Vehicle following target CIPV identification method and system for navigation automatic driving

Abstract

The invention discloses a vehicle-following target CIPV identification method for navigation automatic driving, and relates to the technical field of automatic driving prediction algorithms. The method comprises the following steps: acquiring surrounding vehicle information, and extracting core information of surrounding vehicles; according to a preset screening rule, an important vehicle set with the CIPV candidate qualification is screened out; constructing a vehicle behavior prediction program based on a historical key vehicle set, inputting the screened important vehicle set into the vehicle behavior prediction program, predicting a vehicle behavior trend, and quantitatively predicting key parameters; outputting a complete state prediction sequence of each candidate vehicle; and a track deviation trend, a steering signal state and a vehicle speed adjustment characteristic are extracted, and whether a lane changing intention exists or not is judged through a vehicle behavior prediction program. According to a fixed formula for judging whether the vehicle deviates from the center threshold range of the lane or not, a comprehensive judgment method combining the current judgment formula and the transverse speed of the vehicle is provided, and the judgment accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving prediction algorithm, in particular to a navigation automatic driving car-following target CIPV (Car Intelligent Platform for Vehicles) identification method and system; specifically, the surrounding vehicle information is input, important vehicles are extracted, behavior prediction, vehicle speed, distance from the front vehicle and front vehicle speed prediction are performed on the important vehicles, and finally the vehicle trajectory of the important vehicles in the entire prediction time domain is predicted, whether the front vehicle in the adjacent lane has a lane-changing intention is intelligently judged, and the environmental adaptability and driving safety are ensured. BACKGROUND

[0002] With the increasingly wide application of automatic driving technology, more and more users will turn on the automatic driving function in road driving. In the process of car-following or constant-speed driving of the automatic driving vehicle, various vehicles will cut in or out in front. The current automatic driving prediction algorithm has design defects and deficiencies, mainly the technical limitations of prediction ability and environmental adaptability defects, such as being unable to timely cut in the vehicle or misjudging the lane-changing intention. Based on such problems, a navigation automatic driving car-following target CIPV identification method is proposed, which improves the accuracy through comprehensive judgment of the distance of vehicle deviation from the lane line and the lateral speed.

[0003] Prior art one, Chinese patent, patent number: 202511173528.3 proposes a vehicle automatic driving navigation method and system based on road sign recognition, relating to the field of automatic driving, the method comprising: acquiring image data and point cloud data of road signs, aligning and processing the image data and point cloud data to obtain fusion sign features, and extracting preliminary position information; acquiring actual position information of road signs, comparing the actual position information and the preliminary position information to obtain offset vector features; performing angle correction processing on the image data according to the visual features and the offset vector features to obtain sign posture information, and determining the change trend of the sign posture information; acquiring the vehicle driving trajectory, adjusting the vehicle driving trajectory according to the change trend and the offset vector features to obtain the predicted driving trajectory, and performing vehicle navigation based on the predicted driving trajectory. Although the real-time accuracy and dynamic update of vehicle automatic driving are realized, the screening rules are designed based on conventional road scenes, the navigation path information is not fully combined, leading to missing or misselecting of the CIPV candidate set, and it is difficult to adapt to the dynamic path changes of navigation planning.

[0004] The prior art two, Chinese patent, patent number: 202510562649.0 discloses an automatic driving vehicle motion risk assessment system and method based on multi-mode trajectory prediction, comprising four sensors, a time synchronization module, a test module, a local anomaly degree voting and fault detection counter, a trajectory correction and feature enhancement module and an LSTM network; the four sensors include an INS combined inertial navigation sensor, an inertial navigation IMU sensor, a line control chassis sensor and a visual recognition sensor, the INS combined inertial navigation sensor includes GPS and IMU; the test module includes χ2 test, motion mode consistency test and residual consistency test; the LSTM network includes LSTM encoder, classification decoder and LSTM decoder. The present application proposes a complete "detection-evaluation-decision" closed-loop solution. Although, by constructing a multi-mode prediction network including fault features, not only fault recognition is realized, but also the influence degree of fault on driving safety is accurately evaluated, providing a comprehensive safety decision basis for the automatic driving system; however, the prediction model is trained based on single vehicle type, fixed driving style historical data, which does not cover vehicle behavior characteristics under special vehicle type, abnormal driving behavior and extreme weather, resulting in low prediction accuracy of key parameters and weak ability to provide reliable support for CIPV decision.

[0005] The prior art three, Chinese patent, patent number: 202510619669.7 provides a mileage robust calculation method based on LSTM abnormality identification and compensation, first creates a driving trajectory enhanced data set, then pre-processes the data, determines the training set feature attributes and normalizes and time series formats the data, and then constructs an abnormal data identification and compensation model based on LSTM, trains the model to detect abnormal data in latitude and longitude position data, and uses position time series prediction value for compensation to generate continuous and accurate position data sequence, calculates the Euclidean distance based on the sequence after Gauss-Krug projection conversion, thereby calculating the accurate driving mileage. Although, it can effectively process abnormal data generated in the positioning process, and is suitable for navigation, automatic driving, free flow tolling and other fields; however, only the lane changing intention is judged by local features such as trajectory offset and steering signal, without combining context information such as navigation path and road signs, which is easy to misjudge temporary avoidance and lane adjustment as real lane changing, or miss illegal lane changing behavior without steering signal, affecting the stability of CIPV locking.

[0006] At present, the prior art one, the prior art two and the prior art three have the problems of scene adaptability limitation, weak behavior prediction generalization ability and one-sided lane changing intention judgment. In order to solve the above problems, the present application provides a navigation automatic driving following target CIPV identification method. SUMMARY

[0007] To achieve the above object, the present application adopts the following technical solutions: In one aspect of the present application, a navigation automatic driving follow-up target CIPV identification method is provided, which comprises the following steps: Based on the historical key vehicle set, a vehicle behavior prediction program is constructed, the important vehicle set after screening is input into the vehicle behavior prediction program, the vehicle behavior trend is predicted, and the key parameters are quantitatively predicted; and the complete state prediction sequence of each candidate vehicle is output. The targets in the adjacent lane and the current lane are screened, and the vehicle closest to the current vehicle in the current lane and the adjacent lane is the potential CIPV target; the intention of the potential CIPV target is judged and recorded; whether the vehicle in front of the current lane has a lane-changing intention is displayed by the complete state prediction sequence, and if so, the CIPV target is released; whether the vehicle closest to the adjacent lane has a cutting intention is displayed by the complete state prediction sequence, and if so, the target vehicle is recorded as a new CIPV target.

[0008] In an optional embodiment, the process of outputting the complete state prediction sequence of each candidate vehicle comprises the following steps: The current state quantity of each candidate vehicle is extracted from the important vehicle set, including position coordinates, velocity vectors and acceleration vectors; at the same time, the prediction time domain length and time step interval are set according to the vehicle parameters to form a prediction time axis; The state transition relationship in the pre-constructed vehicle behavior prediction program is used to sequentially deduce the state quantity of each time step by taking the obtained current state quantity as the initial input; by repeatedly applying the state transition relationship, the state quantity of all time steps in the prediction time domain is generated to form a state quantity evolution sequence; From the generated state quantity evolution sequence, the real-time vehicle speed change curve, the dynamic distance from the ego vehicle and the key parameters of the vehicle's own speed fluctuation are extracted; the key parameters and the state quantity sequence are combined to organize a complete state prediction sequence arranged in time sequence.

[0009] In an optional embodiment, the process of intention judgment comprises the following steps: Based on the prediction lateral displacement and lateral velocity data provided by the complete state prediction sequence, combined with the lane center line position, the lane width and the adjustment coefficient, the intention judgment threshold of each potential CIPV target is calculated; the prediction lateral displacement is compared with the intention judgment threshold, and the vehicle intention classification result is output as right lane change, lane keeping or left lane change; When the output intention classification result is the right lane change or left lane change of the vehicle in front of the current lane, it is determined that the vehicle in front of the current lane has a lane-changing intention; then the CIPV target release mechanism is triggered to clear the current follow-up target association. When the output intent classification result is a right-lane-change or left-lane-change of the nearest vehicle in the adjacent lane, and the lane-change direction points to the lane of the ego vehicle, it is determined that the vehicle has a cut-in intention; then a CIPV target updating mechanism is triggered to identify the nearest vehicle in the adjacent lane as a new following target.

[0010] In an alternative embodiment, the process of calculating the intention judgment threshold of each potential CIPV target includes the following steps: First, values are extracted from the lane centerline position, lane width, and a pre-set adjustment coefficient to form a reference lateral range; The obtained reference lateral range is compared with the predicted lateral displacement of each potential CIPV target in the complete state prediction sequence, and the corresponding upper and lower limits of lateral tolerance are generated according to the comparison result; the output is the lateral displacement threshold interval of each target; According to the obtained lateral displacement threshold interval, the intention judgment threshold of each target is determined in combination with the predicted lateral velocity information.

[0011] In an alternative embodiment, the process of determining the intention judgment threshold of each target includes the following steps: Values are extracted from the lane centerline position, lane width, and a pre-set adjustment coefficient, and a reference lateral range is formed through numerical combination processing; the lane centerline position serves as the longitudinal reference, the lane width serves as the lateral space measure, and the pre-set adjustment coefficient is used to adjust the sensitivity of the lateral range; the combination processing is to calculate a lateral distance range by integrating the values; The obtained reference lateral range is compared with the predicted lateral displacement of each potential CIPV target in the complete state prediction sequence, and the upper and lower limits of lateral tolerance are generated according to the comparison result; the comparison processing is based on the boundaries of the reference lateral range to evaluate the deviation of the predicted lateral displacement and determine a lateral displacement threshold interval, which defines the allowed range of vehicle lateral movement; According to the obtained lateral displacement threshold interval, the intention judgment threshold is determined in combination with the predicted lateral velocity information provided in the complete state prediction sequence; the combination processing uses the lateral displacement threshold interval as the spatial limit and the predicted lateral velocity as the dynamic indicator to generate a comprehensive intention judgment threshold for the final judgment of vehicle lane-change intention by evaluating whether the lateral velocity will cause the lateral displacement to exceed the threshold interval.

[0012] In an alternative embodiment, the process of evaluating whether the lateral velocity will cause the lateral displacement to exceed the threshold interval includes the following steps: The upper and lower limits of lateral tolerance in the lateral displacement threshold interval are used as the right and left spatial boundaries of vehicle lateral movement; The lateral motion dynamic feature is formed by extracting the predicted lateral velocity information from the complete state prediction sequence, analyzing the direction attribute and magnitude level of the lateral velocity, and the like. The lateral motion dynamic feature is fused with the right and left spatial boundaries to generate the intention judgment threshold for right-lane changing when the direction attribute points to the right and the magnitude level indicates that the lateral displacement tends to exceed the right spatial boundary, and to generate the intention judgment threshold for left-lane changing when the direction attribute points to the left and the magnitude level indicates that the lateral displacement tends to exceed the left spatial boundary.

[0013] In an alternative embodiment, the process of forming the lateral motion dynamic feature includes the following steps: The lateral velocity data in a future preset time domain is intercepted from the complete state prediction sequence to form a lateral velocity sequence segment, which is derived from the complete state prediction sequence output by the vehicle behavior prediction program and contains continuous change data of the lateral velocity in a future period of time; The dominant direction of the lateral velocity sequence segment is determined to identify the main symbol trend of the lateral velocity values in the segment and obtain the direction attribute, and the determination process is based on the dominant proportion of the data points in the sequence segment to distinguish the direction attribute as pointing to the right or left; The cumulative impact of the lateral velocity sequence segment is evaluated to measure the comprehensive displacement effect of all the lateral velocity data in the sequence segment on the lateral position of the vehicle and obtain the magnitude level, and the cumulative impact evaluation result is used to represent the intensity of the lateral movement and as a component of the lateral motion dynamic feature.

[0014] In an alternative embodiment, the process of determining the dominant proportion of the data points in the sequence segment includes the following steps: Each data point in the lateral velocity sequence segment is subjected to symbol marking processing to assign a direction symbol identifier to each lateral velocity value and generate a symbol marked sequence; The direction weight of the symbol marked sequence is calculated to count the number proportion of the symbol identifiers representing the right direction and the symbol identifiers representing the left direction in the symbol marked sequence and obtain a direction weight value, and the direction weight value reflects the number advantage degree of the different direction symbols in the symbol marked sequence; The direction weight value is compared with a preset direction determination threshold to make a decision, and when the direction weight value exceeds the direction determination threshold, the direction attribute pointing to the right is output, and when the direction weight value is lower than the negative value of the direction determination threshold, the direction attribute pointing to the left is output, and the comparison decision process is based on the relative relationship between the direction weight value and the fixed threshold to finally determine the classification result of the direction attribute.

[0015] In an alternative embodiment, the surrounding vehicle information is acquired, and core information of the surrounding vehicles is extracted; and a preset screening rule is used to eliminate invalid interference targets from the surrounding vehicles, and an important vehicle set with CIPV candidate qualifications is screened out. The core information of the surrounding vehicles includes position coordinates, a driving lane, a real-time vehicle speed, a longitudinal / lateral distance from the ego vehicle, and vehicle dimensions.

[0016] In another aspect of the present application, a navigation automatic driving following target CIPV identification system is provided for implementing the navigation automatic driving following target CIPV identification method. The navigation automatic driving following target CIPV identification system comprises: A surrounding vehicle screening module is configured to acquire surrounding vehicle information, extract core information of the surrounding vehicles, and eliminate invalid interference targets from the surrounding vehicles according to a preset screening rule, so as to screen out an important vehicle set with CIPV candidate qualifications. The core information of the surrounding vehicles includes position coordinates, a driving lane, a real-time vehicle speed, a longitudinal / lateral distance from the ego vehicle, and vehicle dimensions. A vehicle behavior prediction module is configured to construct a vehicle behavior prediction program based on a historical key vehicle set, input the screened important vehicle set into the vehicle behavior prediction program, predict a vehicle behavior trend, and quantitatively predict key parameters. The module outputs a complete state prediction sequence of each candidate vehicle. The predicted key parameters include a real-time vehicle speed change curve, a dynamic distance from a preceding vehicle, and a speed fluctuation of the preceding vehicle in a preset time domain. An adjustment prediction scheme module is configured to screen targets in adjacent lanes and the current lane, and the vehicle closest to the ego vehicle in the current lane and the adjacent lanes is a potential CIPV target. The module performs intention judgment on the potential CIPV target and records the result. If the complete state prediction sequence shows that the vehicle in the current lane has a lane-changing intention, the CIPV target is released. If the complete state prediction sequence shows that the vehicle in the adjacent lane has a lane-changing intention, the target vehicle is recorded as a new CIPV target.

[0017] The application collects core information such as the position coordinates of surrounding vehicles, driving lanes, and the like, eliminates invalid interference targets according to preset screening rules, screens a CIPV candidate important vehicle set; a vehicle behavior prediction program is constructed based on a historical key vehicle set, key parameters such as a future preset time domain vehicle speed change curve and a dynamic distance are quantitatively predicted by inputting the candidate set, and a complete state prediction sequence is output; the final follow-up target CIPV is determined in combination with the prediction sequence, the trajectory deviation trend, the steering signal state and the vehicle speed adjustment characteristics of the front vehicle in the adjacent lane are extracted, and the lane changing intention is judged through the prediction program; the core information extraction and screening rules are combined in this embodiment, which effectively eliminates interference targets and improves the accuracy of the CIPV candidate set; the prediction program trained by the historical data realizes quantitative prediction of key parameters, provides data support for follow-up target determination; the multi-dimensional feature analysis and lane changing intention judgment of the front vehicle in the adjacent lane improve the recognition accuracy of the follow-up target and the reliability of the lane changing prediction, provide accurate and comprehensive reference for the follow-up decision of the ego vehicle, and significantly improve the driving safety and the rationality of the driving decision. BRIEF DESCRIPTION OF DRAWINGS

[0018] The drawings accompanying the specification of this application form a part thereof, serve to provide further understanding of the application, and together with the description of the exemplary embodiments of the application, serve to explain the application, and do not constitute an improper limitation on the application. In the drawings: Figure 1 A step flowchart of an embodiment of the follow-up target CIPV recognition method for navigation automatic driving of the application; Figure 2 A schematic diagram of the principle of an embodiment of the follow-up target CIPV recognition method for navigation automatic driving of the application; Figure 3 A lane changing intention judgment schematic diagram of an embodiment of the follow-up target CIPV recognition method for navigation automatic driving of the application; Figure 4 A step flowchart of screening an important vehicle set with CIPV candidate qualifications of the application; Figure 5 A step flowchart of outputting a complete state prediction sequence of each candidate vehicle of the application; Figure 6 A step flowchart of intention judgment on potential CIPV targets of the application; Figure 7 A functional module schematic diagram of an embodiment of the follow-up target CIPV recognition system for navigation automatic driving of the application. DETAILED DESCRIPTION

[0019] The application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments. It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict.

[0020] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0021] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative in all respects and are not the only ones. All modifications within the scope of this invention or its equivalents are included in this invention.

[0022] like Figure 1 As shown, this embodiment provides an example of a CIPV (Continuous Vehicle Image) identification method for navigation-enabled autonomous driving. In this embodiment, the CIPV identification method specifically includes the following steps: Step S1: Obtain information about surrounding vehicles and extract core information about them; use preset filtering rules to remove invalid interfering targets from the surrounding vehicles and select a set of important vehicles that are qualified as CIPV candidates. The core information about surrounding vehicles includes their location coordinates, driving lane, real-time speed, longitudinal / lateral distance from the vehicle, and vehicle size. Step S2: Construct a vehicle behavior prediction program based on the historical key vehicle set. Input the selected key vehicle set into the vehicle behavior prediction program to predict vehicle behavior trends and quantitatively predict key parameters; output the complete state prediction sequence for each candidate vehicle. Among them, the key parameters for prediction include the real-time vehicle speed change curve within the future preset time domain, the dynamic distance to the vehicle in front, and the speed fluctuation of the vehicle in front itself. Step S3: Filter targets in adjacent lanes and the current lane. The vehicles closest to the current vehicle in the current lane and adjacent lanes are potential CIPV targets; determine the intent of potential CIPV targets and record them; when the complete state prediction sequence shows whether the vehicle in front of the current lane intends to change lanes, release the CIPV target if so; when the complete state prediction sequence shows whether the vehicle closest to the adjacent lane intends to cut in, record the target vehicle as a new CIPV target if so.

[0023] Preferably, the embodiment first marks the surrounding vehicles within a certain distance range as surrounding important vehicles according to the current vehicle speed of the ego vehicle, extracts the current state quantities of the identified surrounding important vehicles: x coordinate, y coordinate, longitudinal speed, lateral speed, longitudinal acceleration, lateral acceleration; within a period, the acceleration is constant, and according to the current state quantities, the state quantities at different times within a future whole period can be deduced; wherein the state quantities X at different times are:

[0024] For the prediction time domain Tp, the state equation at each time is established as follows:

[0025] 0 is the current time; Tp is the prediction time domain, which can be 5-10s (determined according to vehicle parameters); According to the lateral displacement state quantity y and the lateral speed Vy of the surrounding vehicles within a period, the vehicle intention is judged, and the judgment results are right lane change, lane keeping, and left lane change, which are represented as: Figure 3

[0026] wherein, : right lane change; : lane keeping; : left lane change; : center line; the coordinate on the left is positive, and the coordinate on the right is negative; : lane width; : lane width adjustment coefficient; : vehicle lateral speed, positive on the left and negative on the right; : speed adjustment coefficient; First, the targets of the adjacent lane and the current lane are screened, and the vehicle closest to the ego vehicle in the current lane and the adjacent lane is the potential CIPV target; the state equation within the prediction time domain Tp is established, and each state quantity is calculated; the intention is judged, and the record is made; whether the vehicle in front of the current lane has a lane change cut-in intention, if so, the CIPV target is released. Whether the vehicle closest to the ego vehicle in the adjacent lane has a cut-in intention, if so, the target vehicle is recorded as a new CIPV target (for details, refer to the attached Figure 2 ).

[0027] In summary, the principle is described in the attached Figure 2 ​​The embodiment collects core information such as position coordinates of surrounding vehicles, driving lanes and the like, removes invalid interference targets according to preset screening rules, and screens a CIPV candidate important vehicle set; a vehicle behavior prediction program is constructed based on a historical key vehicle set, key parameters such as a future preset time domain vehicle speed change curve and a dynamic distance are quantitatively predicted by inputting the candidate set, and a complete state prediction sequence is output; the final following target CIPV is determined in combination with the prediction sequence, the trajectory deviation trend, the steering signal state and the speed adjustment characteristics of the vehicle in front of the adjacent lane are extracted, and the lane changing intention is judged through the prediction program; the core information extraction and screening rules are combined to effectively exclude interference targets and improve the accuracy of the CIPV candidate set; the prediction program trained by the historical data realizes quantitative prediction of the key parameters, provides data support for the determination of the following target, analyzes the multi-dimensional characteristics of the vehicle in front of the adjacent lane and judges the lane changing intention, improves the accuracy of the following target recognition and the reliability of the lane changing prediction, provides accurate and comprehensive reference for the following decision of the ego vehicle, and significantly improves the driving safety and the rationality of the driving decision.

[0028] Further, as shown in Figure 2 The process of screening the important vehicle set with CIPV candidate qualifications in step S1 specifically includes the following steps: Step S11: Obtain the original data of all vehicles around the ego vehicle from the environment perception system, and extract the core information of each vehicle, including the position coordinates of the longitudinal and lateral coordinates with the ego vehicle as the reference system, the driving lane identifier, the real-time speed vector, the longitudinal and lateral distances from the ego vehicle, and the vehicle size; Step S12: Apply the preset three-level screening rules to remove invalid interference targets; The first level screening is based on a dynamic space region: an elliptical dynamic attention region is defined with the ego vehicle as the center, the long axis of the dynamic attention region extends in the longitudinal direction, and its length is dynamically adjusted by the real-time speed of the ego vehicle, the short axis extends in the lateral direction, and its width is determined based on the current lane width and the number of adjacent lanes; only the vehicles located in the dynamic attention region are retained, and the vehicles outside the region are excluded; The second level screening is based on lane correlation: from the vehicles after the first level screening, only the vehicles driving in the same lane as the ego vehicle or in the directly adjacent left lane and right lane are retained; the vehicles in other lanes are regarded as invalid targets and are removed; The third level screening is based on motion trend consistency: from the vehicles after the second level screening, the speed difference vector of each vehicle and the ego vehicle is calculated, and the deviation angle between the vehicle motion direction and the expected path direction of the ego vehicle is evaluated; only the vehicles with a deviation angle less than a preset threshold are retained to ensure that the motion trend of these vehicles is basically consistent with the path of the ego vehicle; Step S13: After three levels of screening, the output vehicles constitute an important vehicle set and have CIPV candidate qualifications.

[0029] Preferably, the embodiment only retains vehicles that are highly relevant to the ego vehicle's driving path in space, lane and motion trend, significantly reduces the number of targets that the perception system needs to process, and improves the focus and response speed of decision-making. The elliptical focus area scales in real time with the ego vehicle's speed, allowing it to cover farther front targets when driving at high speed, and tightening the range in low-speed or congested situations, ensuring that the screening range always matches the actual driving risk; through lane-related screening, targets far from the lane that are irrelevant to the ego vehicle's driving direction are excluded, avoiding misjudgment caused by crossing-lane noise and improving the reliability of target association; based on the speed difference vector and the deviation angle evaluation, it ensures that the remaining vehicles are basically consistent with the ego vehicle's expected path in the motion direction, reducing the misidentification caused by lateral or opposite motion targets. The position information, vehicle speed vector, size and other core features of each target are still retained during the filtering process, providing complete and high-quality data support for path planning, cooperative control and safety decision-making. The multi-level, space-based, lane-based and motion-based comprehensive screening mechanism can effectively suppress noise interference and false detection in complex traffic environments, improving the robustness and reliability of the overall perception system.

[0030] Further, the process of evaluating the deviation angle of the vehicle motion direction from the ego vehicle's expected path direction in step 12 includes the following steps: Step S121: Based on the ego vehicle's current driving lane identifier, a ego vehicle's expected path reference line is generated; the ego vehicle's expected path reference line is a virtual line parallel to the lane center line, and its curvature is determined by the lane geometry; process each candidate vehicle after the second level screening, extract the real-time vehicle speed vector of each candidate vehicle including size and direction; at the same time, obtain the real-time vehicle speed vector of the ego vehicle; Step S122: Perform vector composition on the candidate vehicle's speed vector and the ego vehicle's speed vector to obtain a relative motion vector that describes the candidate vehicle's motion trend relative to the ego vehicle; project the synthesized relative motion vector onto the ego vehicle's expected path reference line, and measure the directional angle difference between the relative motion vector and the ego vehicle's expected path reference line, which is the deviation angle; Step S123: Compare the calculated deviation angle with a preset angle threshold; only the candidate vehicles with a deviation angle less than the preset threshold are determined to have a motion trend consistent with the ego vehicle's path, and thus retained to enter the important vehicle set; vehicles with a deviation angle exceeding the preset threshold are excluded.

[0031] Preferably, the expected path reference line generated by the ego vehicle lane center line in the embodiment makes the motion trend evaluation always revolve around the actual driving route of the ego vehicle, ensuring that the screening result is highly consistent with the driving direction of the ego vehicle. The relative motion vector is obtained by synthesizing the speed vector of the candidate vehicle and the speed vector of the ego vehicle, eliminating the difference in speed of different vehicles and retaining only the relative motion trend, thus achieving a fair comparison of vehicles with different speeds. The relative motion vector is projected onto the expected path reference line and the directional angle difference between the relative motion vector and the expected path reference line is measured to obtain a clear deviation angle value, which provides a quantifiable criterion for threshold comparison; by comparing the deviation angle with the preset threshold, only vehicles with a directional deviation within the tolerance range are retained, and interference targets with a motion direction inconsistent with the path of the ego vehicle are eliminated, significantly reducing the false detection rate and improving the relevance of the important vehicle set; the evaluation process combines real-time speed vectors and lane geometry properties, and can maintain consistent screening criteria in different road curvatures and traffic scenarios, enhancing the robustness of the system in complex road environments. The remaining vehicles all have a motion trend basically consistent with the path direction of the ego vehicle, providing high-quality and reliable target information for path planning, cooperative control and safety decision-making.

[0032] Further, the process of measuring the directional angle difference between the relative motion vector and the expected path reference line of the ego vehicle in step S122 specifically includes the following steps: Step S1221: Based on the generated expected path reference line of the ego vehicle, a straight line is drawn at the current longitudinal position of the ego vehicle, which is tangent to the expected path reference line of the ego vehicle. The direction of the straight line is the instantaneous tangent direction of the expected path reference line of the ego vehicle at the current point; and a path tangent vector of unit length is constructed according to the instantaneous tangent direction; Step S1222: The relative motion vector obtained by the path tangent vector synthesis process is projected onto the constructed path tangent vector; the projection operation generates a tangent component along the path tangent; at the same time, the difference between the relative motion vector and the tangent component obtained by the projection forms a vertical component perpendicular to the expected path reference line of the ego vehicle; Step S1223: Based on the tangent component and the vertical component obtained, the ratio of the modulus of the vertical component to the modulus of the tangent component is calculated; the ratio is subjected to an inverse tangent operation, and the operation result is the directional angle difference between the relative motion vector and the expected path reference line of the ego vehicle.

[0033] Preferably, this embodiment projects the relative motion vector onto the path tangential vector to obtain the tangential and vertical components, and then obtains the directional deviation between the two through arctangent operation, achieving accurate numerical representation of the motion trend deviation. Using the instantaneous tangential direction of the vehicle's expected path baseline as a reference, the angle measurement is always consistent with the vehicle's current driving direction, ensuring that the evaluation result is highly correlated with the actual driving path. After separating the tangential and vertical components, only the proportional relationship between the two is considered, effectively filtering out noise components in the relative motion vector that are unrelated to the path direction, improving the robustness of angle estimation. The obtained directional angle difference provides a unified and comparable criterion for subsequent threshold comparison, facilitating a consistent screening standard under different traffic scenarios and vehicle speed conditions. Projection and modulus ratio operations are both basic linear algebra operations, with a simple and efficient calculation process, suitable for the high-speed execution requirements of real-time perception systems.

[0034] Furthermore, the process of forming a vertical component perpendicular to the vehicle's expected path baseline in step S1222 specifically includes the following steps: Step S12221: Define the tangential component along the path generated by the projection operation as a vector structure with a clear direction and magnitude; fully preserve the vector properties along the tangential vector direction. Step S12222: Perform a vector subtraction operation between the initially obtained original relative motion vector and the tangential component vector structure recorded in the first process to generate a new vector difference. Step S12223: Compare the new vector difference with the instantaneous normal direction of the vehicle's expected path baseline. When the two directions are consistent, the vector difference is finally confirmed as the vertical component perpendicular to the vehicle's expected path baseline.

[0035] Preferably, in this embodiment, the projection result along the tangential direction of the path is encapsulated into a vector structure with direction and magnitude, ensuring that subsequent calculations are based on unified and complete vector attributes. The difference obtained by subtracting the tangential component from the original relative motion vector is the component of the relative motion vector in the normal direction of the path baseline, achieving accurate separation of vertical information. The difference vector is compared with the instantaneous normal direction of the expected path baseline of the vehicle, and the difference is confirmed as the vertical component only when the directions are consistent, ensuring that the geometric meaning of the vertical component strictly corresponds to the path baseline. Through explicit tangential / vertical separation and normal comparison, motion components unrelated to the path direction are filtered out, improving the reliability and anti-interference ability of angle deviation calculation.

[0036] Furthermore, such as Figure 5 As shown, the process of outputting the complete state prediction sequence for each candidate vehicle in step S2 specifically includes the following steps: Step S21: Extract the current state quantity of each candidate vehicle from the important vehicle set, including position coordinates, velocity vector and acceleration vector. At the same time, set the prediction time domain length and time step interval according to the ego vehicle parameters, and form the prediction time axis; Step S22: Use the state transition relationship in the pre-constructed vehicle behavior prediction program to sequentially deduce the state quantity of each time step with the obtained current state quantity as the initial input; generate the state quantity of all time steps in the prediction time domain by repeatedly applying the state transition relationship, and constitute the state quantity evolution sequence; Step S23: Extract the real-time vehicle speed change curve, dynamic distance from the ego vehicle, and vehicle speed fluctuation from the generated state quantity evolution sequence; combine the key parameters with the state quantity sequence to form a complete state prediction sequence arranged in chronological order.

[0037] Preferably, the embodiment synchronously extracts key state quantities such as position, speed and acceleration from the important vehicle set, and generates a unified prediction time axis according to the ego vehicle parameters, ensuring that the subsequent prediction is performed on the same time reference; uses the state transition relationship of the vehicle behavior prediction program to recursively deduce the state of each time step from the current state, realizes multi-step lookahead of vehicle motion, and forms a complete state evolution sequence. In the complete state sequence, extract the vehicle speed change curve, dynamic distance from the ego vehicle, and speed fluctuation, and integrate the key parameters with the full state data in chronological order to generate a structured complete prediction sequence, providing high-quality and time-consistent input for path planning, cooperative control and safety evaluation.

[0038] Further, the process of generating the state quantity of all time steps in the prediction time domain in step S22 specifically includes the following steps: Step S221: Apply the pre-defined state transition relationship in the vehicle behavior prediction program to the extracted current state quantity of the candidate vehicle as input, and directly output the state quantity of the first time step on the prediction time axis; Step S222: Take the output state quantity of the first time step as new input, and apply the same state transition relationship again to output the state quantity of the second time step on the prediction time axis; Step S223: Take the output state quantity of the second time step as input, continue to apply the state transition relationship in a loop, and sequentially output the state quantity of all subsequent time steps on the prediction time axis until the set prediction time domain endpoint is covered, and finally form a complete state quantity evolution sequence.

[0039] Preferably, the embodiment realizes the continuous evolution of the candidate vehicle in the entire prediction time domain by recursively applying the unified state transition relationship; each step takes the just-generated state as the input of the next step, ensuring the temporal consistency and dynamic coherence of the state sequence; the process can completely cover all time steps within the set prediction range, providing multi-step forward-looking state information from the current time, and providing reliable timing data support for path planning and safety assessment.

[0040] Further, the process of applying the state transition relationship predefined in the vehicle behavior prediction program in step S221 specifically includes the following steps: Step S2211: From the memory unit of the vehicle behavior prediction program, retrieve the state quantity records of the target vehicle at consecutive historical time points, including position coordinates, velocity components, and acceleration components, and arrange them in chronological order to form a historical state sequence; Step S2212: Perform temporal analysis on the formed historical state sequence to identify the change rule between adjacent time state quantities, establish the mapping relationship between the current time state quantity and the next time corresponding state quantity, and form a state change pattern; Step S2213: Parameterize the extracted state change pattern and construct it into a mathematical relationship with a fixed structure, which can receive current state quantity input and output the next time state quantity prediction value, completing the standardized packaging of the state transition relationship.

[0041] Preferably, the embodiment realizes the learnable expression of vehicle behavior through the retrieval of historical state sequence, analysis of temporal rules, and parameterized modeling; abstracts the actual motion data at multiple time points into a unified mathematical mapping relationship, so that the current state can be directly mapped to the predicted state at the next time; the standardized state transition model not only retains the dynamic characteristics of vehicle motion, but also provides a fast and unified calculation interface, providing a reliable foundation for multi-step prediction.

[0042] Further, the process of constructing a mathematical relationship with a fixed structure in step S2213 specifically includes the following steps: Step S22131: Based on the correlation rules between state quantities identified in the state change pattern, set an input variable for each current state quantity, and set an output variable for each next time state quantity; Step S22132: According to the numerical correspondence rule revealed by the state change pattern, determine the combination mode between each output variable and a specific input variable to form a relationship structure framework composed of input variable combinations; Step S22133: Based on the quantified numerical correspondence in the state change pattern, assign specific coefficient values ​​to each combination item in the relational structure framework, multiply the coefficient values ​​with the corresponding variable combinations and then add them together to form a complete mathematical relation.

[0043] Preferably, this embodiment achieves deterministic numerical mapping between vehicle states through systematic modeling of variable mapping, combination structure, and coefficient assignment; it transforms multidimensional inputs into unified algebraic expressions, enabling the direct calculation of the predicted state at the next moment from the current state; the mathematical relation of this fixed structure combines computational efficiency and interpretability, providing a fast and reliable numerical deduction basis for subsequent recursive prediction.

[0044] Furthermore, such as Figure 6 As shown, the process of determining intent in step S3 specifically includes the following steps: Step S31: Based on the predicted lateral displacement and lateral velocity data provided by the complete state prediction sequence, and combined with the lane centerline position, lane width and adjustment coefficient, calculate the intent judgment threshold for each potential CIPV target; compare the predicted lateral displacement with the intent judgment threshold, and output the vehicle intent classification result as changing lanes to the right, keeping lanes, or changing lanes to the left. Step S32: When the output intent classification result is that the vehicle in front of this lane is changing lanes to the right or left, it is determined that the vehicle in front of this lane has the intent to change lanes and cut out; then the CIPV target release mechanism is triggered to clear the current following target association. Step S33: When the output intent classification result is that the nearest vehicle in the adjacent lane is changing lanes to the right or left, and the lane change direction is pointing towards the vehicle's lane, it is determined that the vehicle has the intent to cut in; then the CIPV target update mechanism is triggered, and the nearest vehicle in the adjacent lane is identified as the new following target.

[0045] Preferably, this embodiment calculates the intent threshold by combining the predicted results of lateral displacement and velocity with lane geometry information to achieve real-time intent classification of potential CIPV targets; based on the classification results, it triggers target release or target update mechanisms respectively, so that when a lane change intention appears ahead of the current lane, the existing following association can be released in time to avoid misfollowing; when an intention to cut into an adjacent lane appears, it can be quickly marked as a new following target to ensure that the following object is always the vehicle most likely to enter the current lane; overall, it forms an intent perception closed loop based on predicted trajectory, improving the accuracy of target selection and the system's response speed to lane change behavior.

[0046] Furthermore, the process of calculating the intent judgment threshold for each potential CIPV target in step S31 specifically includes the following steps: Step S311: first extract values from the lane centerline position, lane width and the pre-set adjustment coefficient, and combine them to form a reference lateral range; Step S312: compare the obtained reference lateral range with the predicted lateral displacement provided by each potential CIPV target in the complete state prediction sequence, and generate the corresponding upper and lower limits of lateral tolerance according to the comparison result; the output is the lateral displacement threshold interval of each target; When the predicted lateral velocity direction of the potential CIPV target is away from the lane centerline, and the predicted lateral displacement trend points to the upper limit boundary of the determined lateral displacement threshold interval, it is determined that the vehicle has the intention to change lanes to the right; the basis for this determination is that the vehicle is moving to the right boundary of the lane at a significant speed; when the predicted lateral velocity direction of the potential CIPV target is away from the lane centerline, and the predicted lateral displacement trend points to the lower limit boundary of the determined lateral displacement threshold interval, it is determined that the vehicle has the intention to change lanes to the left; the basis for this determination is that the vehicle is moving to the left boundary of the lane at a significant speed; Step S313: determine the intention judgment threshold of each target according to the obtained lateral displacement threshold interval and the predicted lateral velocity information.

[0047] Preferably, the embodiment realizes the accurate quantification of the lateral behavior limit of each potential CIPV target by constructing the reference lateral range from the lane geometric parameters and the adjustment coefficient, setting the upper and lower tolerances of the predicted lateral displacement using the range and forming the threshold interval, and then determining the final intention judgment threshold in combination with the predicted lateral velocity, thereby providing a unified, forward-looking and dynamically adaptive criterion for intention classification, and significantly improving the accuracy and robustness of target intention judgment.

[0048] Further, the process of determining the intention judgment threshold of each target in step S313 specifically includes the following steps: Step S3131: extract values from the lane centerline position, lane width and pre-set adjustment coefficient, and form a reference lateral range through value combination processing; wherein the lane centerline position is used as the longitudinal reference, the lane width is used as the lateral space measurement, and the pre-set adjustment coefficient is used to adjust the sensitivity of the lateral range; the combination processing is to calculate a lateral distance range by integrating the values; Step S3132: compare the obtained reference lateral range with the predicted lateral displacement provided by each potential CIPV target in the complete state prediction sequence, and generate the upper and lower limits of lateral tolerance according to the comparison result; the comparison processing is based on the boundaries of the reference lateral range to evaluate the deviation degree of the predicted lateral displacement, and determine a lateral displacement threshold interval, which defines the allowed range of vehicle lateral movement; Step S3133: According to the obtained lateral displacement threshold interval, combined with the predicted lateral velocity information provided in the complete state prediction sequence, the determination of the intention judgment threshold is carried out; The processing is to use the lateral displacement threshold interval as the spatial limit, and the predicted lateral velocity as the dynamic index. By evaluating whether the lateral velocity will cause the lateral displacement to exceed the threshold interval, a comprehensive intention judgment threshold is generated, which is used for the final judgment of the lane changing intention of the vehicle.

[0049] Preferably, the embodiment realizes the prospective quantification of the lateral behavior limit of each potential CIPV target by constructing the reference lateral range from the lane geometry and the adjustment coefficient, setting the upper and lower tolerances of the predicted lateral displacement to form the displacement threshold interval using the range, and then jointly evaluating the interval and the predicted lateral velocity to generate a comprehensive intention judgment threshold, providing a unified, dynamic and spatial speed coupling characteristic criterion for lane changing intention classification, which significantly improves the accuracy and robustness of intention judgment.

[0050] Further, the process of evaluating whether the lateral velocity will cause the lateral displacement to exceed the threshold interval in step S3133 specifically includes the following steps: Step S31331: Use the upper limit of the lateral tolerance and the lower limit of the lateral tolerance in the lateral displacement threshold interval as the right spatial boundary and the left spatial boundary of the vehicle lateral movement; Step S31332: Extract the predicted lateral velocity information from the complete state prediction sequence to analyze the direction attribute and size level of the lateral velocity, forming the lateral motion dynamic feature; The direction attribute is used to indicate the trend of the vehicle lateral movement, and the size level is used to represent the intensity of the lateral movement; Step S31333: Fuse and evaluate the lateral motion dynamic feature with the right spatial boundary and the left spatial boundary. When the direction attribute points to the right and the size level indicates that the lateral displacement has a tendency to exceed the right spatial boundary, the right lane changing intention judgment threshold is generated. When the direction attribute points to the left and the size level indicates that the lateral displacement has a tendency to exceed the left spatial boundary, the left lane changing intention judgment threshold is generated.

[0051] Preferably, the embodiment defines the left and right boundaries of the lateral displacement threshold interval as the spatial limit, extracts the direction and intensity of the predicted lateral velocity to form the motion feature, and fuses and evaluates the feature with the left and right boundaries. It can instantly determine whether the vehicle is likely to break through the left and right spatial limits within the prediction time domain, thereby generating the corresponding lane changing intention threshold, and realizing the prospective, spatial speed coupling accurate judgment of the lane changing intention of the potential CIPV target.

[0052] Further, the process of forming the lateral motion dynamic feature in step S31332 specifically includes the following steps: Step S313321: Extracting lateral velocity data in a preset time domain from the complete state prediction sequence to form a lateral velocity sequence segment; the lateral velocity sequence segment is derived from the complete state prediction sequence output by the vehicle behavior prediction program and contains continuous change data of lateral velocity in a future period of time; Step S313322: Determining the dominant direction of the lateral velocity sequence segment to identify the main symbol trend of the lateral velocity values in the segment and obtain the direction attribute; the determination process is based on the dominant proportion of data points in the sequence segment, and the direction attribute is divided into right-pointing or left-pointing; Step S313323: Evaluating the cumulative impact based on the lateral velocity sequence segment to measure the comprehensive displacement effect of all lateral velocity data in the sequence segment on the lateral position of the vehicle and obtain the size level; the cumulative impact evaluation result is used to represent the intensity of lateral movement and serves as a component of the lateral motion dynamic feature.

[0053] Preferably, the embodiment extracts a lateral velocity segment from a complete prediction sequence, determines the dominant direction according to the proportion of positive and negative values in the segment, and then calculates the comprehensive impact of all velocity values in the segment on the lateral displacement and divides the intensity level, forming a lateral motion dynamic feature with direction attribute and intensity level, which provides a prospective, quantitative and spatially velocity-coupled criterion for intent threshold evaluation.

[0054] Further, the determination process in step S313322 based on the dominant proportion of data points in the sequence segment specifically includes the following steps: Step S3133221: Performing symbol marking processing on each data point in the lateral velocity sequence segment to assign a direction symbol identifier to each lateral velocity value and generate a symbol marked sequence; the symbol marked sequence is derived from the mathematical symbol of each data point in the lateral velocity sequence segment; Step S3133222: Calculating the direction weight of the symbol marked sequence to count the number proportion of the symbol identifiers representing the right direction and the symbol identifiers representing the left direction in the symbol marked sequence and obtain a direction weight value; the direction weight value reflects the number advantage degree of different direction symbols in the symbol marked sequence; Step S3133223: Comparing the direction weight value with a preset direction determination threshold to make a decision, when the direction weight value exceeds the direction determination threshold, outputting the direction attribute pointing to the right; when the direction weight value is lower than the negative value of the direction determination threshold, outputting the direction attribute pointing to the left; the comparison decision process is based on the relative relationship between the direction weight value and the fixed threshold to finally determine the classification result of the direction attribute.

[0055] Preferably, the embodiment realizes fast and quantitative determination of the dominant direction of the sequence by symbolizing the transverse speed segment, counting the proportion of the direction symbol and comparing with the preset threshold, forms a clear direction attribute, and provides a reliable space for intention threshold evaluation Speed direction depends on.

[0056] As Figure 7 shown, the embodiment also provides an embodiment of a navigation automatic driving following target CIPV identification system, in which the navigation automatic driving following target CIPV identification system is applied to the navigation automatic driving following target CIPV identification method in the above embodiment. The navigation automatic driving following target CIPV identification system comprises: A surrounding vehicle screening module 1 is configured to acquire surrounding vehicle information, extract core information of the surrounding vehicles, and eliminate invalid interference targets from the surrounding vehicles according to a preset screening rule, and screen out an important vehicle set with CIPV candidate qualifications. The core information of the surrounding vehicles includes position coordinates, a driving lane, a real-time vehicle speed, a longitudinal / lateral distance from the ego vehicle, and vehicle dimensions. A vehicle behavior prediction module 2 is configured to construct a vehicle behavior prediction program based on a historical key vehicle set, input the screened important vehicle set into the vehicle behavior prediction program, predict a vehicle behavior trend, and quantitatively predict key parameters, and output a complete state prediction sequence of each candidate vehicle. The predicted key parameters include a real-time vehicle speed change curve, a dynamic distance from a preceding vehicle, and a speed fluctuation of the preceding vehicle in a future preset time domain. An adjustment prediction scheme module 3 is configured to screen targets in adjacent lanes and the current lane, and the vehicle closest to the ego vehicle in the current lane and the adjacent lanes is a potential CIPV target. The potential CIPV target is subjected to intention judgment and recorded. If the complete state prediction sequence shows that the vehicle in the current lane has a lane-changing intention, the CIPV target is released. If the complete state prediction sequence shows that the vehicle in the adjacent lane has a lane-changing intention, the target vehicle is recorded as a new CIPV target.

[0057] Preferably, the embodiment acquires surrounding vehicle information, extracts core information such as position coordinates, eliminates interference according to a preset rule, screens a CIPV candidate vehicle set, constructs a prediction program based on historical data, inputs the candidate set to quantitatively predict key parameters such as a vehicle speed change curve, and outputs a complete state prediction sequence. The CIPV is determined in combination with the prediction sequence, the trajectory offset of the preceding vehicle in the adjacent lane is extracted, and the prediction program is used to determine the lane-changing intention.

[0058] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0059] The present application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0060] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0062] Finally, it should be noted that the above-mentioned embodiments are merely intended for describing the technical solutions of the present application, but not for limiting it. Although the present application is described in detail with reference to the above embodiments, those skilled in the field should understand that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A method for identifying CIPV (Continuous Vehicle Image) in navigation-enabled autonomous driving, characterized in that, The method for identifying CIPV (Continuous Vehicle Image) targets in navigation-based autonomous driving includes the following steps: A vehicle behavior prediction program is built based on a set of historical key vehicles. The selected set of important vehicles is input into the vehicle behavior prediction program to predict vehicle behavior trends and quantitatively predict key parameters; the program outputs a complete state prediction sequence for each candidate vehicle. Targets in adjacent lanes and the current lane are screened. The vehicles closest to the current vehicle in the current lane and adjacent lanes are identified as potential CIPV targets. The intent of potential CIPV targets is determined and recorded. If the complete state prediction sequence indicates whether the vehicle in front of the current vehicle intends to change lanes, the CIPV target is released. If the complete state prediction sequence indicates whether the vehicle closest to the current vehicle in the adjacent lane intends to cut in, the target vehicle is recorded as a new CIPV target.

2. The CIPV (Continuous Vehicle Image Detection) method for navigation-based autonomous driving as described in claim 1, characterized in that, The process of outputting the complete state prediction sequence for each candidate vehicle includes the following steps: Extract the current state variables of each candidate vehicle from the set of important vehicles, including position coordinates, velocity vector, and acceleration vector; at the same time, set the prediction time domain length and time step interval according to the vehicle parameters to form the prediction time axis; Using the state transition relationship in the pre-built vehicle behavior prediction program, the current state quantity is used as the initial input to derive the state quantity at each time step in turn; by repeatedly applying the state transition relationship, the state quantities at all time steps in the prediction time domain are generated, forming a state quantity evolution sequence. From the generated state variable evolution sequence, extract the key parameters of real-time vehicle speed change curve, dynamic distance from the vehicle, and speed fluctuation of the vehicle itself; merge the key parameters with the state variable sequence and organize them into a complete state prediction sequence arranged in time order.

3. The CIPV (Continuous Vehicle Image Detection) method for navigation-based autonomous driving as described in claim 1, characterized in that, The process of determining intent includes the following steps: Based on the predicted lateral displacement and lateral velocity data provided by the complete state prediction sequence, combined with the lane centerline position, lane width and adjustment coefficient, the intent judgment threshold for each potential CIPV target is calculated; the predicted lateral displacement is compared with the intent judgment threshold, and the vehicle intent classification result is output as changing lanes to the right, keeping lanes, or changing lanes to the left. When the output intent classification result is that the vehicle in front of this lane is changing lanes to the right or left, it is determined that the vehicle in front of this lane has the intent to change lanes and cut out; then the CIPV target release mechanism is triggered to clear the current following target association. When the output intent classification result is that the nearest vehicle in the adjacent lane is changing lanes to the right or left, and the direction of the lane change is pointing towards the vehicle's own lane, it is determined that the vehicle has the intent to cut in; then the CIPV target update mechanism is triggered, and the nearest vehicle in the adjacent lane is identified as the new following target.

4. The CIPV (Continuous Vehicle Image Detection) method for navigation-based autonomous driving as described in claim 3, characterized in that, The process of calculating the intent judgment threshold for each potential CIPV target includes the following steps: First, extract values ​​from the lane centerline position, lane width, and pre-set adjustment coefficients, and combine them to form a baseline lateral range; The obtained baseline lateral range is compared with the predicted lateral displacement provided by each potential CIPV target in the full state prediction sequence. Based on the comparison results, the corresponding upper and lower limits of the lateral tolerance are generated. The output is the lateral displacement threshold range for each target. Based on the obtained lateral displacement threshold range and combined with the predicted lateral velocity information, the intent judgment threshold for each target is determined.

5. The CIPV (Continuous Vehicle Image) identification method for navigation-based autonomous driving as described in claim 4, characterized in that, The process of determining the intent judgment threshold for each target vehicle includes the following steps: The system extracts values ​​from the lane centerline position, lane width, and preset adjustment coefficients, and then combines these values ​​to form a baseline lateral range. The lane centerline position serves as the longitudinal reference benchmark, the lane width serves as the lateral space measurement, the preset adjustment coefficients are used to adjust the sensitivity of the lateral range, and the combined processing calculates a lateral distance range by integrating the values. The obtained baseline lateral range is compared with the predicted lateral displacement provided by each potential CIPV target in the full state prediction sequence. Based on the comparison results, the upper and lower limits of the lateral tolerance are generated. The comparison process is based on the boundary of the baseline lateral range to evaluate the degree of deviation of the predicted lateral displacement and determine a lateral displacement threshold interval. The lateral displacement threshold interval defines the allowable range of lateral movement of the vehicle. Based on the obtained lateral displacement threshold range, combined with the predicted lateral velocity information provided in the complete state prediction sequence, the intent judgment threshold is determined. The combined processing uses the lateral displacement threshold range as a spatial constraint and the predicted lateral velocity as a dynamic indicator. By evaluating whether the lateral velocity will cause the lateral displacement to exceed the threshold range, a comprehensive intent judgment threshold is generated to ultimately determine the vehicle's lane change intent.

6. The CIPV (Continuous Vehicle Image Portal) identification method for navigation-based autonomous driving as described in claim 5, characterized in that, The process of assessing whether lateral velocity would cause lateral displacement to exceed a threshold range includes the following steps: The upper and lower limits of the lateral tolerance within the lateral displacement threshold range are used as the right and left spatial boundaries of the vehicle's lateral movement. The predicted lateral velocity information is extracted from the complete state prediction sequence, and the lateral velocity direction attribute and magnitude level are parsed to form the lateral motion dynamic features; the direction attribute is used to indicate the tendency of the vehicle's lateral movement, and the magnitude level is used to indicate the intensity of the lateral movement. The lateral motion dynamics are fused and evaluated with the right and left spatial boundaries. When the direction attribute points to the right and the magnitude level indicates that the lateral displacement tends to exceed the right spatial boundary, a threshold for judging the intention to change lanes to the right is generated; when the direction attribute points to the left and the magnitude level indicates that the lateral displacement tends to exceed the left spatial boundary, a threshold for judging the intention to change lanes to the left is generated.

7. The CIPV (Continuous Vehicle Image Detection) method for navigation-based autonomous driving as described in claim 6, characterized in that, The process of forming lateral motion dynamic characteristics includes the following steps: A lateral velocity sequence segment is formed by extracting lateral velocity data for a future preset time domain from the complete state prediction sequence. The lateral velocity sequence segment originates from the complete state prediction sequence output by the vehicle behavior prediction program and contains continuous change data of lateral velocity over a future period. The dominant direction of the transverse velocity sequence segment is determined by identifying the main sign tendency of the transverse velocity values ​​in the segment and obtaining the direction attribute. The determination process is based on the dominant proportion of data points in the sequence segment, and the direction attribute is distinguished as pointing to the right or pointing to the left. The cumulative impact assessment is based on lateral velocity sequence segments. By measuring the combined displacement effect of all lateral velocity data within the sequence segment on the vehicle's lateral position, the magnitude of the impact is determined. The cumulative impact assessment results are used to characterize the severity of lateral movement and serve as a component of the dynamic characteristics of lateral motion.

8. The CIPV (Continuous Vehicle Image) identification method for navigation-based autonomous driving as described in claim 7, characterized in that, The determination process, based on the dominance ratio of data points within a sequence segment, includes the following steps: Each data point in the lateral velocity sequence segment is labeled with a symbol, and a direction symbol is assigned to each lateral velocity value to generate a symbol label sequence. The directional weights are calculated for the symbol marker sequence by statistically analyzing the ratio of symbols representing the right direction to those representing the left direction, resulting in a directional weight value. The directional weight value reflects the degree of numerical dominance of symbols representing different directions in the symbol marker sequence. The direction weight value is compared with a preset direction determination threshold. When the direction weight value exceeds the direction determination threshold, the direction attribute pointing to the right is output; when the direction weight value is lower than the negative value of the direction determination threshold, the direction attribute pointing to the left is output. The comparison and determination process is based on the relative relationship between the direction weight value and the fixed threshold to finally determine the classification result of the direction attribute.

9. The CIPV (Continuous Vehicle Image Detection) method for navigation-based autonomous driving as described in claim 1, characterized in that, Acquire information about surrounding vehicles and extract their core information; then, using preset filtering rules, eliminate invalid interfering targets from the surrounding vehicles and select an important set of vehicles that meet CIPV candidate qualifications. The key information about surrounding vehicles includes their location coordinates, driving lane, real-time speed, longitudinal / lateral distance from the vehicle, and vehicle size.

10. A CIPV (Continuous Vehicle Image Recognition) system for navigation-assisted autonomous driving, used to implement the CIPV recognition method for navigation-assisted autonomous driving as described in any one of claims 1-9, characterized in that, The navigation-assisted autonomous driving target CIPV recognition system includes: The surrounding vehicle filtering module is used to obtain surrounding vehicle information and extract core information of surrounding vehicles; and with preset filtering rules, invalid interference targets are removed from surrounding vehicles to filter out an important set of vehicles that meet CIPV candidate qualifications; among them, the core information of surrounding vehicles includes location coordinates, driving lane, real-time vehicle speed, longitudinal / lateral distance from the vehicle and vehicle size. The vehicle behavior prediction module is used to build a vehicle behavior prediction program based on a historical set of key vehicles. The selected set of important vehicles is input into the vehicle behavior prediction program to predict vehicle behavior trends and quantitatively predict key parameters; it outputs a complete state prediction sequence for each candidate vehicle. Among them, the predicted key parameters include the real-time vehicle speed change curve in the future preset time domain, the dynamic distance to the vehicle in front, and the speed fluctuation of the vehicle in front itself. The prediction scheme adjustment module is used to filter targets in adjacent lanes and the current lane. The vehicles closest to the current vehicle in the current lane and adjacent lanes are potential CIPV targets. The intent of potential CIPV targets is judged and recorded. When the complete state prediction sequence shows whether the vehicle in front of the current lane intends to change lanes, if so, the CIPV target is released. When the complete state prediction sequence shows whether the vehicle closest to the adjacent lane intends to cut in, if so, the target vehicle is recorded as a new CIPV target.

Images