Method and system for constructing interactive prediction risk field based on dynamic environment characteristics

Abstract

This invention provides a method and system for constructing an interactive predictive risk field based on dynamic environmental characteristics. The method includes: constructing an obstacle field based on the attributes and predicted behavior of surrounding vehicles; constructing a road field based on road conditions; configuring driving factors based on the driver's own characteristics; and integrating and interacting the obstacle field, road field, and driving factors to construct an interactive predictive risk field. This invention constructs corresponding risk field models for each potential factor in the proposed traffic scenario, including an obstacle field caused by traffic participants, a road field caused by lane lines and boundaries, and driver factors determined by the driver's own characteristics. By incorporating trajectory prediction and relative motion calculation of surrounding vehicles, the model can perceive impending risks earlier, demonstrating the potential of this method for subsequent application in decision-making and planning.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving scenario modeling, and in particular to a method and system for constructing an interactive prediction risk field based on dynamic environmental features. Background Technology

[0002] Intelligent vehicles integrate various technologies, including sensors, computer vision, and artificial intelligence, aiming to achieve autonomous driving and significantly improve driving safety and comfort. Lateral decision-making in intelligent vehicles refers to the appropriate lateral control actions taken by the vehicle during operation, based on acquired environmental information, such as lane keeping, lane changing, and obstacle avoidance. These decisions directly impact the vehicle's driving safety and comfort.

[0003] Lateral decision-making in intelligent vehicles relies on technologies such as environmental perception and scene modeling during vehicle operation, while motion planning is based on accurate assessment of road environmental risks. How to achieve accurate assessment of road environmental risks has become a pressing technical challenge. Summary of the Invention

[0004] One of the objectives of this invention is to provide a method and system for constructing an interactive predictive risk field based on dynamic environmental characteristics, so as to accurately assess the risks of the road environment.

[0005] This invention provides a method for constructing an interactive predictive risk field based on dynamic environmental features, comprising:

[0006] An obstacle course is constructed based on the attributes and predicted behavior of surrounding vehicles;

[0007] Construct a road site based on road conditions;

[0008] Configure driving factors based on the driver's own characteristics;

[0009] By integrating and interacting obstacle courses, road courses, and driving factors, an interactive predictive risk field is constructed.

[0010] Preferably, the steps for constructing the obstacle field are as follows:

[0011] The system identifies surrounding vehicles both statically and dynamically, and categorizes them into static and dynamic groups.

[0012] Based on the vehicle's center of mass and attitude, the risk field strength of the static group of vehicles is generated.

[0013] The trajectory of the vehicles in the dynamic group is predicted and combined with the relative speed between the vehicles to calculate the relative motion trajectory.

[0014] The relative motion trajectory is sampled at points to determine the electric field strength points;

[0015] Based on the trajectory length and vehicle width corresponding to the field strength points, the risk field strength of each field strength point is generated.

[0016] Preferably, the construction steps of the road site are as follows:

[0017] Analyze road conditions to determine the number of lanes, lane width, lane markings, and road curvature;

[0018] The field strength value of the road field is determined based on the number of lanes, lane width, lane line type, and road curvature.

[0019] Preferably, the driving factor configuration steps are as follows:

[0020] Assess driver capabilities and determine driver capability factors;

[0021] The driver's driving style is evaluated to determine the driver style factor;

[0022] Assess the driver's mental state and determine the driver's mental state factors;

[0023] Driving factors are determined based on driver ability factors, driver style factors, and driver state factors.

[0024] Preferably, the steps for the fusion and interaction of the obstacle course, road course, and driving factors are as follows:

[0025] The obstacle field, road field, and driving factors are normalized to obtain the interactive prediction risk field. The normalization formula is as follows:

[0026]

[0027] In the formula, m represents the number of obstacles in the current environment, and n is the number of lane boundaries or lines; σ dr Indicates driving factor; E road(j) (x, y) represents the road field strength at coordinates (x, y); E k(i) (x, y) represents the dynamic field strength of the obstacle field at coordinates (x, y); E s(i) (x, y) represents the static field strength of the obstacle field at coordinates (x, y).

[0028] The present invention also provides an interactive predictive risk field construction system based on dynamic environmental characteristics, including: an obstacle field analysis module, a road field analysis module, a driving factor analysis module, and an interactive fusion module;

[0029] The obstacle field analysis module constructs an obstacle field based on the attributes and predicted behavior of surrounding vehicles; the road field analysis module constructs a road field based on road conditions; the driving factor analysis module configures driving factors based on the driver's own characteristics; and the interactive fusion module integrates the obstacle field, road field, and driving factors to construct an interactive predictive risk field.

[0030] Preferably, the obstacle field analysis module includes: a grouping unit, a static analysis unit, and a dynamic analysis unit;

[0031] The grouping unit performs static and dynamic identification of surrounding vehicles and divides them into static and dynamic groups. The static analysis unit generates the risk field strength of vehicles in the static group based on the vehicle's center of mass and attitude. The dynamic analysis unit predicts the trajectory of vehicles in the dynamic group and calculates the relative motion trajectory by combining it with the relative speed between vehicles. It then samples the relative motion trajectory to determine the field strength points and generates the risk field strength of each field strength point based on the trajectory length and vehicle width corresponding to the field strength points.

[0032] Preferably, the road field analysis module includes a data parsing unit and a road field determination unit; wherein, the data parsing unit parses road conditions and determines the number of lanes, lane width, lane line type, and road curvature; the road field determination unit determines the field strength value of the road field based on the number of lanes, lane width, lane line type, and road curvature.

[0033] Preferably, the driving factor analysis module includes: a capability assessment unit, a style assessment unit, a mental state assessment unit, and a driving factor determination unit;

[0034] The assessment unit evaluates the driver's ability and determines the driver's ability factor; the style assessment unit evaluates the driver's driving style and determines the driver's style factor; the mental state assessment unit evaluates the driver's mental state and determines the driver's mental state factor; and the driving factor determination unit determines the driving factor based on the driver's ability factor, driver's style factor, and driver's mental state factor.

[0035] Preferably, the interactive fusion module performs the following operations:

[0036] The obstacle field, road field, and driving factors are normalized to obtain the interactive prediction risk field. The normalization formula is as follows:

[0037]

[0038] In the formula, m represents the number of obstacles in the current environment, and n is the number of lane boundaries or lines; σ dr Indicates driving factor; E road(j) (x, y) represents the road field strength at coordinates (x, y); E k(i)(x, y) represents the dynamic field strength of the obstacle field at coordinates (x, y); E s(i) (x, y) represents the static field strength of the obstacle field at coordinates (x, y).

[0039] The present invention has the following advantages:

[0040] For each potential factor in the proposed traffic scenario, a corresponding risk field model is constructed, including an obstacle field caused by traffic participants, a road field caused by lane lines and boundaries, and a driver factor determined by the driver's own characteristics. Furthermore, during the construction of the obstacle field, to further improve the predictive power and accuracy, predicted trajectories are introduced into the dynamic prediction field, and the relative trajectory method is used to calculate the relative motion states between vehicles. The established interactive predictive risk field can intuitively and comprehensively describe the risks generated by different traffic elements in the current environment, providing a new foundation for situational assessment in complex environments. By incorporating the prediction of surrounding vehicle trajectories and the calculation of relative motion, the model can perceive impending risks earlier, demonstrating the potential of this method for subsequent application in decision-making and planning.

[0041] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0042] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0043] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0044] Figure 1 This is a schematic diagram of an interactive prediction risk field construction method based on dynamic environmental features in an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of a static attribute field;

[0046] Figure 3 A schematic diagram for designing the shape of the dynamically predicted field;

[0047] Figure 4 This is a schematic diagram of the dynamic prediction field;

[0048] Figure 5 This is a schematic diagram of a lane-changing scenario;

[0049] Figure 6 This is a schematic diagram of the road site;

[0050] Figure 7 This is a schematic diagram of the risk field boundary in an embodiment of the present invention;

[0051] Figure 8 This is a schematic diagram of an interactive predictive risk field construction system based on dynamic environmental characteristics, as described in an embodiment of the present invention. Detailed Implementation

[0052] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0053] This invention provides a method for constructing an interactive predictive risk field based on dynamic environmental features, such as... Figure 1 As shown, it includes:

[0054] Step 1: Construct an obstacle course based on the attributes and predicted behavior of surrounding vehicles;

[0055] Surrounding vehicles are the most significant factor affecting the vehicle's movement and a major source of risk. The model built to address this is the obstacle field. Based on the motion state of the obstacle vehicles, it can be divided into a static attribute field and a dynamic prediction field. The static field is influenced by the objective attributes of the vehicles, equivalent to the influence of stationary vehicles on the vehicle itself; the field strength mainly depends on the obstacle's size, position, and attitude. For moving obstacle vehicles, the dynamic field not only describes the risk arising from their relative motion with the vehicle but also incorporates the prediction results, spreading the risk and projecting it onto the predicted trajectory. The magnitude and direction of the field strength are mainly determined by the relative speed and acceleration between the obstacle vehicle and the vehicle, as well as the predicted future trajectory.

[0056] Step 2: Construct the road site based on road conditions;

[0057] When vehicles travel on a road, they are inevitably constrained by the lane and lane markings, and must travel in an orderly manner according to established constraints and rules. The road field reflects the impact of these road conditions on driving safety. To make the constructed road field model more universal, the field strength value is mainly related to parameters such as the number of lanes, lane width, lane marking type, and road curvature.

[0058] Step 3: Configure driving factors based on the driver's own characteristics;

[0059] Drivers exhibit distinct behavioral characteristics and driving abilities, leading to vastly different risk assessments and personalized decision-making even in the same driving scenarios. For example, in terms of driving style, conservative drivers, compared to aggressive drivers, tend to overestimate environmental risks due to their timidity; in terms of driving ability, experienced drivers, compared to novices, handle various driving situations with greater ease, resulting in a relatively lower risk factor; and in terms of mental state, fatigued drivers, compared to those at their peak performance, are more likely to underestimate environmental risks. Therefore, driver factors describe the impact of these three types of driver factors on risk.

[0060] Step 4: Integrate and interact the obstacle course, road course, and driving factors to construct an interactive predictive risk field.

[0061] The steps for fusing and interacting the obstacle field, road field, and driving factors are as follows: Normalize the obstacle field, road field, and driving factors to obtain the interactive prediction risk field. The normalization formula is as follows:

[0062]

[0063] In the formula, m represents the number of obstacles in the current environment, and n is the number of lane boundaries or lines; σ dr Indicates driving factor; E road(j) (x, y) represents the road field strength at coordinates (x, y); E k(i) (x, y) represents the dynamic field strength of the obstacle field at coordinates (x, y); E s(i) (x, y) represents the static field strength of the obstacle field at coordinates (x, y);

[0064] For surrounding vehicles in the environment, when a vehicle is stationary on the road, the risk it poses is equivalent to a static obstacle, and the field strength is only related to its own size, pose, and other objective attributes. When a vehicle is moving on the road, in addition to the risk generated by its static objective attributes, its relative motion with other vehicles also generates risk, and its future driving intentions and trajectory will change the distribution and shape of the field strength. Therefore, the obstacle field is further refined into a static attribute field related to objective attributes and a dynamic prediction field related to motion state. In one embodiment, the obstacle field construction steps are as follows: static and dynamic identification of surrounding vehicles is performed, and they are divided into static and dynamic groups; based on the vehicle's center of mass and attitude, the risk field strength of the vehicles in the static group is generated; trajectory prediction is performed on the vehicles in the dynamic group and combined with the relative speed between them to calculate the relative motion trajectory; point sampling is performed on the relative motion trajectory to determine the field strength points; based on the trajectory length and vehicle width corresponding to the field strength points, the risk field strength of each field strength point is generated.

[0065] The design of the static attribute field function follows these rules: the overall risk is distributed around the vehicle's external dimensions, and the risk gradually decreases as the distance from the vehicle's center increases. The peak field strength occurs across the entire surface of the vehicle. Due to road and traffic regulations, the vehicle's travel direction is limited, meaning its lateral velocity is usually much smaller than its longitudinal velocity, making the longitudinal risk range of the obstacle vehicle much larger than its lateral risk range. The location of the risk field changes with the vehicle's center of mass, and its orientation changes with the vehicle's attitude. Based on the above analysis, the risk field strength of vehicle i in a stationary state can be expressed as:

[0066]

[0067] Where, x obs(i) y obs(i) Let A be the x and y coordinates of the obstacle vehicle's center of mass; (x, y) be the position of any point in space. obs σ is the field strength coefficient; β is a higher-order coefficient used to adjust the shape of the potential field peak; x and σ y The obstacle shape coefficients are used to adjust the relationship between the field intensity distribution and the vehicle's lateral and longitudinal dimensions. Furthermore, considering different obstacle poses, the field intensity distribution is rotated and translated through projection transformation based on the vehicle's current heading angle ψ. A schematic diagram of the static attribute field is shown below. Figure 2 As shown, the field strength increases as the distance from the obstacle center decreases. For points near the obstacle edge, the field strength increases rapidly, and the field strength is equal and at its maximum value across the entire space occupied by the obstacle. On the other hand, the overall shape of the risk field is determined by the dimensions of the obstacle vehicle. The obstacle shape coefficient can be determined using a pre-configured shape coefficient determination library; that is, based on the vehicle's brand, model, etc., the corresponding σ is determined from the shape coefficient determination library. x and σ y A obs Both β and β are pre-configured.

[0068] The dynamic prediction field is mainly used to describe the relative state between moving traffic participants and the voluntary vehicle, and the potential risks arising therefrom. It has the following characteristics: the greater the relative speed and the smaller the relative distance, the greater the possibility of a traffic accident, and the stronger the dynamic prediction field. The direction of the relative speed affects the orientation of the field. When the obstacle vehicle is faster than the voluntary vehicle, the obstacle vehicle is essentially "moving away" from the voluntary vehicle, and the field is oriented forward; conversely, when it is slower, the field is oriented backward. In addition to the relative motion state between the obstacle vehicle and the voluntary vehicle, the dynamic prediction field also incorporates the predicted trajectory, allowing the shape of the risk field to change and expand along its predicted trajectory direction. This invention, based on the trajectory characteristics of the obstacle vehicle, constructs the shape of the dynamic prediction field using the following four steps: Figure 3 As shown. First, as Figure 3As shown in (a), the same method as for static attribute fields is used to bend the risk distribution along the prediction trajectory to form a torus; secondly, the risk should gradually decrease with distance in the direction perpendicular to the trajectory, therefore the cross-section of the field is designed as a Gaussian cross-section, as... Figure 3 As shown in (b); next, considering that the risk should gradually decrease along the predicted trajectory direction, and that the rate of decrease should gradually slow down with the increase of arc length in the direction perpendicular to the trajectory, this paper designs height and width coefficients and correlates them with the length of the predicted trajectory, thereby causing the risk field to gradually expand and decrease in both the horizontal and vertical directions along the trajectory direction, as shown in (b). Figure 3 The effects of (c) and (d). Based on the above analysis, as Figure 3 As shown in (a), for a point (x, y) in space where the risk to be calculated is to be projected onto the predicted trajectory, the resulting projection point (x, y) can be obtained. n y n )satisfy:

[0069]

[0070] In the formula, Let p be all the points in the predicted trajectory sequence of the obstacle vehicle at time t, i.e., the predicted trajectory of the obstacle vehicle within 5 seconds.

[0071] At the same time, the trajectory length s corresponding to this point n This can be expressed as:

[0072]

[0073] Based on trajectory length s n The height coefficient A is defined. d This determines the steepness and decay characteristics of the risk field as the predicted trajectory progresses, and is expressed as:

[0074] A d =k s (s n -d pre )(s n -s total );

[0075] Where, k s d is the steepness coefficient. pre =t p v obs(i) Forward aiming distance; t p The forward aiming time is configured to 2 seconds. total The total length of the predicted trajectory at the current moment; v obs(i) The speed of the obstacle vehicle;

[0076] For the design of the field width, a width coefficient σ is defined. vThis determines the field's horizontal expansion characteristics, expressed as follows:

[0077] σ v =(k l +k ψ |ψ|)s n +c obs ;

[0078] Among them, c obs This represents the basic lateral extent of the obstacle vehicle's location field, which is related to the vehicle's width; k l k is the field diffusion slope when the obstacle vehicle is traveling in a straight line. ψ The steering gain coefficient causes the width of the field to increase or decrease as the vehicle's heading changes;

[0079] Therefore, the dynamic prediction field of the obstacle vehicle labeled i can be initially represented as:

[0080]

[0081] A schematic diagram of the dynamic prediction field is shown below. Figure 4 As shown, the risk field extends entirely along the predicted trajectory. Along the trajectory direction, the risk gradually decreases with increasing distance from the vehicle. Perpendicular to the trajectory direction, the field strength reaches its maximum at each trajectory point and decays towards both sides. Simultaneously, the maximum field strength gradually decreases as the trajectory extends. At this point, the shape of the risk field is no longer confined to a traditional ellipse or ring, but rather expands entirely based on the obstacle vehicle's motion state.

[0082] Although the dynamic prediction field described above can well describe the risk characteristics of the obstacle vehicle during its driving process and takes into account the obstacle vehicle's speed, trajectory and other characteristics, since the ultimate purpose of the risk field is to serve the vehicle's subsequent decision-making, the relative motion between the vehicle and other vehicles still needs to be taken into account in the dynamic prediction field.

[0083] A relative trajectory method was employed to incorporate relative motion into the risk field without altering the field function. For example... Figure 5 As shown, assume the vehicle speed is V. obs The obstacle vehicle's trajectory during the entire lane-changing process is a curve. At time T1, the predicted trajectory is output. However, from the driver's perspective, due to the relative motion between the two vehicles, the resulting risks do not entirely follow the predicted trajectory, but rather a so-called "relative trajectory." For example, if the obstacle vehicle ahead is much faster than the driver or the vehicle behind is much slower, even if the obstacle vehicle is very close to the driver, it will not have much impact on the driver's operation; conversely, only when there is a slow vehicle ahead and a fast vehicle behind will there be a significant potential risk.

[0084] Therefore, this paper bases the predicted trajectory of the output on the relative velocity ΔV = V ego -V obs Adjustments and optimizations were made to construct a relative trajectory sequence p. relv :

[0085]

[0086] like Figure 5 As shown, there is a predicted trajectory of the obstacle vehicle within 5 seconds at time T1. This trajectory is also equivalent to the relative trajectory of the vehicle when it is stationary. When the vehicle's speed is slower than the obstacle vehicle, the obstacle vehicle is moving forward relative to the vehicle, and the relative trajectory is shown as the dashed line. When the speeds of the vehicle and the obstacle vehicle are equal, the relative motion of the obstacle vehicle exists only in the lateral direction, and the relative trajectory is perpendicular to the horizontal direction. When the vehicle's speed is faster than the obstacle vehicle, the obstacle vehicle is moving backward relative to the vehicle, and the relative trajectory shifts to the rear of the obstacle vehicle.

[0087] Therefore, the predicted trajectory sequence p in the dynamic prediction field function calculation is replaced with the relative trajectory sequence p. relv This is the final dynamic prediction field model;

[0088] When vehicles travel on a road, they are inevitably affected by lane markings. Lane markings not only separate vehicles on the road, ensuring orderly traffic flow, but also, in particular, double yellow lines serve as a warning to vehicles in the opposite lane, prohibiting crossing them unless under special circumstances. Therefore, from a potential field design perspective, the road field should possess the following characteristics: road boundaries have the highest level of danger, followed by double yellow lines, with ordinary lane markings posing the lowest risk. Vehicles should stay as close to the center of each lane as possible, as the center of each lane carries the lowest risk. The road risk field is solely related to the road structure and should extend along the lanes, independent of traffic participants. After analysis, the steps for constructing the road field are as follows: analyze road conditions to determine the number of lanes, lane width, lane marking type, and road curvature; based on the number of lanes, lane width, lane marking type, and road curvature, determine the field strength value of the road field.

[0089]

[0090] constant η LANE The lane marking type coefficient is used to distinguish the degree of danger of different lane markings. The white dashed line has the lowest value, the double yellow line has a higher value, and the road boundary has the highest value. In this invention, the road boundary, double yellow line, and dashed line are set to 8, 3, and 1, respectively. d is the minimum Euclidean distance from the vehicle's current position to the center line of the lane. LANE σ is the lane width; LANE The road adjustment coefficient ensures that the field strength varies within a reasonable range and changes its rate of change with lateral distance; it is pre-configured.

[0091] Road site schematic diagram as follows Figure 6 As shown, the risk remains at its minimum at the center of each lane, visually represented by dark blue areas. The risk gradually increases with distance from the lane center, reaching its maximum at the boundaries. For ordinary road boundary dashed lines, vehicles are allowed to cross them; therefore, even on the boundary lines, the risk transitions with a slightly higher field strength, represented by light blue areas. This allows the road field to guide vehicles along the current lane center while also meeting the needs for vehicle movement planning when crossing lanes, enabling vehicles to complete driving tasks such as lane changes. However, for the lowermost road boundary (near Y=0), vehicles should avoid entering or approaching these areas during driving; therefore, the potential field color is darkest, the field strength is greatest, and the increase is also greatest. Similarly, in the double yellow line area near Y=12, vehicles should also stay as far away from this area as possible to avoid collision risk; therefore, the risk is also relatively high, with a significant increase.

[0092] Differences in behavioral characteristics and driving abilities among drivers lead to drastically different risk assessments. Risk assessments determined by driver characteristics can be categorized into three types: driving ability, driving style, mental state, and driver factor σ. dr This is a dimensionless coefficient between 0 and 1, determined based on the driver's performance in the three aspects mentioned above. Its value reflects the driver's sensitivity to and acceptance of environmental risks during driving. A larger value indicates that the driver is more concerned about environmental risks at that moment. The steps for configuring driving factors are as follows: assess the driver's ability to determine the driver's ability factor; assess the driver's driving style to determine the driver's style factor; assess the driver's mental state to determine the driver's mental state factor; and determine the driving factor based on the driver's ability factor, driver's style factor, and driver's mental state factor. The driver factors affecting the risk field distribution are defined as:

[0093] σ dr =η1σ skill +η2σ style +η3σ phy&psy ;

[0094] Where, σ skill σ is the driver's ability factor. style For the driver style factor, σ phy&psy η1, η2, and η3 are the driver's mental state factors, and η1, η2, and η3 are the weight coefficients corresponding to each factor.

[0095] When constructing a risk field for complex traffic scenarios, the size of the risk field needs to be comprehensively considered, taking into account the motion relationships between surrounding vehicles and the driver's field of vision. The design of the risk field's extent requires a balance between predictive performance and computational efficiency. A larger field size increases the amount of computation required, leading to poorer real-time algorithm performance; conversely, a smaller size results in less comprehensive description of features within the field, making subsequent decisions prone to getting trapped in local optima. Therefore, as... Figure 7 As shown, the field's range is defined from two aspects: the feasible area for the vehicle in the short term and the driver's observation area.

[0096] The feasible area is mainly determined by the speed limit of the road and the acceleration performance of the vehicle, and is defined as the area from the vehicle's current speed v. e Accelerate to maximum speed v max and decelerate to minimum speed v min The encompassed range. Within this range, every location in the risk field can be reached through appropriate acceleration and deceleration behavior of the vehicle, ensuring the feasibility of subsequent strategy calculations and control execution. To ensure that the established risk field can move along with the vehicle, this value is calculated using relative distance.

[0097] From current position x e Starting from the designated location, the area the vehicle will reach by accelerating is:

[0098]

[0099] Where a e Let the acceleration of the vehicle be 2 m / s². 2 ;v max It is determined by the larger of the road's maximum speed limit and the vehicle's desired speed.

[0100] Similarly, the rear region can also be obtained:

[0101]

[0102] Where d e For the deceleration of the vehicle, take -3m / s². 2 ;v min It is determined by the smaller of the road's minimum speed limit and the vehicle's desired speed.

[0103] Drivers typically allow sufficient forward and rearward visibility distances during daily driving to ensure they can analyze environmental conditions in advance and react accordingly. Therefore, in addition to the feasible area calculated based on vehicle motion, an observation area is defined to reflect the driver's observation range of the environment. The observation area is determined by the headway (TH). Considering that vehicles behind generally adjust their strategies autonomously based on their own vehicle's motion, posing a lower level of danger, while the sudden actions of the vehicle in front pose a significant threat, the observation area for the rear is shortened. The headway for both the front and rear is set to TH. f =3,TH r =-1. The observation area is calculated as follows:

[0104] x r =v e ·TH r ;

[0105] x f =v e ·TH f ;

[0106] Ultimately, the risk field is determined by taking the larger value between the feasible region and the observation region within its longitudinal range, in order to simultaneously satisfy the requirements of driver behavior and vehicle control.

[0107] X r =max(Δx) r x r );

[0108] X f =max(Δx) f x f );

[0109] In the lateral direction, the extent of the risk field is related to the number of lanes and lane width, and is defined as:

[0110] Y = nW;

[0111] Where n is the number of lanes and W is the lane width.

[0112] The scope of the risk field is as follows Figure 7 As shown, the state of the vehicle at time t can be represented as:

[0113]

[0114] in, This represents the horizontal and vertical coordinates of the vehicle. Indicates vehicle speed. Indicates acceleration. Indicates the heading angle. Indicates the steering wheel angle.

[0115] At time t-1, the risk field area is shown in the dashed box. For vehicles and static obstacles outside the window, it is difficult to pose an imminent collision risk, so they are temporarily ignored in the calculation. As the vehicle travels to time t, the risk field also shifts to the solid box area. At this time, obstacle vehicles and static obstacles enter the vehicle's field of vision, and the risk field needs to consider multiple traffic participants at the same time.

[0116] To achieve an accurate assessment of a driver's various abilities, in one embodiment, the assessment steps for the driving ability factor are as follows:

[0117] Obtain the driver's basic data; the basic data includes: the number of years the driver's license has been held, age, medical examination data, etc.

[0118] Obtain the driver's historical driving data; historical driving data includes: vehicle type data, total driving time or total mileage for each type of vehicle, etc.

[0119] Using a pre-configured first feature parameter extraction library, first feature parameters are extracted from driving data and historical driving data to obtain multiple first feature parameters. Among them, the first feature parameters include: parameters representing the number of years the driver's license has been obtained, parameters representing age, parameters representing the medical examination data evaluation score, parameters representing the total driving time or total mileage of various vehicle types, etc.

[0120] Using each first feature parameter as an index item, a pre-configured driving ability factor evaluation library is retrieved to determine the corresponding driving ability factor; the first feature parameters in the driving ability factor evaluation library are associated with the corresponding driving ability factors.

[0121] The steps for evaluating driving style factors are as follows:

[0122] Obtain the driver's historical driving data; for example, use driving data from the past year as the basis for analysis.

[0123] Based on a pre-defined event database, historical driving data is filtered to obtain multiple event data points. These event data points include: whether the driver ran a yellow light while passing through a traffic light; whether the driver ran a red light while passing through a traffic light; whether the driver yielded to another vehicle at a traffic intersection; whether the driver frequently changed lanes (i.e., the number of lane changes exceeded a pre-defined threshold); and whether the driver frequently overtook other vehicles (i.e., the number of overtaking maneuvers exceeded a pre-defined threshold).

[0124] Multiple event data are grouped and statistically analyzed, and the statistical data is quantified. Based on the quantified parameter values, an analysis vector is constructed. This includes the proportion of red light violations and yellow light violations to the total number of traffic light violations, the number of times drivers yielded at intersections, the ratio of frequent lane changes to the total number of driving times, and the ratio of frequent overtaking to the total number of driving times.

[0125] The analysis vector is used to retrieve the corresponding driving style factor from the pre-configured driving style factor evaluation library; the analysis vector and driving style factor in the driving style factor evaluation library are associated one-to-one.

[0126] The steps for analyzing driver mental state factors are as follows:

[0127] Obtain the current driver's driving data; driving data includes: driving duration;

[0128] Acquire facial images of the driver during driving; input the facial images into a pre-trained and converged neural network to obtain fatigue assessment data;

[0129] Based on driving time and fatigue assessment data, the corresponding driver mental state factors are retrieved from a pre-configured mental state factor analysis library.

[0130] To ensure the accuracy of the dynamic field construction, in one embodiment, after predicting the trajectory of the dynamic obstacle, a risk field can be constructed with the dynamic obstacle as the target to correct the trajectory. The specific correction steps are as follows:

[0131] Associate each trajectory point on the predicted trajectory with the risk field within its preset range.

[0132] When the field strength of each point in the risk field corresponding to the trajectory point is less than or equal to the preset threshold, the position of the trajectory point will not be adjusted.

[0133] When there is a point in the risk field corresponding to the trajectory point where the field strength is greater than a preset threshold, the nearest point is extracted as a reference point; the trajectory point is adjusted in a direction away from the reference point, and the adjustment distance is the difference between the distance between the reference point and the trajectory point and the threshold.

[0134] This invention also provides an interactive predictive risk field construction system based on dynamic environmental characteristics, such as... Figure 8 As shown, it includes: obstacle field analysis module 1, road field analysis module 2, driving factor analysis module 3, and interactive fusion module 4;

[0135] Among them, obstacle field analysis module 1 constructs an obstacle field based on the attributes and predicted behavior of surrounding vehicles; road field analysis module 2 constructs a road field based on road conditions; driving factor analysis module 3 configures driving factors based on the driver's own characteristics; and interactive fusion module 4 integrates and interacts the obstacle field, road field, and driving factors to construct an interactive predictive risk field.

[0136] The obstacle field analysis module includes: grouping unit, static analysis unit, and dynamic analysis unit;

[0137] The grouping unit performs static and dynamic identification of surrounding vehicles and divides them into static and dynamic groups. The static analysis unit generates the risk field strength of vehicles in the static group based on the vehicle's center of mass and attitude. The dynamic analysis unit predicts the trajectory of vehicles in the dynamic group and calculates the relative motion trajectory by combining it with the relative speed between vehicles. It then samples the relative motion trajectory to determine the field strength points and generates the risk field strength of each field strength point based on the trajectory length and vehicle width corresponding to the field strength points.

[0138] The road field analysis module includes a data parsing unit and a road field determination unit. The data parsing unit analyzes road conditions and determines the number of lanes, lane width, lane line type, and road curvature. The road field determination unit determines the field strength value of the road field based on the number of lanes, lane width, lane line type, and road curvature.

[0139] The driving factor analysis module includes: a capability assessment unit, a style assessment unit, a mental state assessment unit, and a driving factor determination unit.

[0140] The assessment unit evaluates the driver's ability and determines the driver's ability factor; the style assessment unit evaluates the driver's driving style and determines the driver's style factor; the mental state assessment unit evaluates the driver's mental state and determines the driver's mental state factor; and the driving factor determination unit determines the driving factor based on the driver's ability factor, driver's style factor, and driver's mental state factor.

[0141] The interactive integration module performs the following operations:

[0142] The obstacle field, road field, and driving factors are normalized to obtain the interactive prediction risk field. The normalization formula is as follows:

[0143]

[0144] In the formula, m represents the number of obstacles in the current environment, and n is the number of lane boundaries or lines; σ dr Indicates driving factor; E road(j) (x, y) represents the road field strength at coordinates (x, y); E k(i) (x, y) represents the dynamic field strength of the obstacle field at coordinates (x, y); E s(i) (x, y) represents the static field strength of the obstacle field at coordinates (x, y).

[0145] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for constructing an interactive predictive risk field based on dynamic environmental characteristics, characterized in that, include: An obstacle course is constructed based on the attributes and predicted behavior of surrounding vehicles; Construct a road site based on road conditions; Configure driving factors based on the driver's own characteristics; By integrating and interacting obstacle courses, road courses, and driving factors, an interactive predictive risk field is constructed. The steps for constructing the obstacle field are as follows: The system identifies surrounding vehicles both statically and dynamically, and categorizes them into static and dynamic groups. Based on the vehicle's center of mass and attitude, the risk field strength of the static group of vehicles is generated. The trajectory of the vehicles in the dynamic group is predicted and combined with the relative speed between the vehicles to calculate the relative motion trajectory. The relative motion trajectory is sampled at points to determine the electric field strength points; Based on the trajectory length and vehicle width corresponding to the field strength points, the risk field strength of each field strength point is generated; After predicting the trajectory of dynamic obstacles, the trajectory can be corrected by constructing a risk field with the dynamic obstacles as the target. The specific correction steps are as follows: Associate each trajectory point on the predicted trajectory with the risk field within its preset range. When the field strength of each point in the risk field corresponding to the trajectory point is less than or equal to the preset threshold, the position of the trajectory point will not be adjusted. When there is a point in the risk field corresponding to the trajectory point where the field strength is greater than a preset threshold, the nearest point is extracted as a reference point; the trajectory point is adjusted in a direction away from the reference point, and the adjustment distance is the difference between the distance between the reference point and the trajectory point and the threshold.

2. The method for constructing an interactive predictive risk field based on dynamic environmental characteristics as described in claim 1, characterized in that, The construction steps for the road site are as follows: Analyze road conditions to determine the number of lanes, lane width, lane markings, and road curvature; The field strength value of the road field is determined based on the number of lanes, lane width, lane line type, and road curvature.

3. The method for constructing an interactive predictive risk field based on dynamic environmental characteristics as described in claim 1, characterized in that, The steps for configuring driving factors are as follows: Assess driver capabilities and determine driver capability factors; The driver's driving style is evaluated to determine the driver style factor; Assess the driver's mental state and determine the driver's mental state factors; Driving factors are determined based on driver ability factors, driver style factors, and driver state factors.

4. The method for constructing an interactive predictive risk field based on dynamic environmental characteristics as described in claim 1, characterized in that, The steps for the integration and interaction of obstacle courses, road courses, and driving factors are as follows: The obstacle field, road field, and driving factors are normalized to obtain the interactive prediction risk field. The normalization formula is as follows: ； In the formula, This indicates the number of obstacles in the current environment. It refers to the number of lane boundaries or lines; Indicates driving factor; Representing coordinates The road field strength; Representing coordinates The dynamic field strength of the obstacle field; Representing coordinates The static field strength of the obstacle field.

5. A system for constructing an interactive predictive risk field based on dynamic environmental characteristics, characterized in that, include: The system includes an obstacle field analysis module, a road field analysis module, a driving factor analysis module, and an interactive fusion module. The obstacle field analysis module constructs an obstacle field based on the attributes and predicted behavior of surrounding vehicles; the road field analysis module constructs a road field based on road conditions; the driving factor analysis module configures driving factors based on the driver's own characteristics; and the interactive fusion module integrates the obstacle field, road field, and driving factors to construct an interactive predictive risk field. The obstacle field analysis module includes: grouping unit, static analysis unit, and dynamic analysis unit; The grouping unit performs static and dynamic identification of surrounding vehicles and divides them into static and dynamic groups. The static analysis unit generates the risk field strength of vehicles in the static group based on the vehicle's center of mass and attitude. The dynamic analysis unit predicts the trajectory of vehicles in the dynamic group and calculates the relative motion trajectory by combining it with the relative speed between vehicles. The relative motion trajectory is sampled to determine the field strength points. Based on the trajectory length and vehicle width corresponding to the field strength points, the risk field strength of each field strength point is generated. After predicting the trajectory of dynamic obstacles, the trajectory can be corrected by constructing a risk field with the dynamic obstacles as the target. The specific correction steps are as follows: Associate each trajectory point on the predicted trajectory with the risk field within its preset range. When the field strength of each point in the risk field corresponding to the trajectory point is less than or equal to the preset threshold, the position of the trajectory point will not be adjusted. When there is a point in the risk field corresponding to the trajectory point where the field strength is greater than a preset threshold, the nearest point is extracted as a reference point; the trajectory point is adjusted in a direction away from the reference point, and the adjustment distance is the difference between the distance between the reference point and the trajectory point and the threshold.

6. The interactive predictive risk field construction system based on dynamic environmental characteristics as described in claim 5, characterized in that, The road field analysis module includes a data parsing unit and a road field determination unit. The data parsing unit analyzes the road conditions and determines the number of lanes, lane width, lane line type, and road curvature. The road field determination unit determines the field strength value of the road field based on the number of lanes, lane width, lane line type, and road curvature.

7. The interactive predictive risk field construction system based on dynamic environmental characteristics as described in claim 5, characterized in that, The driving factor analysis module includes: a capability assessment unit, a style assessment unit, a mental state assessment unit, and a driving factor determination unit; The assessment unit evaluates the driver's ability and determines the driver's ability factor; the style assessment unit evaluates the driver's driving style and determines the driver's style factor; the mental state assessment unit evaluates the driver's mental state and determines the driver's mental state factor; and the driving factor determination unit determines the driving factor based on the driver's ability factor, driver's style factor, and driver's mental state factor.

8. The interactive predictive risk field construction system based on dynamic environmental characteristics as described in claim 5, characterized in that, The interactive fusion module performs the following operations: The obstacle field, road field, and driving factors are normalized to obtain the interactive prediction risk field. The normalization formula is as follows: ； In the formula, This indicates the number of obstacles in the current environment. It refers to the number of lane boundaries or lines; Indicates driving factor; Representing coordinates The road field strength; Representing coordinates The dynamic field strength of the obstacle field; Representing coordinates The static field strength of the obstacle field.

Images