal speed of the suspicious vehicle.

[0131] Part Three: The vehicle controller determines the risk value of the suspicious vehicle based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle.

[0132] In one possible implementation, the vehicle controller normalizes the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle to obtain a distance risk factor corresponding to the predicted minimum lateral distance, a time risk factor corresponding to the predicted lane intrusion time, and an angle risk factor corresponding to the predicted lane intrusion angle. The vehicle controller then weights and fuses these three risk factors to obtain a comprehensive risk value. This comprehensive risk value is then determined as the risk value for the suspicious vehicle.

[0133] Normalization refers to converting raw parameters with different physical units and magnitudes into dimensionless values ​​of a uniform scale. This can be achieved using min-max normalization or sigmoid standardization methods, aiming to eliminate assessment biases caused by inconsistent units and ensure comparability between parameters. Distance risk factor is an indicator that quantifies the risk contribution of the minimum predicted lateral distance, used to convert physical distance into a standardized risk metric. Time risk factor and angle risk factor correspond to the risk quantification of predicted lane incursion time and angle, respectively, used to independently convert time and angle parameters into risk contribution values. Weighted fusion refers to assigning weight coefficients to each risk factor based on its importance in the ETC (Electronic Toll Collection) scenario, used to comprehensively consider the differentiated impact of multiple factors on risk.

[0134] The principle of the above implementation method is as follows: First, the predicted minimum lateral distance, predicted lane intrusion time, and predicted lane intrusion angle are normalized, converting the original parameters into dimensionless risk factors within the range of 0 to 1, making parameters with different physical meanings comparable. Then, the distance risk factor, time risk factor, and angle risk factor are weighted and summed according to preset weighting coefficients to generate a comprehensive risk value that reflects multi-dimensional risks. Finally, this comprehensive risk value is directly used as the risk value output for suspicious vehicles, providing a unified and reliable risk quantification basis for the control decisions of the ETC antenna, thus forming a complete technical chain from parameter standardization to comprehensive risk assessment.

[0135] The above implementation method can solve the problem of assessment distortion caused by unit differences in multi-dimensional risk parameters, achieve accurate quantification of risk level, and thus provide a reliable decision basis for the control of ETC antenna, so as to shut down the antenna in time when the risk of ETC fraud is too high to prevent illegal behavior.

[0136] For example, the vehicle controller normalizes the predicted minimum lateral distance using a lateral distance risk threshold to obtain a distance risk factor corresponding to that predicted minimum lateral distance. The vehicle controller normalizes the predicted lane intrusion time using a time risk threshold to obtain a time risk factor corresponding to that predicted lane intrusion time. The vehicle controller normalizes the predicted lane intrusion angle using the maximum effective cut-in angle to obtain an angle risk factor corresponding to that predicted lane intrusion angle. The vehicle controller then weights and fuses these distance risk factors, time risk factors, and angle risk factors to obtain a comprehensive risk value. The vehicle controller determines this comprehensive risk value as the risk value for the suspicious vehicle.

[0137] For example, the vehicle controller normalizes the predicted minimum lateral distance using the following formula (5), normalizes the predicted lane intrusion time using the following formula (6), normalizes the predicted lane intrusion angle using the following formula (7), and weights and fuses the distance risk factor, the time risk factor, and the angle risk factor using the following formula (8).

[0138] F1(D lat min ) = 1 / (1 + e^(k1·(D lat min -d0)))(5) F2(TTC lat ) = 1 / (1 + e^(k2·(TTC lat -t0))) (6) F3(θ cut in ) = min(θ cut in / θ max ,1.0)(7) R total =w1·F1(D lat min ) + w2·F2(TTC lat )+w3·F3(θ cut in (8) Among them, D lat min This represents the predicted minimum lateral distance, d0 represents the lateral distance risk threshold (e.g., 1.5m), t0 represents the time risk threshold (e.g., 2s), and θ... max This represents the maximum effective entry angle, for example, 30°. F1() and F2() represent the Sigmoid function, F3() represents the minimum-maximum normalization function, k1 and k2 are preset coefficients, and w1, w2, and w3 are also preset coefficients, obtained through training and calibration with a large amount of real vehicle data (e.g., w1=0.4, w2=0.4, w3=0.2), representing the importance of each parameter. R total This represents the overall risk value.

[0139] Optionally, after step 306, the vehicle controller may execute steps 307, 309, or 310 as appropriate.

[0140] 307. If the risk value of the suspicious vehicle is greater than or equal to the first risk value threshold, the vehicle controller controls the ETC antenna of the target vehicle to be turned off or put into sleep mode.

[0141] In this embodiment, when the risk value exceeds a preset threshold, the ETC antenna can be controlled by software commands to trigger it into a low-power sleep state, or by hardware circuitry to cut off the antenna power supply and interrupt communication. This allows for immediate blocking of illegal transactions. Therefore, the technical solution provided in this application, by predicting suspicious vehicle behavior in real time and dynamically assessing risk, initiates a targeted intervention mechanism when the target vehicle is at the ETC lane entrance, avoiding the processing delays associated with post-event license plate recognition and achieving proactive prevention of ETC-grabbing behavior. In some embodiments, the ETC antenna is a 5.8GHz DSRC antenna. After the ETC antenna is turned off or enters sleep mode, the RSU cannot complete a transaction with the target vehicle's OBU, the barrier will not lift, and vehicles attempting to use the ETC lane will fail.

[0142] In one possible implementation, if the risk value of the suspected vehicle is greater than or equal to a risk value threshold, the vehicle controller determines whether the first predicted driving trajectory intersects with a virtual connecting line, which is obtained by extending a preset distance from the target vehicle in the direction of its lane. If the first predicted driving trajectory intersects with the virtual connecting line, the vehicle controller controls the target vehicle's ETC antenna to turn off or go into sleep mode.

[0143] The risk threshold is a preset risk assessment benchmark used to judge the risk of suspicious vehicles attempting to use ETC (Electronic Toll Collection) without authorization. The virtual connection line is a virtual geometric reference line used to verify whether the trajectory of a suspicious vehicle encroaches on the target vehicle's passage area. It can be implemented as a straight line segment extending a reasonable distance from the target vehicle's current position along the lane centerline, aiming to transform abstract risk judgment into intuitive path conflict verification. Determining intersection refers to judging whether there is a geometric intersection point between the first predicted driving trajectory and the virtual connection line in a two-dimensional coordinate system, used to identify potential path conflicts. Controlling the ETC antenna to turn off or go into sleep mode means stopping the target vehicle's ETC antenna communication function or putting it into a low-power state, used to block interference from unauthorized ETC use on normal transactions.

[0144] The principle of the above implementation method is as follows: when the risk value of a suspicious vehicle reaches the risk threshold, the antenna is not directly shut down. Instead, a secondary verification is performed based on the geometric relationship between the first predicted driving trajectory and the virtual connecting line. The virtual connecting line extends a preset distance along the lane direction from the target vehicle. This design maps the expected movement range of the target vehicle at the entrance of the ETC lane, transforming the risk value judgment into a geometric problem of path conflict. By analyzing the relative positional relationship between the first predicted driving trajectory and the virtual connecting line, when the two intersect, it indicates that the suspicious vehicle is cutting into the target vehicle's passage path, posing a risk of actual transaction interference; conversely, it indicates that the suspicious vehicle is only in a side lane or deviating from the path, requiring no intervention. Based on this, the antenna shutdown or hibernation operation is only performed under the condition of intersection, ensuring that risk intervention measures are strictly limited to the critical moment when the suspicious vehicle is about to invade the target vehicle's transaction area. This retains the rapid screening function of the risk value threshold and filters out the risk of misjudgment through geometric verification, thus forming a complete risk decision-making closed loop.

[0145] The above implementation method avoids the problem of false triggering caused by the broad risk value threshold judgment, and enables the ETC antenna to work normally in harmless scenarios such as suspicious vehicles only being in the side lane or briefly deviating from the path. This ensures the continuity of ETC transactions for legitimate vehicles and reduces traffic interruptions and poor user experience.

[0146] For example, the virtual connecting line can be specifically implemented as a straight line segment extending from the center point of the rear axle of the target vehicle along the lane centerline, with the direction consistent with the lane direction. If the risk value of the suspected vehicle is greater than or equal to a risk value threshold, the vehicle controller discretizes the first predicted driving trajectory into multiple trajectory points and determines the perpendicular distance from each trajectory point to the virtual connecting line. If the minimum distance is less than a preset distance threshold, it is determined to be an intersection. After determining an intersection, the vehicle controller sends a shutdown command or a sleep command to the ETC antenna, causing the ETC antenna to shut down or enter a low-power state until the risk is eliminated.

[0147] Optionally, after step 307, the vehicle controller may execute step 308 as follows, depending on the actual situation.

[0148] 308. If the risk value of the suspicious vehicle becomes less than the first risk value threshold and continues for a preset duration, the vehicle controller controls the ETC antenna of the target vehicle to turn on or off and triggers a second prompt message, which is used to prompt the suspicious vehicle to give up using the ETC.

[0149] In this embodiment, the risk value of the suspicious vehicle is monitored and updated in real time. If the lateral distance between the suspicious vehicle and the target vehicle continues to increase, or if the suspicious vehicle leaves the threat area, the first predicted driving trajectory of the suspicious vehicle will change, and consequently, the risk value determined based on the first predicted driving trajectory will also change. By monitoring the risk value change trend of the suspicious vehicle in real time, when the risk value drops below the first risk value threshold and remains below it for a preset duration, the risk is substantially eliminated based on the dual verification of the dynamic change trend and the duration of time. Subsequently, the ETC antenna of the target vehicle is controlled to turn on or off, allowing the antenna function to promptly return to normal operation. Simultaneously, a second prompt message is triggered to remind the customer to avoid an unauthorized ETC use.

[0150] In one possible implementation, when the risk value of a suspected vehicle drops below a first risk value threshold and remains below the threshold for a preset duration, the vehicle controller automatically sends an activation command to the ETC antenna controller of the target vehicle to release the antenna from sleep mode and triggers a second prompt message on the display screen of the on-board unit, which reads "The attempt to use ETC has been abandoned. Please proceed normally."

[0151] In addition, when the target vehicle is detected leaving the toll station, the vehicle controller can also control the ETC antenna of the target vehicle to turn on or off to ensure that the ETC antenna is available in the future.

[0152] 309. If the risk value of the suspicious vehicle is greater than or equal to the second risk value threshold and less than the first risk threshold, the vehicle controller triggers a first prompt message, which is used to prompt attention to the suspicious vehicle.

[0153] The second risk threshold refers to the critical value used to define the safety threshold of the risk value. It can be a fixed value preset by the system or a value dynamically adjusted based on historical traffic data. This is used to avoid generating redundant prompts when the risk value is low, thereby reducing the cognitive burden on the user. The first prompt message refers to the prompting method to deliver warning information to the user. It can be implemented through visual prompts on the vehicle dashboard, voice prompts from the in-vehicle audio system, or seat vibration feedback. The purpose is to enable users to pay attention to the dynamics of suspicious vehicles through intuitive perception and provide a basis for decision-making to take preventive measures. By continuously monitoring the risk value of suspicious vehicles, when the risk value falls between the second risk threshold and the first risk threshold, the first prompt message is automatically triggered, thereby providing the user with a targeted warning in the early stage of risk accumulation.

[0154] In one possible implementation, when the risk value of a suspected vehicle is detected to be higher than a second risk value threshold but lower than a first risk value threshold, the vehicle controller controls the warning light on the dashboard to flash at a specific frequency, and at the same time plays a voice prompt "Pay attention to risky vehicles, ETC protection in progress" through the vehicle audio system, reminding the driver to observe the movement of the suspected vehicle through the rearview mirror.

[0155] The above implementation method can provide timely warnings when the risk of suspicious vehicles is at a moderate level, enabling drivers to take preventive actions such as slowing down or maintaining a safe distance in advance, thereby reducing the possibility of ETC fraud.

[0156] 310. If the risk value of the suspicious vehicle is less than the second risk value threshold, the vehicle controller will not trigger a prompt.

[0157] Specifically, when the risk value is lower than the second risk value threshold, the triggering of the prompt is suppressed, so that the prompt message is only generated when necessary.

[0158] In one possible implementation, if the risk value is lower than a second risk value threshold, the vehicle controller does not generate any prompts to maintain the simplicity of the driving environment.

[0159] By implementing the above methods, false alarms in low-risk scenarios are avoided, thereby improving reliability and user experience.

[0160] The technical solution provided in this application enables proactive intervention against unauthorized ETC (Electronic Toll Collection) use. The mechanism is activated when the target vehicle is at the entrance of an ETC lane, a high-risk scenario. Based on vehicle movement information, trajectory prediction captures dynamic intent, improving the accuracy of risk assessment. By integrating multi-source data on predicted trajectories and lane relationships, the risk assessment closely reflects the actual scenario. Ultimately, when the risk value exceeds the limit, the ETC antenna status is directly controlled to prevent illegal vehicles from overtaking, ensuring the normal operation of the toll collection system and protecting the rights of legitimate vehicle owners.

[0161] Figure 4 This is a schematic diagram of the structure of a control device for an ETC antenna provided in an embodiment of this application. See also... Figure 4 The device includes: a trajectory determination module 401, a risk value determination module 402, and an antenna control module 403.

[0162] The trajectory determination module 401 is used to determine the first predicted driving trajectory of a suspicious vehicle based on the vehicle movement information of suspicious vehicles around the target vehicle when the target vehicle is at the entrance of the ETC lane. The suspicious vehicle is a vehicle that has the risk of using ETC without authorization.

[0163] The risk value determination module 402 is used to determine the risk value of the suspicious vehicle based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle.

[0164] Antenna control module 403 is used to control the ETC antenna of the target vehicle to be turned off or put into sleep mode when the risk value of the suspicious vehicle is greater than or equal to a first risk value threshold.

[0165] In one possible implementation, the trajectory determination module 401 is used to determine the lateral and longitudinal movement trajectories of the suspicious vehicle relative to the target vehicle within a preset future time period, based on the vehicle motion information of the suspicious vehicle. Based on the lateral and longitudinal movement trajectories, the first predicted driving trajectory is generated.

[0166] In one possible implementation, the trajectory determination module 401 is used to determine the lateral movement trajectory of the suspicious vehicle relative to the target vehicle within a preset future time period based on the current lateral position, current lateral velocity, and current heading angle in the vehicle motion information. It also determines the longitudinal movement trajectory of the suspicious vehicle relative to the target vehicle within the preset future time period based on the current longitudinal position, current longitudinal velocity, and current heading angle in the vehicle motion information.

[0167] In one possible implementation, the risk value determination module 402 is used to determine the predicted minimum lateral distance between the suspicious vehicle and the target vehicle based on the first predicted driving trajectory and the second predicted driving trajectory. Based on the first predicted driving trajectory and the lane where the target vehicle is located, it determines the predicted lane intrusion time and predicted lane intrusion angle of the suspicious vehicle. Based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle, it determines the risk value of the suspicious vehicle.

[0168] In one possible implementation, the risk value determination module 402 is used to acquire a plurality of first trajectory points on the first predicted driving trajectory and a plurality of second trajectory points on the second predicted driving trajectory. It determines the lateral distance between each of the plurality of first trajectory points and a second trajectory point with a corresponding timestamp among the plurality of second trajectory points. From the determined lateral distances, the minimum value is selected as the predicted minimum lateral distance.

[0169] In one possible implementation, the risk value determination module 402 is used to determine the predicted intersection point of the first predicted driving trajectory and the lane boundary line of the lane where the target vehicle is located. The time required for the suspicious vehicle to reach the predicted intersection point is determined as the predicted lane intrusion time. The angle between the first predicted driving trajectory and the lane at the predicted intersection point is determined as the predicted lane intrusion angle.

[0170] In one possible implementation, the risk value determination module 402 is used to normalize the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle to obtain a distance risk factor corresponding to the predicted minimum lateral distance, a time risk factor corresponding to the predicted lane intrusion time, and an angle risk factor corresponding to the predicted lane intrusion angle. The distance risk factor, the time risk factor, and the angle risk factor are then weighted and fused to obtain a comprehensive risk value. This comprehensive risk value is determined as the risk value of the suspicious vehicle.

[0171] In one possible implementation, the antenna control module 403 is configured to determine whether the first predicted driving trajectory intersects with a virtual connecting line when the risk value of the suspected vehicle is greater than or equal to a risk value threshold. The virtual connecting line is obtained by extending a preset distance from the target vehicle towards its lane. If the first predicted driving trajectory intersects with the virtual connecting line, the module controls the target vehicle's ETC antenna to be turned off or put into sleep mode.

[0172] In one possible implementation, the device further includes: The alert module is used to trigger a first alert message when the risk value of a suspicious vehicle is greater than or equal to a second risk value threshold and less than a first risk threshold. This first alert message is used to prompt attention to the suspicious vehicle. If the risk value of the suspicious vehicle is less than the second risk value threshold, no alert is triggered.

[0173] In one possible implementation, the control module is further configured to control the ETC antenna of the target vehicle to turn on or off from sleep mode and trigger a second prompt message when the risk value of the suspicious vehicle becomes less than the first risk value threshold and continues for a preset duration. The second prompt message is used to prompt the suspicious vehicle to give up trying to use the ETC.

[0174] In one possible implementation, the device further includes: The suspicious vehicle identification module is used to determine the relative position and relative speed between a candidate vehicle and the target vehicle when candidate vehicles exist around the target vehicle. If the relative position between the candidate vehicle and the target vehicle meets a first preset condition, and the relative speed meets a second preset condition, the candidate vehicle is identified as a suspicious vehicle.

[0175] In one possible implementation, the device further includes: The candidate vehicle recognition module is used to acquire first visual perception information about the surroundings of the target vehicle. Based on this first visual perception information, it determines whether there are candidate vehicles around the target vehicle.

[0176] In one possible implementation, the device further includes: The location recognition module is used to acquire the vehicle position of the target vehicle and / or second visual perception information in front of the target vehicle. Based on the vehicle position and / or the second visual perception information, it determines whether the target vehicle is at the entrance of the ETC lane.

[0177] It should be noted that the control device for the ETC antenna provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the control device for the ETC antenna provided in the above embodiments and the control method embodiments for the ETC antenna belong to the same concept; the specific implementation process is detailed in the method embodiments and will not be repeated here.

[0178] This application also provides a vehicle. Figure 5 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0179] Typically, vehicle 500 includes one or more processors 501 and one or more memories 502.

[0180] Processor 501 may include one or more processing cores, such as a quad-core processor, a penta-core processor, etc. Processor 501 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 501 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 501 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 501 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0181] The memory 502 may include one or more computer-readable storage media, which may be non-transitory. The memory 502 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 502 are used to store at least one computer program, which is executed by the processor 501 to implement the ETC antenna control method provided in the method embodiments of this application.

[0182] Those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on vehicle 500 and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0183] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the ETC antenna control method provided in the above embodiments.

[0184] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the method for controlling an ETC antenna provided in the above embodiment.

[0185] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the method for controlling an ETC antenna provided in the above embodiment.

[0186] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0187] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0188] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0189] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A control method for an ETC antenna, characterized in that, The method includes: When the target vehicle is at the entrance of the ETC lane, based on the vehicle movement information of suspicious vehicles around the target vehicle, the first predicted driving trajectory of the suspicious vehicle is determined, and the suspicious vehicle is a vehicle that has the risk of using ETC without authorization. Based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle, the risk value of the suspicious vehicle is determined; If the risk value of the suspected vehicle is greater than or equal to the first risk value threshold, the ETC antenna of the target vehicle shall be turned off or put into sleep mode.

2. The method according to claim 1, characterized in that, The step of determining the first predicted driving trajectory of the suspicious vehicle based on the vehicle movement information of suspicious vehicles around the target vehicle includes: Based on the vehicle motion information of the suspicious vehicle, determine the lateral and longitudinal motion trajectories of the suspicious vehicle relative to the target vehicle within a preset future time period; The first predicted driving trajectory is generated based on the lateral movement trajectory and the longitudinal movement trajectory.

3. The method according to claim 2, characterized in that, The step of determining the lateral and longitudinal movement trajectories of the suspicious vehicle relative to the target vehicle within a preset future time period based on the vehicle movement information of the suspicious vehicle includes: Based on the current lateral position, current lateral velocity, and current heading angle in the vehicle motion information, the lateral movement trajectory of the suspicious vehicle relative to the target vehicle within the preset future time period is determined; Based on the current longitudinal position, current longitudinal speed, and current heading angle in the vehicle motion information, the longitudinal trajectory of the suspicious vehicle relative to the target vehicle within the preset future time period is determined.

4. The method according to claim 1, characterized in that, The step of determining the risk value of the suspicious vehicle based on the first predicted driving trajectory, the lane where the target vehicle is located, and the second predicted driving trajectory of the target vehicle includes: Based on the first predicted driving trajectory and the second predicted driving trajectory, the predicted minimum lateral distance between the suspicious vehicle and the target vehicle is determined; Based on the first predicted driving trajectory and the lane where the target vehicle is located, the predicted lane intrusion time and predicted lane intrusion angle of the suspicious vehicle are determined. The risk value of the suspicious vehicle is determined based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle.

5. The method according to claim 4, characterized in that, The step of determining the predicted lane incursion time and predicted lane incursion angle of the suspicious vehicle based on the first predicted driving trajectory and the lane where the target vehicle is located includes: Determine the predicted intersection point between the first predicted driving trajectory and the lane boundary line of the lane where the target vehicle is located; The time required for the suspicious vehicle to reach the predicted junction point is determined as the predicted lane intrusion time; The angle between the first predicted driving trajectory and the predicted intersection point and the lane is determined as the predicted lane intrusion angle.

6. The method according to claim 4, characterized in that, The process of determining the risk value of the suspicious vehicle based on the predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle includes: The predicted minimum lateral distance, the predicted lane intrusion time, and the predicted lane intrusion angle are normalized to obtain the distance risk factor corresponding to the predicted minimum lateral distance, the time risk factor corresponding to the predicted lane intrusion time, and the angle risk factor corresponding to the predicted lane intrusion angle. The distance risk factor, the time risk factor, and the angle risk factor are weighted and fused to obtain a comprehensive risk value. The comprehensive risk value is determined as the risk value of the suspicious vehicle.

7. The method according to claim 1, characterized in that, The step of controlling the ETC antenna of the target vehicle to turn off or go into sleep mode when the risk value of the suspected vehicle is greater than or equal to the risk value threshold includes: If the risk value of the suspected vehicle is greater than or equal to the risk value threshold, determine whether the first predicted driving trajectory intersects with the virtual connecting line, wherein the virtual connecting line is obtained by extending a preset distance from the target vehicle in the direction of the lane. If the first predicted driving trajectory intersects with the virtual connection line, the ETC antenna of the target vehicle is controlled to be turned off or put into sleep mode.

8. The method according to claim 1, characterized in that, The method further includes: If the risk value of the suspicious vehicle is greater than or equal to the second risk value threshold and less than the first risk threshold, a first prompt message is triggered, which is used to prompt attention to the suspicious vehicle. If the risk value of the suspicious vehicle is less than the second risk value threshold, no alert will be triggered.

9. The method according to claim 1, characterized in that, Before determining the first predicted driving trajectory of the suspicious vehicle based on the vehicle movement information of suspicious vehicles around the target vehicle, the method further includes: When candidate vehicles exist around the target vehicle, determine the relative position and relative speed between the candidate vehicles and the target vehicle. If the relative position between the candidate vehicle and the target vehicle meets a first preset condition, and the relative speed meets a second preset condition, the candidate vehicle is determined to be a suspicious vehicle.

10. A vehicle, characterized in that, The vehicles include: Memory, used to store executable program code; A processor is configured to call and run the executable program code from the memory, causing the vehicle to perform the control method for the ETC antenna as described in any one of claims 1 to 9.

Images