Automatic danger-avoiding steering auxiliary system based on visual inspection

Abstract

The invention relates to the field of risk-avoiding steering assistance, in particular to an automatic risk-avoiding steering assistance system based on visual detection. The system comprises a risk area determination module used for acquiring and analyzing a real-time environment image of a driving vehicle and determining a dynamic risk area and a road slippery degree; the obstacle information determination module is used for acquiring vehicle information of the running vehicle and determining risk obstacle information in combination with the dynamic risk area; the motion information determination module is used for determining the motion information of the risk obstacle according to the real-time environment image and the risk obstacle information; and the steering information determination module is used for determining steering auxiliary information according to the motion information, the road slippery degree, the vehicle information and the real-time environment image. By acquiring and analyzing the real-time environment image and the vehicle information, the risk condition of the vehicle in the driving process is accurately determined, and the steering auxiliary information is accurately determined in combination with the actual road condition, so that the driving safety and reliability of the vehicle during avoidance are improved.

Description

Technical Field

[0001] The present application relates to the field of automatic risk avoidance steering assistance, and in particular to an automatic risk avoidance steering assistance system based on visual detection. Background Art

[0002] With the rapid development of the new energy vehicle industry, the automatic avoidance steering assistance technology of new energy vehicles is becoming more and more mature. When the driver drives a new energy vehicle on the road, the automatic avoidance steering assistance technology of new energy vehicles can help the driver deal with complex road conditions that the driver himself cannot deal with, thereby ensuring the driving safety of new energy vehicles and the safety and smoothness of roads, and reducing the risk of road traffic accidents.

[0003] However, due to the great uncertainty of road traffic conditions, the automatic evasive steering assistance technology of new energy vehicles may be unable to cope with the complex actual road traffic conditions, resulting in the failure of the evasive steering operation of new energy vehicles and increasing the risk of road traffic accidents. Summary of the invention

[0004] The present application provides an automatic risk avoidance steering assistance system based on visual detection to solve the above problems.

[0005] In a first aspect, the present application provides an automatic risk avoidance steering assistance system based on visual detection, the system comprising a risk area determination module, an obstacle information determination module, a motion information determination module and a steering information determination module, the risk area determination module and the obstacle information determination module are connected in sequence, the motion information determination module and the steering information determination module are connected in sequence, the motion information determination module is connected to the risk area determination module and the obstacle information determination module respectively, the risk area determination module and the steering information determination module are connected in sequence, and the obstacle information determination module and the steering information determination module are connected in sequence: A risk area determination module is used to obtain a real-time environmental image of a moving vehicle and determine a dynamic risk area and a degree of road slipperiness based on the real-time environmental image; An obstacle information determination module, used to obtain vehicle information of the traveling vehicle, and determine risk obstacle information according to the dynamic risk area and the vehicle information; A motion information determination module, used to determine the motion information of the risk obstacle according to the real-time environment image and the risk obstacle information; The steering information determination module is used to determine the steering assistance information according to the motion information, the road wetness, the vehicle information and the real-time environment image, and perform auxiliary steering operation according to the steering assistance information.

[0006] Through this solution, the risk area determination module obtains the real-time environmental image of the moving vehicle, identifies the dynamic risk area and evaluates the slippery degree of the road, ensuring accurate determination of the risk area. The obstacle information determination module combines the real-time environmental image and vehicle information to analyze and identify potential risk obstacles, and provides accurate risk obstacle positioning and feature information. The motion information determination module determines the motion information of the risk obstacle through the real-time environmental image and risk obstacle information, so as to judge the threat it poses to the normal driving of the vehicle. The steering information determination module generates accurate steering assistance information based on the motion information of the risk obstacle, the real-time environmental image, the slippery degree of the road and the vehicle information, to assist the driver in making effective steering and avoiding collision risks.

[0007] Optionally, when determining the dynamic risk area and the degree of road slipperiness based on the real-time environmental image, the risk area determination module is used to: determine the texture uniformity of the road during the vehicle's driving, the degree of road reflection, a number of static targets, a number of dynamic targets, the locations of a number of static targets and the motion trajectories of a number of dynamic targets based on the real-time environmental image; determine the degree of road slipperiness based on the texture uniformity and the degree of road reflection; predict the static motion trends of the number of static targets based on the locations and the real-time environmental image; predict the dynamic motion trends of the number of dynamic targets based on the real-time environmental image and the motion trajectories; determine whether the dynamic motion trend and the static motion trend affect the normal driving of the vehicle, and if so, determine the dynamic position coordinates of the dynamic target and the static position coordinates of the static target based on the dynamic motion trend and the static motion trend; determine the target formation range based on the dynamic position coordinates and the static position coordinates, and determine the target formation range as the dynamic risk area.

[0008] Through this solution, the texture uniformity and reflectivity of the road are analyzed through real-time environmental images. These factors directly affect the slipperiness of the road, so as to determine whether there is a risk of sliding. The positions and movement trajectories of several static and dynamic targets are identified and analyzed, and the movement trends of these targets in the future are predicted. In particular, for static targets, their static movement trends are predicted by analyzing their positions and real-time environmental images; for dynamic targets, their dynamic movement trends are predicted based on their movement trajectories. Determine whether the movement trends of these targets pose a threat to vehicle driving. If there is a potential risk, further calculate the precise position coordinates of the target. Combining the movement trajectories and position coordinates of dynamic and static targets, a target formation range is determined. This range is the dynamic risk area.

[0009] Optionally, the vehicle information includes the vehicle's driving speed and driving direction. When the obstacle information determination module determines the risk obstacle information based on the dynamic risk area and the vehicle information, it is used to: determine the static relative positions of the several static targets, the dynamic relative positions of the several dynamic targets, the movement speeds of the several dynamic targets, and the movement directions of the several dynamic targets in the dynamic risk area based on the real-time environmental image; determine the static risk targets and dynamic risk targets that affect the vehicle's driving based on the static relative positions and the dynamic relative positions and according to the movement speed, the movement direction, the driving speed, and the driving direction, and determine the static risk targets and the dynamic risk targets as the risk obstacles; determine the first position of the static risk target, the static target volume of the static risk target, the first relative position of the dynamic risk target, and the dynamic target volume of the dynamic risk target based on the real-time environmental image; and incorporate the first position, the static target volume, the first relative position, and the dynamic target volume into the risk obstacle information.

[0010] Through this solution, the relative positions and motion characteristics of static targets and dynamic targets in the dynamic risk area are identified through real-time environmental images, with a focus on the static relative positions of static targets and the dynamic relative positions of dynamic targets. The movement speed and direction of the dynamic targets are obtained, and combined with the vehicle's driving speed and direction, it is determined whether these dynamic targets will affect the vehicle's driving. If certain static or dynamic targets pose a threat, they are marked as risk obstacles. Through further analysis, the first position and static target volume of the static risk target, as well as the first relative position and dynamic target volume of the dynamic risk target, are accurately determined, and then this information is integrated into risk obstacle information. While improving driving safety, it accurately identifies and responds to potential risk obstacles, providing a key basis for vehicle automation decision-making.

[0011] Optionally, when the motion information determination module determines the motion information of the risk obstacle based on the real-time environmental image and the risk obstacle information, it is used to: determine the static real-time motion speed and the static real-time motion direction of the static risk target when it moves at the first position based on the real-time environmental image; determine the motion speed as the dynamic motion speed of the dynamic risk target at the first relative position, and determine the motion direction as the dynamic motion direction of the dynamic risk target at the first relative position; and incorporate the dynamic motion speed, the dynamic motion direction, the static real-time motion speed and the static real-time motion direction into the motion information.

[0012] Through this solution, the movement of static risk targets is analyzed through real-time environmental images to determine their static real-time movement speed and static real-time movement direction at the first position. For dynamic risk targets, their dynamic movement speed and dynamic movement direction are calculated based on their changes in the first relative position. Combining this information, the movement status of static risk targets and dynamic risk targets at a specific time point is accurately depicted. According to the relative position, movement speed and movement direction of the dynamic risk target, its movement prediction is adjusted in time. After integrating these data, comprehensive movement information is formed, which provides accurate movement trajectory prediction for subsequent risk assessment and dynamic decision-making, greatly improves the vehicle's ability to respond to potential obstacles, and effectively supports the safety of driving steering assistance.

[0013] Optionally, when the motion information determination module determines the static real-time motion speed and the static real-time motion direction of the static risk target when it moves at the first location based on the real-time environmental image, it is used to: determine the dynamic motion target in the vehicle motion perception area, the dynamic target speed of the dynamic motion target, the target motion direction of the dynamic motion target, the motion target volume of the dynamic motion target, the static target volume of the static risk target and the static target type of the static risk target based on the real-time environmental image; determine the aerodynamic flow intensity of the dynamic motion target based on the dynamic motion speed, the dynamic motion target volume and the dynamic motion direction; determine the static target weight of the static target based on the static target volume and the static target type; determine whether the static target moves based on the aerodynamic flow intensity, the location and the static target weight; if the static target moves, determine the static real-time motion speed and the static real-time motion direction of the several static targets when they start to move based on the real-time environmental image.

[0014] Through this solution, dynamic moving targets in the vehicle motion perception area are identified based on real-time environmental images, and the key features of dynamic moving targets are extracted. By analyzing the key features of dynamic moving targets, their aerodynamic flow intensity is calculated. The volume and type information of static risk targets are extracted and used to calculate the weight of their static targets. By combining the aerodynamic flow intensity, location and static target weight, it is determined whether the static target has moved. If the static target has moved, its static real-time motion speed and static real-time motion direction are further re-evaluated based on the real-time environmental image to ensure accurate tracking of the motion state of the static target. By comprehensively analyzing the various physical characteristics of dynamic targets and static targets, the potential risks in the environment can be accurately judged, effectively improving the safety and responsiveness of the intelligent driving system.

[0015] Optionally, the vehicle information includes the vehicle motion trajectory of the vehicle, the real-time weight of the vehicle and the vehicle acceleration. When the steering information determination module determines the steering assistance information according to the motion information, the road slipperiness, the vehicle information and the real-time environmental image, it is used to: determine whether the normal driving state of the vehicle is affected according to the motion trajectory and the vehicle motion trajectory; if affected, determine the dynamic relative distance between the vehicle and the dynamic risk target, the dynamic acceleration of the dynamic risk target and the dynamic relative acceleration between the dynamic risk target and the vehicle according to the real-time environmental image; obtain the impact moment of the dynamic risk target affecting the normal driving state and the turning moment when the driver starts to turn, and determine the driver's reaction time according to the impact moment and the turning moment; determine the steering avoidance angle of the vehicle according to the driving speed, the dynamic relative acceleration, the vehicle acceleration, the real-time weight, the reaction time, the dynamic relative distance, the dynamic acceleration, the road slipperiness and the dynamic target volume, and calculate it according to the following formula: ; in, represents the steering avoidance angle, represents the reaction time, represents the dynamic acceleration, Indicates the slipperiness of the road. represents the dynamic target volume, represents the dynamic relative distance, represents the driving speed, represents the vehicle acceleration, represents the reaction time, represents the real-time weight, Representing the dynamic relative acceleration; and incorporating the steering avoidance angle into the steering assist information.

[0016] Through this solution, by comparing the vehicle's motion trajectory with the motion trajectory, it is determined whether an abnormality occurs; if affected, the relative distance between the vehicle and the dynamic risk target, the dynamic acceleration, and the dynamic relative acceleration relative acceleration are evaluated, and the driver's reaction time is calculated. Through these data, combined with multiple factors such as the vehicle's driving speed and acceleration, the best steering avoidance angle is calculated using a formula to ensure that the risk target can be avoided in time in critical moments. The various parameters involved in the formula help simulate the complex motion state in real situations, making the steering angle decision more accurate. These data will be converted into steering assistance information and provided to the driver to assist him in making a faster and more effective response, thereby effectively enhancing driving safety, especially in complex or dangerous driving environments.

[0017] Optionally, the real-time environment image includes a road edge image, and the vehicle information includes a vehicle width. When the steering information determination module determines the steering assist information according to the motion information, the road wetness, the vehicle information and the real-time environment image, it is used to: determine the road edge curvature of the vehicle when turning, the real-time position of the vehicle on the road, and the real-time interval distance between the vehicle and the road edge according to the road edge image; determine the real-time turning radius of the vehicle when turning according to the road edge curvature, the real-time position, the driving speed and the real-time interval distance; determine the turning speed of the vehicle when turning according to the driving speed, the dynamic motion speed and the dynamic target volume; determine the turning speed reduction of the vehicle when turning according to the turning speed and the driving speed; determine the turning time of the vehicle according to the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road wetness and the driving speed, and calculate it according to the following formula: ; in, represents the duration of the turn, represents the width of the vehicle, represents the dynamic target volume, Indicates the preset volume avoidance coefficient, represents the driving speed, represents the reaction time, represents the dynamic motion speed, represents the real-time turning radius, Indicates the slipperiness of the road. Indicates the reduction amount of the steering speed; and incorporates the steering duration and the steering speed into the steering assistance information.

[0018] Through this solution, the road curvature, the real-time position of the vehicle, and the real-time distance from the road edge are obtained through the road edge image, and then the real-time turning radius of the vehicle when turning is calculated. At the same time, combined with factors such as driving speed and dynamic target volume, the turning speed is determined, and the reduction in turning speed is evaluated. These data are combined with factors such as vehicle acceleration to further calculate the turning time to ensure that obstacle avoidance behavior in complex environments fully considers the changes in turning time and speed. The various parameters in the formula help simulate the dynamic changes during actual driving, thereby providing accurate steering assistance information, helping drivers make safe and efficient steering decisions under complex road and environmental conditions, and improving driving safety and operational responsiveness.

[0019] Optionally, when determining the turning time of the vehicle according to the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, the turning information determination module is used to: determine whether the road edge boundary is clear according to the real-time environmental image; if the road edge boundary is not clear, determine the object volume and object position in the shoulder area according to the real-time environmental image; determine the actual road width according to the object volume and the object position; determine the turning time according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, and calculate according to the following formula: ; in, represents the duration of the turn, represents the actual width of the road, represents the width of the vehicle, represents the dynamic target volume, represents the driving speed, represents the reaction time, represents the dynamic motion speed, represents the real-time turning radius, Indicates the slipperiness of the road. represents the reduction in steering speed, represents the volume of the object, represents the vehicle acceleration.

[0020] Through this solution, when the edge of the road is unclear, objects in the shoulder area are identified through real-time environmental image analysis, and the actual width of the road is estimated. By comprehensively considering variables such as vehicle acceleration, object volume, road wetness, and real-time turning radius, the turning time required for steering can be accurately determined. The various parameters in the formula, such as dynamic target volume, vehicle width, turning radius, etc., reflect the actual status of the road and vehicle, ensuring that the calculation of turning time can adapt to different environmental conditions. The accurate calculation of turning time takes into account the road conditions, object obstacles, and vehicle motion characteristics, effectively optimizes the driver's operational response, and improves safety and driving comfort. Through dynamic adjustment and accurate prediction, the steering assist function under complex or unclear road conditions is enhanced.

[0021] Optionally, after determining the turning duration according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, the turning information determination module is further used to: determine a first turning duration according to the actual road width, the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the road slipperiness and the driving speed, and calculate according to the following formula: ; in, represents the first turn duration, represents the actual width of the road, represents the dynamic target volume, represents the preset volume safety factor, represents the driving speed, represents the dynamic motion speed, represents the reaction time, represents the vehicle acceleration, Indicates the slipperiness of the road. Indicates the amount of reduction in steering speed.

[0022] Through this solution, the first turning time is calculated according to a formula that incorporates key parameters such as road width, vehicle speed, dynamic target volume, and reaction time, reflecting the impact of different driving conditions on the turning time. Specifically, the variables in the formula reflect environmental conditions (such as wetness, road width) and the characteristics of the vehicle itself (such as acceleration and width). Through reasonable parameter adjustment, the time required for turning is accurately estimated to ensure the smoothness and safety of the vehicle's steering action. In this way, the vehicle can improve the accuracy of steering control in complex driving environments and enhance the dynamic stability and safety of the vehicle.

[0023] Optionally, when the steering information determination module determines the steering assistance information according to the motion information, the road slipperiness, the vehicle information and the real-time environmental image, it is used to: determine a first slipperiness of the road after the vehicle turns according to the real-time environmental image; determine a first driving speed of the vehicle after the vehicle turns according to the vehicle information; determine a real-time offset of the center of gravity of the vehicle according to the real-time weight, the first driving speed and the first slipperiness; obtain a center of gravity offset safety range, match the real-time offset with the center of gravity offset safety range, and determine whether the real-time offset is within the center of gravity offset safety range; if the real-time offset is not within the center of gravity offset safety range, determine an offset difference according to the real-time offset and the center of gravity offset safety range; determine a maximum turning speed of the vehicle according to the offset difference, the first driving speed and the real-time weight; determine a secondary steering angle of the vehicle after the turn according to the maximum steering speed, the first slipperiness and the real-time weight, and incorporate the secondary steering angle into the steering assistance information.

[0024] Through this solution, the first slippery degree of the road after the vehicle turns is determined based on the real-time environmental image, laying the foundation for subsequent steering adjustments; the first driving speed is calculated using vehicle information to reflect vehicle state changes; the real-time offset degree of the vehicle's center of gravity is determined in combination with the real-time weight, slippery degree and first driving speed to evaluate steering safety. By matching the real-time offset degree with the safe range of the center of gravity offset, the vehicle stability is effectively monitored. If the offset exceeds the safe range, the offset degree difference is further calculated to provide data support for the maximum steering speed adjustment. According to the maximum steering speed, the first slippery degree and the real-time weight, the secondary steering angle is determined to make the vehicle steering more accurate and safe. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0026] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application; Figure 2 A schematic diagram of the structure of an automatic risk avoidance steering assistance system based on visual detection provided in one embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0029] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0030] However, due to the great uncertainty of road traffic conditions, the automatic evasive steering assistance technology of new energy vehicles may be unable to cope with the complex actual road traffic conditions, resulting in the failure of the evasive steering operation of new energy vehicles and increasing the risk of road traffic accidents.

[0031] Based on this, the present application provides an automatic risk avoidance steering assistance system based on visual detection. The risk area determination module obtains the real-time environmental image of the moving vehicle, identifies the dynamic risk area and evaluates the degree of road slipperiness, ensuring accurate determination of the risk area. The obstacle information determination module combines the real-time environmental image and vehicle information to analyze and identify potential risk obstacles, and provide accurate risk obstacle positioning and feature information. The motion information determination module determines the motion information of the risk obstacle through the real-time environmental image and risk obstacle information, so as to judge the threat it poses to the normal driving of the vehicle. The steering information determination module generates accurate steering assistance information based on the motion information of the risk obstacle, the real-time environmental image, the degree of road slipperiness and the vehicle information, to assist the driver in effective steering and avoid collision risks.

[0032] Figure 1 A schematic diagram of an application scenario provided for the present application. When performing risk avoidance steering assistance for a vehicle, the method provided by the present application is applied to obtain and analyze real-time environmental images and vehicle information, determine the steering assistance information of the vehicle when encountering driving risks, and perform assisted steering operations based on the steering assistance information.

[0033] Specifically, the method provided by the present application is applied to any server, and the server interacts with the vehicle monitoring device and the driving control system. The driving control system can be a system for controlling and managing the driving operation and driving state of the vehicle. The driving control system can include a dynamically stored database, which can store dynamic data of the vehicle during driving. The vehicle monitoring device can be a device that monitors the driving environment of the vehicle when the vehicle is driving. The server determines the steering assistance information of the vehicle by acquiring and analyzing the vehicle information of the driving control system and the real-time environmental image of the vehicle monitoring device, and sends the steering assistance information to the driving control system, and the driving control system performs steering assistance control on the vehicle according to the steering assistance information. By acquiring and analyzing the vehicle information and the real-time environmental image, the steering assistance information is accurately determined to ensure the safety and stability of the steering process. After the steering assistance information is sent to the driving control system, the driving control system adjusts the steering control of the vehicle in real time according to the steering assistance information to avoid the vehicle from being out of control or unstable due to factors such as road conditions and vehicle speed, thereby ensuring driving safety.

[0034] For specific implementation methods, please refer to the following embodiments.

[0035] Figure 2 This is a schematic diagram of the structure of an automatic risk avoidance steering assistance system based on visual detection provided by an embodiment of the present application. The system of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the system includes a risk area determination module 201, an obstacle information determination module 202, a motion information determination module 203 and a steering information determination module 204: The risk area determination module 201 is used to obtain a real-time environment image of a moving vehicle, and determine a dynamic risk area and a degree of road slipperiness based on the real-time environment image.

[0036] The real-time environment image may be an image obtained by monitoring the objects contained in the driving area of ​​the vehicle. The dynamic risk area may be an area formed by connecting the positions corresponding to objects that have potential impacts and risks on the normal driving state of the vehicle in a certain order. The road slipperiness may be a specific value indicating that the friction between the wheels and the road is reduced when the vehicle is driving, thereby increasing the risk of vehicle slippage.

[0037] Specifically, a real-time environment image is obtained from a vehicle monitoring device. A feature recognition algorithm is used to identify the real-time environment image to obtain dynamic targets and static targets in the real-time environment image. An image segmentation algorithm is used to analyze the real-time environment image to obtain several environment area images. An object detection and tracking algorithm is used to analyze several environment area images to obtain potential motion trajectories of dynamic targets and potential static motion trajectories of static targets. An image analysis algorithm is used to analyze the real-time environment image to obtain the real-time motion trajectory of the vehicle. The real-time motion trajectory is matched with the potential static motion trajectory and the potential motion trajectory respectively, and the area where the static target corresponds to the potential static motion trajectory that intersects with the real-time motion trajectory is determined as a dynamic risk area, and the area where the dynamic target corresponds to the potential motion trajectory that intersects with the real-time motion trajectory is determined as a dynamic risk area. An image analysis algorithm is used to analyze the real-time environment image to obtain the degree of road slipperiness.

[0038] The obstacle information determination module 202 is used to obtain vehicle information of a moving vehicle and determine risk obstacle information according to the dynamic risk area and vehicle information.

[0039] The vehicle information may be information generated by the vehicle while it is driving and the parameter information of the vehicle itself. The risk obstacle information may be information corresponding to objects that actually affect and pose risks to the normal driving state of the vehicle while the vehicle is driving.

[0040] Specifically, vehicle information is obtained from the driving control system. The real-time speed and direction of the vehicle are extracted from the vehicle information, and the real-time speed and direction of the vehicle are analyzed using the optical flow method to obtain the actual motion trajectory of the vehicle. The actual motion trajectory is matched with the potential static motion trajectory and the potential motion trajectory respectively, and the static target corresponding to the potential static motion trajectory that intersects with the actual motion trajectory is determined as a static risk target, and the dynamic target corresponding to the potential motion trajectory that intersects with the actual motion trajectory is determined as a dynamic risk target. A feature recognition algorithm is used to analyze several environmental area images to obtain the dynamic target volume of the dynamic risk target and the dynamic target position of the dynamic risk target, the static target volume of the static risk target and the static target position of the static risk target. The static target volume, dynamic target position, static target position and dynamic target volume are included in the risk obstacle information.

[0041] The motion information determination module 203 is used to determine the motion information of the risk obstacle according to the real-time environment image and the risk obstacle information.

[0042] The motion information may be information generated when an object is in motion and has a real impact and risk on the normal driving of the vehicle.

[0043] Specifically, the target recognition algorithm is used to analyze the dynamic risk targets and static risk targets in the environmental area images obtained in the above steps to obtain the dynamic movement speed and dynamic movement direction of the dynamic risk targets, the static movement speed and static movement direction of the static risk targets. The dynamic movement speed, dynamic movement direction, static movement speed and static movement direction are included in the movement information.

[0044] The steering information determination module 204 is used to determine steering assistance information according to motion information, road wetness, vehicle information and real-time environmental images, and perform auxiliary steering operations according to the steering assistance information.

[0045] The steering assist information may be information determined when assisting operation is performed on the steering state of the vehicle to ensure safe steering of the vehicle when the vehicle encounters risks during normal driving.

[0046] Specifically, when a vehicle encounters a risk and makes a turn, the actual road conditions, the vehicle's own factors, and the actual state of the risk obstacles will affect the vehicle driver's judgment and operation on safe steering, resulting in the driver's steering operation being unable to cope with the actual road conditions.

[0047] Therefore, the vehicle width is extracted from the vehicle information. The road slipperiness, static motion speed, dynamic motion speed, vehicle width, static target volume and dynamic target volume obtained in the above steps are analyzed using a mathematical analysis method to obtain the real-time steering speed of the vehicle when turning, and the real-time steering speed is included in the steering assistance information. The driving control system controls the steering of the vehicle according to the steering assistance information.

[0048] Through this solution, the risk area determination module obtains the real-time environmental image of the moving vehicle, identifies the dynamic risk area and evaluates the slippery degree of the road, ensuring accurate determination of the risk area. The obstacle information determination module combines the real-time environmental image and vehicle information to analyze and identify potential risk obstacles, and provides accurate risk obstacle positioning and feature information. The motion information determination module determines the motion information of the risk obstacle through the real-time environmental image and risk obstacle information, so as to judge the threat it poses to the normal driving of the vehicle. The steering information determination module generates accurate steering assistance information based on the motion information of the risk obstacle, the real-time environmental image, the slippery degree of the road and the vehicle information, to assist the driver in making effective steering and avoiding collision risks.

[0049] In some embodiments, when the risk area determination module 201 determines the dynamic risk area and the degree of road slipperiness based on the real-time environmental image, it determines the texture uniformity of the road during the vehicle's driving, the degree of road reflection, a number of static targets, a number of dynamic targets, the locations of a number of static targets, and the motion trajectories of a number of dynamic targets based on the real-time environmental image; determines the degree of road slipperiness based on the texture uniformity and the degree of road reflection; predicts the static motion trends of a number of static targets based on the locations and the real-time environmental image; predicts the dynamic motion trends of a number of dynamic targets based on the real-time environmental image and the motion trajectories; determines whether the dynamic motion trends and the static motion trends affect the normal driving of the vehicle, and if so, determines the dynamic position coordinates of the dynamic targets and the static position coordinates of the static targets based on the dynamic motion trends and the static motion trends; determines the target formation range based on the dynamic position coordinates and the static position coordinates, and determines the target formation range as the dynamic risk area.

[0050] The texture uniformity may be a value representing a distribution state of the texture of the surface of the road on which the vehicle is traveling.

[0051] The degree of road reflectivity may be a numerical value indicating the intensity of light reflection from the surface of the road on which the vehicle is traveling when the vehicle is traveling. The static motion trend may be a tendency or possibility generated when a static target that may potentially affect or pose a risk to the normal driving of the vehicle begins to move. The dynamic motion trend may be a trend generated when a dynamic target that may potentially affect or pose a risk to the normal driving of the vehicle is already in motion and exhibits a continuous or changing direction and speed of motion. The dynamic position coordinates may be the coordinates of a moving target object that may affect the normal driving of the vehicle relative to the position of the vehicle itself. The static position coordinates may be the coordinates of a static moving target that may affect the normal driving of the vehicle relative to the position of the vehicle itself. The target formation range may be the area formed by connecting the dynamic position coordinates of a dynamic target that may potentially affect the normal driving of the vehicle and the static position coordinates of a static target in a certain order.

[0052] Specifically, the real-time environment image is analyzed using an image analysis algorithm to obtain the texture uniformity and road reflectivity of the road on which the vehicle is traveling. A spatial rectangular coordinate system is established with the vehicle itself as the coordinate origin. The real-time environment image is analyzed using a target detection algorithm to obtain the location of the static target. The real-time environment information is analyzed using a target tracking algorithm to obtain the motion trajectory of the dynamic target. The texture uniformity and the road reflectivity are analyzed using a mathematical analysis method to obtain the degree of road slipperiness. Based on the motion trajectory of the dynamic target, the dynamic objects in the real-time environment image are analyzed using an optical flow method to obtain the dynamic motion trend of the dynamic target. Based on the location, the static targets in the real-time environment image are analyzed using an image registration technology to obtain the static motion trend of the static target. The static motion trend, the actual motion trajectory of the vehicle and the dynamic motion trend are marked in the spatial rectangular coordinate system obtained in the above steps, and it is determined whether the actual motion trajectory has an intersection with the dynamic motion trend and the static motion trend. If an intersection is generated, the position of the dynamic target in the spatial rectangular coordinate system is determined as the dynamic position coordinate based on the dynamic motion trend. Based on the static motion trend, the position of the static target in the spatial rectangular coordinate system is determined as the static position coordinate. The static position coordinate and the dynamic position coordinate at the edge are connected to obtain a coordinate area, and while connecting, all the static position coordinates and dynamic position coordinates are ensured to be within this coordinate area. This coordinate area is the target formation range, and the target formation range is determined as the dynamic risk area.

[0053] Through this solution, the texture uniformity and reflectivity of the road are analyzed through real-time environmental images. These factors directly affect the slipperiness of the road, so as to determine whether there is a risk of sliding. The positions and movement trajectories of several static and dynamic targets are identified and analyzed, and the movement trends of these targets in the future are predicted. In particular, for static targets, their static movement trends are predicted by analyzing their positions and real-time environmental images; for dynamic targets, their dynamic movement trends are predicted based on their movement trajectories. Determine whether the movement trends of these targets pose a threat to vehicle driving. If there is a potential risk, further calculate the precise position coordinates of the target. Combining the movement trajectories and position coordinates of dynamic and static targets, a target formation range is determined. This range is the dynamic risk area.

[0054] In some embodiments, when the obstacle information determination module 202 determines the risk obstacle information based on the dynamic risk area and vehicle information, it determines the static relative positions of several static targets, the dynamic relative positions of several dynamic targets, the movement speeds of several dynamic targets, and the movement directions of several dynamic targets in the dynamic risk area based on the real-time environmental image; based on the static relative position and the dynamic relative position, according to the movement speed, movement direction, driving speed, and driving direction, it determines the static risk targets and dynamic risk targets that affect the vehicle's driving, and determines the static risk targets and dynamic risk targets as risk obstacles; according to the real-time environmental image, it determines the first position of the static risk target, the static target volume of the static risk target, the first relative position of the dynamic risk target, and the dynamic target volume of the dynamic risk target; and incorporates the first position, the static target volume, the first relative position, and the dynamic target volume into the risk obstacle information.

[0055] The static relative position may be the position of a static risk target that has an impact or risk on the normal driving of the vehicle relative to the vehicle itself. The dynamic relative position may be the position of a dynamic risk target that has an impact or risk on the normal driving of the vehicle relative to the vehicle itself. A risk obstacle may be information corresponding to an object that has an impact or risk on the normal driving of the vehicle when it is moving or stationary. A static risk target may be an object that is stationary and has an impact or risk on the normal driving of the vehicle. A dynamic risk target may be an object that is in motion and has an impact or risk on the normal driving of the vehicle. The first position may be the position of a static risk target relative to the vehicle. The first relative position may be the position of a dynamic risk target relative to the vehicle.

[0056] Specifically, the target detection algorithm is used to analyze the real-time environment image to obtain the static relative position of the static target and the dynamic relative position of the dynamic target in the dynamic risk area. The target tracking algorithm is used to analyze the real-time environment image to obtain the movement speed and movement direction of the dynamic target. The driving speed and driving direction of the vehicle are extracted from the vehicle information. The driving direction and the movement direction are respectively marked in the spatial rectangular coordinate system obtained in the above steps, and it is determined whether the driving direction and the movement direction have a direction intersection. If a direction intersection is generated, the driving direction and the movement direction are analyzed respectively using a mathematical analysis method to obtain the dynamic direction distance between the dynamic target and the direction intersection and the vehicle direction distance between the vehicle and the direction intersection. The dynamic direction distance and the vehicle direction distance are divided by the movement speed and the driving speed respectively to obtain the dynamic intersection time when the dynamic target arrives at the direction intersection and the vehicle intersection time when the vehicle arrives at the direction intersection. It is determined whether the dynamic intersection time and the vehicle intersection time are consistent. If they are consistent, it is determined that the dynamic target corresponding to the dynamic intersection time causes risk to the vehicle driving, and the dynamic target corresponding to the dynamic intersection time is determined as a dynamic risk target. Mark the static relative position in the spatial rectangular coordinate system obtained in the above steps, and determine whether the driving direction intersects with the static relative position. If so, the static target corresponding to the static relative position is determined as a static risk target. Use the image analysis algorithm to analyze the real-time environmental image to obtain the first position of the static risk target and the static target volume of the static risk target, the first relative position of the dynamic risk target and the dynamic target volume of the dynamic risk target. The first position, static target volume, first relative position and dynamic target volume are included in the risk obstacle information.

[0057] Through this solution, the relative positions and motion characteristics of static targets and dynamic targets in the dynamic risk area are identified through real-time environmental images, with a focus on the static relative positions of static targets and the dynamic relative positions of dynamic targets. The movement speed and direction of the dynamic targets are obtained, and combined with the vehicle's driving speed and direction, it is determined whether these dynamic targets will affect the vehicle's driving. If certain static or dynamic targets pose a threat, they are marked as risk obstacles. Through further analysis, the first position and static target volume of the static risk target, as well as the first relative position and dynamic target volume of the dynamic risk target, are accurately determined, and then this information is integrated into risk obstacle information. While improving driving safety, it accurately identifies and responds to potential risk obstacles, providing a key basis for vehicle automation decision-making.

[0058] In some embodiments, when the motion information determination module 203 determines the motion information of the risk obstacle based on the real-time environmental image and the risk obstacle information, it determines the static real-time motion speed and the static real-time motion direction of the static risk target when it moves at the first position based on the real-time environmental image; determines the motion speed as the dynamic motion speed of the dynamic risk target when it is at the first relative position, and determines the motion direction as the dynamic motion direction of the dynamic risk target when it is at the first relative position; and incorporates the dynamic motion speed, dynamic motion direction, static real-time motion speed and static real-time motion direction into the motion information.

[0059] The static real-time motion speed may be the speed corresponding to the motion of the static risk target that affects the normal driving of the vehicle. The static real-time motion direction may be the direction corresponding to the motion of the static risk target that affects the normal driving of the vehicle. The dynamic motion speed may be the speed corresponding to the motion of the dynamic risk target that affects the normal driving of the vehicle. The dynamic motion direction may be the direction corresponding to the motion of the dynamic risk target that affects the normal driving of the vehicle.

[0060] Specifically, in the dynamic risk area, the movement speed and direction of dynamic targets and static targets at corresponding positions that have an impact or risk on the normal driving of the vehicle will affect the driver's safe steering judgment and operation.

[0061] Therefore, the optical flow method is used to analyze the real-time environment image to obtain the static real-time motion speed and static real-time motion direction of the static risk target when it moves at the first position. The motion speed of the dynamic target obtained in the above steps is determined as the dynamic motion speed of the dynamic risk target at the first relative position, and the motion direction of the dynamic target obtained in the above steps is determined as the dynamic motion direction of the dynamic risk target at the first relative position. The dynamic motion speed, dynamic motion direction, static real-time motion speed and static real-time motion direction are included in the motion information.

[0062] Through this solution, the movement of static risk targets is analyzed through real-time environmental images to determine their static real-time movement speed and static real-time movement direction at the first position. For dynamic risk targets, their dynamic movement speed and dynamic movement direction are calculated based on their changes in the first relative position. Combining this information, the movement status of static risk targets and dynamic risk targets at a specific time point is accurately depicted. According to the relative position, movement speed and movement direction of the dynamic risk target, its movement prediction is adjusted in time. After integrating these data, comprehensive movement information is formed, which provides accurate movement trajectory prediction for subsequent risk assessment and dynamic decision-making, greatly improves the vehicle's ability to respond to potential obstacles, and effectively supports the safety of driving steering assistance.

[0063] In some embodiments, when the motion information determination module 203 determines the static real-time motion speed and static real-time motion direction of the static risk target when it moves at the first position based on the real-time environmental image, it determines the dynamic motion targets in the vehicle motion perception area, the dynamic target speed of the dynamic motion targets, the target motion direction of the dynamic motion targets, the motion target volume of the dynamic motion targets, the static target volume of the static risk targets and the static target type of the static risk targets based on the real-time environmental image; determines the aerodynamic flow intensity of the dynamic motion targets based on the dynamic motion speed, the dynamic motion target volume and the dynamic motion direction; determines the static target weight of the static target based on the static target volume and the static target type; determines whether the static target moves based on the aerodynamic flow intensity, the position and the static target weight; if the static target moves, determines the static real-time motion speed and static real-time motion direction of several static targets when they start to move based on the real-time environmental image.

[0064] The motion sensing area may be the area range presented in the real-time environment image. The dynamic motion target may be a target that is moving in the motion sensing area but does not cause an impact or risk to the normal driving of the vehicle. The dynamic target speed may be the speed corresponding to the object that is moving in the motion sensing area but does not cause an impact or risk to the normal driving of the vehicle. The target motion direction may be the direction corresponding to the object that is moving in the motion sensing area and does not cause an impact or risk to the normal driving of the vehicle. The motion target volume may be the volume corresponding to the object that is moving in the motion sensing area and does not cause an impact or risk to the normal driving of the vehicle. The static target volume may be the volume corresponding to the static target that affects the normal driving of the vehicle. The static target type may be the type corresponding to the static target that affects the normal driving of the vehicle. The aerodynamic vortex intensity may be a value corresponding to the strength of the change in air flow caused by the movement of the dynamic motion target when the vehicle is driving. The static real-time motion speed may be the speed at which the static target moves due to the aerodynamic vortex intensity. The static real-time motion direction may be the direction in which the static target moves due to the aerodynamic vortex intensity.

[0065] Specifically, a feature recognition algorithm is used to analyze the real-time environmental image to obtain the dynamic motion target in the motion perception area, the dynamic target speed of the dynamic motion target, the target motion direction of the dynamic motion target, the motion target volume of the dynamic motion target, the static target volume of the static risk target, and the static target type of the static risk target. The dynamic motion speed, the dynamic motion target volume, and the dynamic motion direction are analyzed using a mathematical analysis method to obtain the aerodynamic flow intensity of the dynamic motion target. The types of all possible static risk targets and their corresponding densities are obtained from the driving control system, and the types of all possible static risk targets and their corresponding densities are integrated to obtain a preset type density set. The static target type obtained in the above steps is matched with the preset type density set to obtain the object density corresponding to the static target type, and the object density corresponding to the static target type is determined as the static target density. The static target density is multiplied by the static target volume to obtain the static target weight. All aerodynamic flow intensities and the weight of objects they can drive are obtained from the airflow research website, and all aerodynamic flow intensities and the weight of objects they can drive are integrated to obtain a preset strength weight set. The preset intensity weight set is matched with the static target weight. If the aerodynamic flow intensity corresponding to the static target weight is greater than or equal to the aerodynamic flow intensity obtained in the above steps, the target tracking algorithm is used to analyze the real-time environment image to obtain the static real-time motion speed and static real-time motion direction.

[0066] Through this solution, dynamic moving targets in the vehicle motion perception area are identified based on real-time environmental images, and the key features of dynamic moving targets are extracted. By analyzing the key features of dynamic moving targets, their aerodynamic flow intensity is calculated. The volume and type information of static risk targets are extracted and used to calculate the weight of their static targets. By combining the aerodynamic flow intensity, location and static target weight, it is determined whether the static target has moved. If the static target has moved, its static real-time motion speed and static real-time motion direction are further re-evaluated based on the real-time environmental image to ensure accurate tracking of the motion state of the static target. By comprehensively analyzing the various physical characteristics of dynamic targets and static targets, the potential risks in the environment can be accurately judged, effectively improving the safety and responsiveness of the intelligent driving system.

[0067] In some embodiments, when the steering information determination module 204 determines the steering assistance information according to the motion information, the road slipperiness, the vehicle information and the real-time environment image, it determines whether the normal driving state of the vehicle is affected according to the motion trajectory and the vehicle motion trajectory; if affected, the dynamic relative distance between the vehicle and the dynamic risk target, the dynamic acceleration of the dynamic risk target and the dynamic relative acceleration between the dynamic risk target and the vehicle are determined according to the real-time environment image; the impact moment of the dynamic risk target affecting the normal driving state and the steering moment when the driver starts to turn are obtained, and the driver's reaction time is determined according to the impact moment and the steering moment; according to the driving speed, dynamic relative acceleration, vehicle acceleration, real-time weight, reaction time, dynamic relative distance, dynamic acceleration, road slipperiness and dynamic target volume, the steering avoidance angle of the vehicle is determined according to formula (1), and the steering avoidance angle is included in the steering assistance information: (1); in, Indicates the steering avoidance angle, Represents the reaction time, represents the dynamic acceleration, Indicates the slipperiness of the road. represents the dynamic target volume, Represents the dynamic relative distance, Indicates the driving speed, represents the vehicle acceleration, Represents the reaction time, Indicates the real-time weight, Indicates dynamic relative acceleration.

[0068] The dynamic relative distance may be the distance of a dynamic risk target that has an impact or risk on the normal driving state of the vehicle relative to the vehicle when in motion. The dynamic acceleration may be the acceleration of a dynamic risk target that has an impact or risk on the normal driving state of the vehicle relative to the vehicle when in motion. The dynamic relative acceleration may be the difference between the acceleration of a dynamic risk target that has an impact or risk on the normal driving state of the vehicle and the acceleration of the vehicle itself when in motion. The turning moment may be the moment when the driver starts to perform a turning operation when the vehicle is avoiding an obstacle. The impact moment may be the moment corresponding to when a risk obstacle starts to affect the normal driving state of the vehicle. The reaction time may be the time from when a risk obstacle starts to affect the normal driving of the vehicle to when the driver starts to perform a turning operation. The steering avoidance angle may be the angle corresponding to the steering that enables the vehicle's steering avoidance to reach a safe state.

[0069] Specifically, the vehicle motion trajectory, the real-time weight of the vehicle and the vehicle acceleration are extracted from the vehicle information. The vehicle motion trajectory is matched with the motion trajectory to determine whether the motion trajectory and the vehicle motion trajectory have an intersection. If so, it is determined that the normal driving state of the vehicle is affected. The real-time environmental image is analyzed using a target tracking algorithm to obtain the dynamic relative distance between the vehicle and the dynamic risk target, the dynamic acceleration of the dynamic risk target and the dynamic relative acceleration between the dynamic risk target and the vehicle. The moment when the dynamic risk target begins to affect the normal driving state and the turning moment when the driver begins to turn are obtained from the driving control system, and the moment when the dynamic risk target begins to affect the normal driving state is determined as the impact moment. The steering moment is subtracted from the impact moment to obtain the driver's reaction time. The driving speed, dynamic relative acceleration, vehicle acceleration, real-time weight, vehicle volume, reaction time, dynamic relative distance, dynamic acceleration, road slipperiness and dynamic target volume are analyzed using a mathematical analysis algorithm, and the steering avoidance angle of the vehicle is determined according to formula (1), and the steering avoidance angle is included in the steering assistance information.

[0070] Among them, the numerator of formula (1) middle: It determines the initial space between the vehicle and the dynamic risk target. The closer the distance, the more difficult it is to avoid and the larger the avoidance angle is required. It reflects the change in the movement speed of the dynamic risk target. The faster the dynamic risk target, the larger the steering angle is required to avoid collision. The larger the dynamic risk target, the more space it will occupy, and the more space is needed to avoid it when turning. Therefore, the introduction of the dynamic target volume is to reflect the impact of the physical existence of the dynamic moving target on the avoidance angle. The larger it is, the greater the avoidance angle the vehicle needs.

[0071] Denominator middle: The larger the angle, the more difficult it is for the vehicle to avoid an obstacle. When driving at high speeds, the vehicle's reaction time to the surrounding environment will be affected, so a larger steering angle is required to ensure safety. It indicates the change in the vehicle's motion state within the driver's reaction time, helping to calculate whether the vehicle can avoid dynamic risk targets in time. middle: It affects the friction between the vehicle and the road. The larger the coefficient of friction, the greater the friction and the better the vehicle's steering control. Conversely, a heavier vehicle may need a larger avoidance angle when encountering danger. A higher coefficient of friction means the road is less slippery and the vehicle is easier to control. Conversely, slippery roads can make steering difficult, so it needs to be taken into account in the calculation. . Reflects the change in relative motion between the vehicle and the target. The greater the acceleration between the dynamic risk target and the vehicle, the greater the angle required for avoidance. (Inverse tangent function) is used to calculate the steering avoidance angle , which is determined by the ratio of the target's influencing factors in the numerator to the vehicle's control capability factors in the denominator. This calculation method conforms to the trigonometric function relationship in physical motion and reflects the angle that the vehicle needs to turn during the avoidance process.

[0072] Through this solution, by comparing the vehicle's motion trajectory with the motion trajectory, it is determined whether an abnormality occurs; if affected, the relative distance between the vehicle and the dynamic risk target, the dynamic acceleration, and the dynamic relative acceleration relative acceleration are evaluated, and the driver's reaction time is calculated. Through these data, combined with multiple factors such as the vehicle's driving speed and acceleration, the best steering avoidance angle is calculated using a formula to ensure that the risk target can be avoided in time in critical moments. The various parameters involved in the formula help simulate the complex motion state in real situations, making the steering angle decision more accurate. These data will be converted into steering assistance information and provided to the driver to assist him in making a faster and more effective response, thereby effectively enhancing driving safety, especially in complex or dangerous driving environments.

[0073] In some embodiments, when determining the steering assistance information according to the motion information, the road slipperiness, the vehicle information and the real-time environment image, the steering information determination module 204 determines the road edge curvature, the real-time position of the vehicle on the road and the real-time interval distance between the vehicle and the road edge when the vehicle is turning according to the road edge image; determines the real-time turning radius of the vehicle when turning according to the road edge curvature, the real-time position, the driving speed and the real-time interval distance; determines the turning speed of the vehicle when turning according to the driving speed, the dynamic motion speed and the dynamic target volume; determines the turning speed reduction of the vehicle when turning according to the turning speed and the driving speed; determines the turning time of the vehicle according to the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, and incorporates the turning time and the turning speed into the steering assistance information: (2); in, Indicates the turning time. Indicates the width of the vehicle, represents the dynamic target volume, Indicates the preset volume avoidance coefficient, Indicates the driving speed, Represents the reaction time, Indicates the dynamic motion speed, Indicates the real-time turning radius, Indicates the slipperiness of the road. Indicates the amount of steering speed reduction.

[0074] The road edge image may be an image of the edge of the road on which the vehicle is traveling when the normal driving of the vehicle is affected by a risk obstacle. The preset volume avoidance coefficient may be a value indicating the influence of the volume of the dynamic risk target on the calculation result of the turning time. The road edge curvature may be a value indicating the degree of curvature of the road edge line. The real-time position may be the position of the vehicle width on the overall width of the road. The real-time interval distance may be the distance between the wheels of the vehicle and the edge of the road when the vehicle is turning to avoid.

[0075] The real-time turning radius may be the turning radius of the vehicle when the vehicle is performing a steering avoidance. The steering speed may be the driving speed of the vehicle when the vehicle is performing a steering avoidance. The steering duration may be the duration of time the vehicle spends performing a steering avoidance to ensure vehicle driving safety.

[0076] Specifically, when a vehicle is turning to avoid an obstacle, it needs to plan the vehicle's driving path based on the lane lines of the road. However, when there are no lane lines on the road, the path planning of the vehicle when turning is difficult to adapt to the actual road traffic conditions. Therefore, a road edge image is obtained from the real-time environment image, and the road edge image is analyzed using an image analysis algorithm to obtain the road edge curvature, the real-time position of the vehicle on the road, and the real-time interval between the vehicle and the road edge when the vehicle is turning. The vehicle width is extracted from the vehicle information. The road edge curvature, real-time position, driving speed and real-time interval are analyzed using a mathematical analysis algorithm to obtain the real-time turning radius of the vehicle when turning. The driving speed, dynamic motion speed and dynamic target volume are analyzed using a mathematical analysis algorithm to determine the turning speed of the vehicle when turning. The driving speed is subtracted from the turning speed to obtain the turning speed reduction. The vehicle acceleration, turning speed reduction, dynamic motion speed, dynamic target volume, vehicle width, real-time turning radius, road slipperiness and driving speed are analyzed using a mathematical analysis method, and the turning time is determined according to formula (2), and the turning time and turning speed are included in the steering assistance information.

[0077] Among them, in formula (2), the numerator is: This directly affects the space the vehicle needs when turning. The wider the vehicle, the longer it takes to turn. The most basic requirement that directly affects the turning time. is the dynamic target volume With driving speed function, and by presetting the volume avoidance coefficient When turning, the vehicle not only needs to consider its own width, but also needs to avoid according to the dynamic volume of the surrounding environment. Therefore, the difficulty of avoidance will affect the turning time. The larger the avoidance target, the longer the turning time will be. Medium reaction time Directly affects the driver or system's ability to adjust steering. Real-time turning radius The smaller it is, the longer the vehicle needs to turn. This affects the friction, The smaller it is, the greater the resistance the vehicle has when turning, resulting in an increase in turning time.

[0078] Denominator: Steering speed reduction It reflects the dynamic behavior of the vehicle when turning. If the vehicle slows down too much, the turning time will be prolonged. The larger it is, the longer the vehicle will take to turn.

[0079] Through this solution, the road curvature, the real-time position of the vehicle, and the real-time distance from the road edge are obtained through the road edge image, and then the real-time turning radius of the vehicle when turning is calculated. At the same time, combined with factors such as driving speed and dynamic target volume, the turning speed is determined, and the reduction in turning speed is evaluated. These data are combined with factors such as vehicle acceleration to further calculate the turning time to ensure that obstacle avoidance behavior in complex environments fully considers the changes in turning time and speed. The various parameters in the formula help simulate the dynamic changes during actual driving, thereby providing accurate steering assistance information, helping drivers make safe and efficient steering decisions under complex road and environmental conditions, and improving driving safety and operational responsiveness.

[0080] In some embodiments, when determining the turning time of the vehicle according to the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, the turning information determination module 204 determines whether the road edge boundary is clear according to the real-time environment image; if the road edge boundary is not clear, the object volume and the object position in the shoulder area are determined according to the real-time environment image; the actual road width is determined according to the object volume and the object position; the turning time is determined according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed according to formula (3): (3); in, Indicates the turning time. Indicates the actual width of the road. Indicates the width of the vehicle, represents the dynamic target volume, Indicates the driving speed, Represents the reaction time, Indicates the dynamic motion speed, Indicates the real-time turning radius, Indicates the slipperiness of the road. Indicates the reduction in steering speed, Represents the volume of an object, Indicates the vehicle acceleration.

[0081] The shoulder area may be a buffer area on both sides of the main driving lane of the road on which the vehicle is traveling. The object volume may be the volume corresponding to the object in the shoulder area when the vehicle is turning to avoid an obstacle. The object position may be the position of the object in the shoulder area relative to the vehicle when the vehicle is turning to avoid an obstacle.

[0082] Specifically, the design width of the current road when the vehicle is traveling on the current road is obtained from the traffic information website. The real-time environment image is analyzed using an image analysis algorithm to obtain the actual width of the road. The actual width of the road is compared with the design width. If the actual width of the road is less than the design width, it is determined that the edge boundary of the road is unclear. The real-time environment image is analyzed using a feature recognition algorithm to obtain the object volume and object position in the shoulder area. Based on the object volume, the distance between the object positions in the shoulder area is determined as the actual width of the road. The mathematical analysis algorithm is used to analyze the vehicle acceleration, actual road width, object volume, steering speed reduction, dynamic motion speed, dynamic target volume, vehicle width, real-time turning radius, road slipperiness and driving speed, and the steering time is determined according to formula (3).

[0083] Among them, in formula (3), the numerator is: Comprehensively consider the contribution of various influencing factors to the turning time. It directly affects the feasible space for turning. The narrower the road width, the longer it takes for the vehicle to turn. Reflects the size of the shoulder or obstacle. Larger objects will take up more road space, forcing vehicles to avoid them, thereby increasing the time it takes to turn. Taking into account the dynamic motion of the vehicle, when the dynamic motion speed of the vehicle is high or the acceleration is low, the turning time of the vehicle will increase accordingly. reflects the rate of change of the vehicle, while the vehicle acceleration It indicates the vehicle's ability to change speed. It can be regarded as a correction to the dynamic state of the vehicle when turning. It reflects the effect of the size of the object that the vehicle needs to avoid at the current speed on the turning time. Larger objects require more time to avoid. : Usually, the turning radius The larger it is, the shorter the steering time is, because the vehicle can turn at a larger angle, reducing the time to avoid obstacles. This means that the time required for a vehicle to turn decreases as the speed increases. This indicates that slippery roads make it harder to control a vehicle, so it takes more time to turn. When the slipperiness is high, the vehicle reacts more slowly when turning, and acceleration limits cause the turning time to increase. Wider vehicles also require longer turning times to ensure avoidance safety.

[0084] Denominator: The larger it is, the more time the vehicle needs to slow down and complete the turn. It is an important indicator of speed change during the dynamic adjustment of the vehicle. It interacts with multiple factors such as road slipperiness, acceleration, steering angle, etc., and jointly determines the steering time of the vehicle during the steering process.

[0085] Through this solution, when the edge of the road is unclear, objects in the shoulder area are identified through real-time environmental image analysis, and the actual width of the road is estimated. By comprehensively considering variables such as vehicle acceleration, object volume, road wetness, and real-time turning radius, the turning time required for steering can be accurately determined. The various parameters in the formula, such as dynamic target volume, vehicle width, turning radius, etc., reflect the actual status of the road and vehicle, ensuring that the calculation of turning time can adapt to different environmental conditions. The accurate calculation of turning time takes into account the road conditions, object obstacles, and vehicle motion characteristics, effectively optimizes the driver's operational response, and improves safety and driving comfort. Through dynamic adjustment and accurate prediction, the steering assist function under complex or unclear road conditions is enhanced.

[0086] In some embodiments, after determining the turning time according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic movement speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, the turning information determination module 204 determines the first turning time according to the actual road width, the vehicle acceleration, the turning speed reduction, the dynamic movement speed, the dynamic target volume, the vehicle width, the road slipperiness and the driving speed by calculating according to formula (4): (4); in, Indicates the duration of the first turn. Indicates the actual width of the road. represents the dynamic target volume, Indicates the preset volume safety factor, Indicates the driving speed, Indicates the dynamic motion speed, Represents the reaction time, represents the vehicle acceleration, Indicates the slipperiness of the road. Indicates the amount of steering speed reduction.

[0087] The first turning time may be the length of time it takes for the vehicle to make a secondary turn after it encounters a risk obstacle and needs to turn to avoid it.

[0088] Specifically, a mathematical analysis algorithm is used to analyze the actual road width, vehicle acceleration, steering speed reduction, dynamic motion speed, dynamic target volume, vehicle width, road slipperiness and driving speed, and the first steering time is determined by calculation according to formula (4).

[0089] In formula (4), the numerator is middle, It directly determines the space when the vehicle turns. When turning, the vehicle needs to avoid obstacles and other objects around it, so the smaller the road width, the longer the turning time. Since the second turn occurs after the first turn, the road width may become smaller or occupied by objects, so it is necessary to re-evaluate the road width available to the vehicle when turning. middle, The larger the value, the more time the vehicle needs to avoid obstacles when turning. In practice, vehicles usually add some safety space to obstacles, so the moving target volume is multiplied by a safety factor . Driving speed The faster it is, the shorter the time required to turn, so The combined effect of object size and vehicle speed is taken into account. Larger targets take longer to avoid, but at the same time, faster vehicle speeds reduce the time required to steer. middle, Reflects the relative speed of the vehicle to the dynamic risk target. If the dynamic risk target is approaching or the vehicle itself is moving faster, it will take longer to avoid it. The longer the vehicle is, the more time it takes to turn. The vehicle's acceleration ability affects the time it takes to turn. The larger it is, the faster the vehicle can change its state of motion, so the turning time will be shortened accordingly. Indicates the effect of road slipperiness on the vehicle's ability to turn. Slippery roads reduce the vehicle's grip, which increases the time it takes to turn. If the road is slipperier, the vehicle's ability to turn is weakened and it takes more time to complete the turn.

[0090] Denominator: It is a measure of the change in vehicle speed during the turning process, reflecting the degree of vehicle deceleration during the turning process. When turning, if the turning speed decreases by a large amount, the vehicle needs more time to complete the deceleration and turning actions, thus increasing the turning time.

[0091] Through this solution, the first turning time is calculated according to a formula that incorporates key parameters such as road width, vehicle speed, dynamic target volume, and reaction time, reflecting the impact of different driving conditions on the turning time. Specifically, the variables in the formula reflect environmental conditions (such as wetness, road width) and the characteristics of the vehicle itself (such as acceleration and width). Through reasonable parameter adjustment, the time required for turning is accurately estimated to ensure the smoothness and safety of the vehicle's steering action. In this way, the vehicle can improve the accuracy of steering control in complex driving environments and enhance the dynamic stability and safety of the vehicle.

[0092] In some embodiments, when the steering information determination module 204 determines the steering assist information based on the motion information, the road slipperiness, the vehicle information and the real-time environmental image, it determines the first slipperiness of the road after the vehicle turns based on the real-time environmental image; determines the first driving speed of the vehicle after the vehicle turns based on the vehicle information; determines the real-time offset of the center of gravity of the vehicle based on the real-time weight, the first driving speed and the first slipperiness; obtains the center of gravity offset safety range, matches the real-time offset degree with the center of gravity offset safety range, and determines whether the real-time offset degree is within the center of gravity offset safety range; if the real-time offset degree is not within the center of gravity offset safety range, determines the offset degree difference based on the real-time offset degree and the center of gravity offset safety range; determines the maximum turning speed of the vehicle based on the offset degree difference, the first driving speed and the real-time weight; determines the secondary steering angle of the vehicle after the turn based on the maximum steering speed, the first slipperiness and the real-time weight, and incorporates the secondary steering angle into the steering assist information.

[0093] The first degree of slipperiness may be a value indicating the degree of slippage of the road on which the vehicle is traveling after the vehicle turns to avoid a risk obstacle. The first driving speed may be the speed at which the vehicle is traveling after turning to avoid a risk obstacle. The real-time degree of deviation may be the distance by which the center of gravity of the vehicle deviates from its ideal or designed center of gravity position after the vehicle turns to avoid a risk obstacle. The center of gravity deviation safety range may be the range corresponding to the degree of center of gravity deviation that puts the vehicle in a safe driving state when the vehicle is traveling. The maximum steering speed may be the speed that ensures that the vehicle is in a safe driving state when turning when the vehicle turns to avoid a risk obstacle. The secondary steering angle may be the angle at which the vehicle needs to perform another steering operation when the degree of slipperiness of the road changes suddenly after turning to avoid a risk obstacle.

[0094] Specifically, a feature recognition algorithm is used to analyze the real-time environmental image to obtain the first slippery degree of the road after the vehicle turns. The first driving speed is extracted from the vehicle information. A mathematical analysis method is used to analyze the real-time weight, the first driving speed and the first slippery degree to obtain the real-time offset degree. The center of gravity offset safety range is obtained from the driving control system, and the real-time offset degree is matched with the center of gravity offset safety range. If the real-time offset degree is not within the center of gravity offset safety range, the maximum center of gravity offset degree is extracted from the center of gravity offset safety range, and the maximum center of gravity offset degree is subtracted from the real-time offset degree to obtain the offset degree difference. A mathematical analysis method is used to analyze the offset degree difference, the first driving speed and the real-time weight to obtain the maximum turning speed. A mathematical analysis method is used to analyze the maximum turning speed, the real-time turning radius, the first slippery degree and the real-time weight, and the secondary steering angle is determined according to formula (5), and the secondary steering angle is incorporated into the steering assist information: (5); in, represents the secondary steering angle, Indicates the maximum turning speed, Indicates the first slippery degree, Indicates the real-time weight, Indicates the real-time turning radius.

[0095] Through this solution, the first slippery degree of the road after the vehicle turns is determined based on the real-time environmental image, laying the foundation for subsequent steering adjustments; the first driving speed is calculated using vehicle information to reflect vehicle state changes; the real-time offset degree of the vehicle's center of gravity is determined in combination with the real-time weight, slippery degree and first driving speed to evaluate steering safety. By matching the real-time offset degree with the safe range of the center of gravity offset, the vehicle stability is effectively monitored. If the offset exceeds the safe range, the offset degree difference is further calculated to provide data support for the maximum steering speed adjustment. According to the maximum steering speed, the first slippery degree and the real-time weight, the secondary steering angle is determined to make the vehicle steering more accurate and safe.

[0096] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. An automatic risk avoidance steering assistance system based on visual detection, characterized in that: The system comprises a risk area determination module, an obstacle information determination module, a motion information determination module and a turning information determination module, wherein the risk area determination module and the obstacle information determination module are sequentially connected, the motion information determination module and the turning information determination module are sequentially connected, the motion information determination module and the risk area determination module and the obstacle information determination module are respectively connected, the risk area determination module and the turning information determination module are sequentially connected, and the obstacle information determination module and the turning information determination module are sequentially connected: A risk area determination module is used to obtain a real-time environmental image of a moving vehicle and determine a dynamic risk area and a degree of road slipperiness based on the real-time environmental image; An obstacle information determination module, used to obtain vehicle information of the traveling vehicle, and determine risk obstacle information according to the dynamic risk area and the vehicle information; A motion information determination module, used to determine the motion information of the risk obstacle according to the real-time environment image and the risk obstacle information; The steering information determination module is used to determine the steering assistance information according to the motion information, the road wetness, the vehicle information and the real-time environment image, and perform auxiliary steering operation according to the steering assistance information.

2. The system according to claim 1, characterized in that When determining the dynamic risk area and the road slipperiness according to the real-time environment image, the risk area determination module is used to: Determine, based on the real-time environment image, the texture uniformity of the road, the degree of road reflectivity, a number of static targets, a number of dynamic targets, the locations of the number of static targets, and the motion trajectories of the number of dynamic targets during the driving process of the vehicle; Determining the slipperiness of the road according to the texture uniformity and the road reflectivity; Predicting static motion trends of the plurality of static targets according to the locations and the real-time environment image; Predicting the dynamic motion trends of the plurality of dynamic targets according to the real-time environment image and the motion trajectory; Determine whether the dynamic motion trend and the static motion trend affect the normal driving of the vehicle, and if so, determine the dynamic position coordinates of the dynamic target and the static position coordinates of the static target according to the dynamic motion trend and the static motion trend; A target formation range is determined according to the dynamic position coordinates and the static position coordinates, and the target formation range is determined as the dynamic risk area.

3. The system according to claim 2, characterized in that The vehicle information includes the driving speed and driving direction of the vehicle. When the obstacle information determination module determines the risk obstacle information according to the dynamic risk area and the vehicle information, it is used to: Determine, according to the real-time environment image, the static relative positions of the static targets, the dynamic relative positions of the dynamic targets, the movement speeds of the dynamic targets, and the movement directions of the dynamic targets in the dynamic risk area; Based on the static relative position and the dynamic relative position, according to the movement speed, the movement direction, the driving speed and the driving direction, determine the static risk target and the dynamic risk target that affect the driving of the vehicle, and determine the static risk target and the dynamic risk target as the risk obstacle; Determining, according to the real-time environment image, a first location of the static risk target, a static target volume of the static risk target, a first relative location of the dynamic risk target, and a dynamic target volume of the dynamic risk target; The first location, the static target volume, the first relative location, and the dynamic target volume are included in the risk obstacle information.

4. The system according to claim 3, characterized in that When the motion information determination module determines the motion information of the risk obstacle according to the real-time environment image and the risk obstacle information, it is used to: Determining, according to the real-time environment image, a static real-time motion speed and a static real-time motion direction of the static risk target when it moves at the first location; Determine the movement speed as the dynamic movement speed of the dynamic risk target when it is at the first relative position, and determine the movement direction as the dynamic movement direction of the dynamic risk target when it is at the first relative position; The dynamic motion speed, the dynamic motion direction, the static real-time motion speed and the static real-time motion direction are incorporated into the motion information.

5. The system according to claim 4, characterized in that When the motion information determination module determines the static real-time motion speed and the static real-time motion direction of the static risk target when it moves at the first position according to the real-time environment image, it is used to: Determine, according to the real-time environment image, a dynamic moving target in a vehicle motion perception area, a dynamic target speed of the dynamic moving target, a target motion direction of the dynamic moving target, a moving target volume of the dynamic moving target, a static target volume of the static risk target, and a static target type of the static risk target; Determining the aerodynamic flow intensity of the dynamic motion target according to the dynamic motion speed, the dynamic motion target volume and the dynamic motion direction; Determining a static target weight of the static target according to the static target volume and the static target type; Determining whether the static target is moving according to the aerodynamic flow intensity, the position and the weight of the static target; If the static targets move, the static real-time movement speed and the static real-time movement direction of the static targets when they start to move are determined according to the real-time environment image.

6. The system according to claim 5, characterized in that The vehicle information includes a vehicle motion track, a real-time weight and a vehicle acceleration of the vehicle. When the steering information determination module determines the steering assistance information according to the motion information, the road slipperiness, the vehicle information and the real-time environment image, it is used to: Determining whether the normal driving state of the vehicle is affected according to the motion trajectory and the vehicle motion trajectory; If affected, determining the dynamic relative distance between the vehicle and the dynamic risk target, the dynamic acceleration of the dynamic risk target, and the dynamic relative acceleration between the dynamic risk target and the vehicle according to the real-time environment image; Obtaining the impact moment of the dynamic risk target affecting the normal driving state and the turning moment when the driver starts to turn, and determining the driver's reaction time according to the impact moment and the turning moment; The steering avoidance angle of the vehicle is determined according to the driving speed, the dynamic relative acceleration, the vehicle acceleration, the real-time weight, the reaction time, the dynamic relative distance, the dynamic acceleration, the road slipperiness and the dynamic target volume, and is calculated according to the following formula: ； in, represents the steering avoidance angle, represents the reaction time, represents the dynamic acceleration, Indicates the slipperiness of the road. represents the dynamic target volume, represents the dynamic relative distance, represents the driving speed, represents the vehicle acceleration, represents the reaction time, represents the real-time weight, represents the dynamic relative acceleration; The steering avoidance angle is incorporated into the steering assist information.

7. The system according to claim 6, characterized in that The real-time environment image includes a road edge image, the vehicle information includes a vehicle width, and the steering information determination module determines the steering assistance information according to the motion information, the road wetness, the vehicle information and the real-time environment image, for: Determine, based on the road edge image, the curvature of the road edge when the vehicle is turning, the real-time position of the vehicle on the road, and the real-time interval distance between the vehicle and the road edge; Determining a real-time turning radius of the vehicle when turning according to the road edge curvature, the real-time position, the driving speed and the real-time interval distance; determining a turning speed of the vehicle when turning according to the driving speed, the dynamic motion speed and the dynamic target volume; determining a reduction in the turning speed of the vehicle when turning according to the turning speed and the driving speed; The turning time of the vehicle is determined according to the vehicle acceleration, the reduction in turning speed, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road wetness and the driving speed, and is calculated according to the following formula: ； in, Indicates the duration of the turn, represents the width of the vehicle, represents the dynamic target volume, Indicates the preset volume avoidance coefficient, represents the driving speed, represents the reaction time, represents the dynamic motion speed, represents the real-time turning radius, Indicates the slipperiness of the road. represents the reduction amount of the steering speed; The steering duration and the steering speed are incorporated into the steering assistance information.

8. The system according to claim 7, characterized in that The steering information determination module is used to determine the steering duration of the vehicle according to the vehicle acceleration, the steering speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed: Judging whether the road edge boundary is clear according to the real-time environment image; If the road edge boundary is unclear, determining the volume and position of the object in the shoulder area according to the real-time environment image; Determining the actual width of the road according to the volume of the object and the position of the object; The turning time is determined according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, and is calculated according to the following formula: ； in, Indicates the duration of the turn, represents the actual width of the road, represents the width of the vehicle, represents the dynamic target volume, represents the driving speed, represents the reaction time, represents the dynamic motion speed, represents the real-time turning radius, Indicates the slipperiness of the road. represents the reduction in steering speed, represents the volume of the object, represents the vehicle acceleration.

9. The system according to claim 8, characterized in that After determining the turning time according to the vehicle acceleration, the actual road width, the object volume, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the real-time turning radius, the road slipperiness and the driving speed, the turning information determination module is further used to: The first turning time is determined according to the actual road width, the vehicle acceleration, the turning speed reduction, the dynamic motion speed, the dynamic target volume, the vehicle width, the road slipperiness and the driving speed, and is calculated according to the following formula: ； in, represents the first turn duration, represents the actual width of the road, represents the dynamic target volume, represents the preset volume safety factor, represents the driving speed, represents the dynamic motion speed, represents the reaction time, represents the vehicle acceleration, Indicates the slipperiness of the road. Indicates the amount of reduction in steering speed.

10. The system according to claim 9, characterized in that When the steering information determination module determines the steering assistance information according to the motion information, the road wetness, the vehicle information and the real-time environment image, it is used to: determining, according to the real-time environment image, a first slippery degree of the road after the vehicle turns; Determining a first driving speed of the vehicle after turning according to the vehicle information; determining a real-time offset degree of the center of gravity of the vehicle according to the real-time weight, the first driving speed and the first slippery degree; Acquire a center of gravity offset safety range, match the real-time offset degree with the center of gravity offset safety range, and determine whether the real-time offset degree is within the center of gravity offset safety range; If the real-time offset degree is not within the center-of-gravity offset safety range, determining an offset degree difference according to the real-time offset degree and the center-of-gravity offset safety range; determining a maximum turning speed of the vehicle according to the deviation degree difference, the first driving speed and the real-time weight; A secondary steering angle of the vehicle after turning is determined according to the maximum steering speed, the first slippery degree and the real-time weight, and the secondary steering angle is included in the steering assist information.

Images