A vehicle side blind spot collision warning method

Abstract

The present application designs a vehicle side blind spot collision avoidance warning method, comprising: step S1, obtaining the vehicle's real-time driving information; step S2, when it is determined that the vehicle is turning, calculating the inner wheel difference area of ​​the vehicle in the current state in real time based on the real-time driving information; step S3, determining the blind spot detection area ROI based on the inner wheel difference area and the detection range of the obstacle detection sensor; step S4, performing obstacle detection based on the determined blind spot detection area ROI; step S5, when it is determined that there is an obstacle in the blind spot detection area ROI, performing a side blind spot collision avoidance warning in the turning direction. The present application determines the first intersection of the inner wheel difference area with the horizontal vertical line of the vehicle's driving speed direction or the long line along the horizontal direction of the vehicle's front according to the vehicle's real-time driving information; and based on the line between the first intersection and the obstacle detection sensor, it can determine the blind spot detection area ROI within the detection range in real time, thereby achieving the technical effect of safety monitoring.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving, and specifically to a vehicle side blind spot collision warning method. Background Art

[0002] When the driver is turning, due to the structural design of the vehicle body, there will be different degrees of visual blind spots in the driver's field of view. Especially for large trucks and large buses, the influence of the vehicle's inner wheel difference must also be considered. When the driver observes the surrounding information of the vehicle through the rearview mirror, the visual blind spot is further expanded, seriously affecting driving safety.

[0003] In addition, for large semi-trailers, since the tractor and trailer are connected by a non-rigid body, the inner wheel difference area is large when turning, and the trailer appears in the lateral blind spot of the tractor. There is a possibility that the tractor will be identified as an obstacle (such as the vehicle in front), causing a false alarm of the automatic driving system.

[0004] Chinese patent CN207697623U discloses a driving side blind spot warning system, including a radar detection module, a video signal acquisition module, a central control module, and a driving warning module. This system primarily provides warnings for side blind spots, improving driving safety. The system uses ultrasonic radar detection and warning combined with infrared camera monitoring to inspect blind spots. While radar provides distance measurement and warning, the infrared camera visually displays blind spot road conditions on the onboard display, allowing the driver to observe the blind spot conditions in real time, reducing the possibility of false alarms and missed alerts from multiple angles. LED lights and a buzzer provide the driver with a visual and auditory warning system, significantly reducing accidents caused by blind spot issues while driving.

[0005] Chinese patent CN107672592B discloses a lane keeping assist system based on active lane changing or steering of a vehicle, relating to the field of active vehicle safety technology. It is primarily used to keep the vehicle from changing lanes or turning when there is a risk of collision with the vehicle's active lane changing or steering. It includes a lane keeping assist unit, an active hazard warning system, and a controller. The controller is used to control the on and off of the lane keeping assist system. It is configured so that when the lane keeping assist system is on and an active lane change or steering signal is initiated, the active hazard warning system detects that there is a risk of collision with the vehicle when the vehicle changes lanes or turns to one side, and keeps the lane keeping assist unit active.

[0006] Although the above patents are all aimed at detecting visual blind spots, they all increase side sensors and directly designate part of the detection area as a blind spot detection area. This method not only has high hardware costs, but also has the problem of false alarms for non-obstacle targets in a certain area behind the vehicle, resulting in reduced reliability of the autonomous driving system. Summary of the Invention

[0007] In order to solve the problems of large errors and false alarms in visual blind spot detection and to achieve the technical effect of accurately and real-time updating of blind spot monitoring, this application designs a vehicle side blind spot collision prevention warning method.

[0008] A vehicle side blind spot collision avoidance warning method comprises the following steps: step S1, acquiring real-time vehicle driving information; step S2, when it is determined that the vehicle is turning, calculating in real time an inner wheel difference area in the vehicle's current state based on the real-time driving information; step S3, determining a blind spot detection area (ROI) based on the inner wheel difference area and a detection range of an obstacle detection sensor; step S4, performing obstacle detection based on the determined blind spot detection area (ROI); and step S5, when it is determined that an obstacle exists within the blind spot detection area (ROI), performing a side blind spot collision avoidance warning in the vehicle's turning direction.

[0009] Preferably, in step S3, the method for determining the blind spot detection area ROI includes: step 301, determining the first intersection of a first longitudinal line and an outer edge of an inner wheel difference area in a vehicle turning state, wherein the first longitudinal line is a horizontal longitudinal line perpendicular to the vehicle's driving speed direction or a horizontal longitudinal line in front of the vehicle head; step 302, determining the blind spot detection area ROI within the detection range of the obstacle detection sensor based on the first intersection, the vehicle head position point, and the center coordinate point of the field of view coordinate system of the obstacle detection sensor.

[0010] Preferably, the vehicle front position point is a front edge point of the vehicle front on the side on which the vehicle is turning.

[0011] Preferably, the vehicle is a large vehicle, and the method for determining the blind spot detection area ROI in step S3 also includes: determining the second long line and the revised front position point in the vehicle turning state according to the vehicle length revision value L3; determining the second intersection of the second long line and the outer edge of the inner wheel difference area in the vehicle turning state; and determining the blind spot detection area ROI within the detection range of the obstacle detection sensor according to the second intersection, the revised front position point, and the center coordinate point of the field of view coordinate system of the obstacle detection sensor.

[0012] Preferably, the revised vehicle front position point is a position point in front of the vehicle at a distance from the vehicle front position point by the revised vehicle length value L3, and the second along-line is a along-line passing through the revised vehicle front position point and parallel to the first along-line.

[0013] Preferably, the vehicle is a semi-trailer, which includes at least a tractor and a trailer. In step S3, the method for determining the blind spot detection area ROI includes: step 311, recording the center coordinate point of the field of view coordinate system of the obstacle detection sensor at the rear of the trailer as point P; step 312, calculating the intersection point between the tractor and the trailer through trigonometric relationship based on the tractor length, the trailer length, and the angle β between the tractor and the trailer, and recording it as point N; step 313, calculating point Z through trigonometric relationship based on the tractor length, the coordinates of point N, and the vehicle length revision value L3; step 314, determining a third longitudinal line at point Z, and recording the intersection point of the third longitudinal line and the outer edge of the inner wheel difference area as point W, wherein the third longitudinal line is a longitudinal line passing through point Z and parallel to the first longitudinal line; step 315, determining the blind spot detection area ROI based on point P, point N, point Z, point W, and the detection range of the obstacle detection sensor.

[0014] Preferably, step S315 further includes: determining the curve WP by curve fitting according to point P, point W and the outer edge of the inner wheel difference area corresponding to the current driving state; and determining the blind spot detection area ROI according to the curve WP, point N, point Z and the detection range of the obstacle detection sensor.

[0015] Preferably, in step S1, the real-time driving information of the vehicle is obtained through an IMU unit, or through a gyroscope and a speed sensor.

[0016] Preferably, the obstacle detection sensor is installed at the rear left and / or rear right of the rear of the vehicle.

[0017] Preferably, in step S5, the side blind spot collision avoidance warning is mainly implemented through a warning display, and the warning display is connected to the obstacle detection sensor through a data processing device.

[0018] The advantages and effects of this application are as follows:

[0019] 1. This application designs a vehicle side blind spot collision warning method, which establishes a world coordinate system through obstacle detection sensors to form an active safety system, which can achieve the technical effect of monitoring blind spots.

[0020] 2. After determining the first intersection of the first long line and the outer edge of the inner wheel difference area when the vehicle is turning, this application determines the blind spot detection area ROI within the detection range of the obstacle detection sensor based on the first intersection, the vehicle's front position point, and the center coordinate point of the field of view coordinate system of the obstacle detection sensor, and can update the ROI area in real time to achieve the technical effect of safety monitoring.

[0021] 3. This application increases the vehicle length revision value L3 and redefines the second intersection with the inner wheel difference area 2. Based on the blind spot caused by the inner wheel difference, it increases the side blind spot detection area ROI during vehicle turning, thereby solving the blind spot problem caused by the cab height of large vehicles and the vehicle A-pillar, and can effectively improve the reliability of side blind spot collision warning.

[0022] 4. This application determines the blind spot detection area ROI by calculating the curve WP, point N, point Z and the detection range of the obstacle detection sensor 8, thereby solving the problem of identifying the head of the tractor 6 within the detection range as an obstacle during the turning process of a large semi-trailer.

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiment of the present application in conjunction with the accompanying drawings.

[0024] Based on the detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. 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, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0026] Figure 1 A flowchart of a vehicle side blind spot collision warning method provided by this application;

[0027] Figure 2 This is a schematic diagram of the existing large vehicle blind spot detection area ROI;

[0028] Figure 3 A schematic diagram of the vehicle blind spot detection area ROI provided by this application;

[0029] Figure 4 A schematic diagram of a blind spot detection area ROI for a semi-trailer provided in this application;

[0030] Figure 5 A schematic diagram of another blind spot detection area ROI for a semi-trailer provided in this application;

[0031] Figure 6 A schematic diagram of the blind spot detection area ROI for large vehicles provided in this application;

[0032] Figure 7 Simulation test diagram provided for this application;

[0033] Figure numerals: 1. Vehicle; 2. Inner wheel difference area; 3. Warning designated area; 4. Target; 5. Blind spot detection area ROI; 6. Tractor; 7. Trailer; 8. Obstacle detection sensor; 9. Sensor detection range. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions 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, rather than all of the embodiments. In the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the present application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. In addition, for clarity and brevity, the description of known functions and structures has been omitted in the embodiments.

[0035] It should be understood that references throughout this specification to "one embodiment" or "this embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present application. Therefore, the appearance of "one embodiment" or "this embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0036] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0037] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" in this article describes another type of association object relationship, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the previous and subsequent associated objects are in an "or" relationship.

[0038] The term "at least one" in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, at least one of A and B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0039] It should also be noted that, in this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprises," or any other variations thereof are intended to cover non-exclusive inclusion.

[0040] Example 1

[0041] This embodiment mainly shows a basic design of a vehicle side blind spot collision warning method, which is applicable to all large vehicles.

[0042] In this embodiment, a vehicle side blind spot collision warning system is established. The system includes an obstacle detection sensor 8 such as an RGBD camera, an infrared camera, a radar, a monocular or multi-camera, etc., a host computer, a warning display, and other related equipment, such as an inertial measurement unit. The warning display can display the blind spot image detected by the obstacle detection sensor 8 and can emit a sound. It is installed on the A-pillar of the cockpit to facilitate the driver to view and clearly hear the warning sound. The obstacle detection sensor 8 is installed on the left rear and / or right rear of the rear of the vehicle 1; the obstacle detection sensor 8 is calibrated with external parameters, and vehicle 1 information such as track width, wheelbase, and vehicle length is input into the host computer; the system is kept running in real time, and the blind spot monitoring is updated in real time based on vehicle body information and driving information, such as IMU information, to monitor the blind spot in real time.

[0043] For large vehicles, if vehicle 1 is traveling at speed V and there is an inner wheel difference area 2 when turning, if only the warning designated area 3 is used for blind spot warning, there is a possibility that target 4 that does not affect the normal turning of vehicle 1 will be identified as an obstacle, triggering a collision alarm. For detailed detection, please refer to Figure 2 In response to this situation, this embodiment shows a vehicle side blind spot collision warning method. Please refer to the method flow chart. Figure 1 , including the following steps:

[0044] Step S1: Acquire real-time driving information of vehicle 1. This real-time driving information can be obtained by an IMU unit installed on the vehicle, or by other sensors on the vehicle, such as a gyroscope and a speed sensor. This real-time driving information includes at least speed information, acceleration information, and steering information.

[0045] Step S2: when it is determined that the vehicle 1 is turning, the inner wheel difference area 2 of the vehicle 1 in the current state is calculated in real time based on the real-time driving information;

[0046] Step S3: determining a blind spot detection area ROI 5 based on the inner wheel difference area 2 and the detection range 9 of the obstacle detection sensor 8;

[0047] Step S4: performing obstacle detection according to the determined blind spot detection area ROI 5;

[0048] Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI 5, a side blind spot collision avoidance warning in the vehicle turning direction is performed.

[0049] Furthermore, the method for determining the blind spot detection area ROI 5 in step 3 includes:

[0050] Step 301: Determine a first intersection point B between a first longitudinal line and an outer edge of the inner wheel difference area 2 when the vehicle 1 is in a turning state, wherein the first longitudinal line is a horizontal longitudinal line perpendicular to the direction of the vehicle 1's travel speed or a horizontal longitudinal line in front of the vehicle's front end;

[0051] When the vehicle turns right, the calculated inner wheel difference area 2 is located on the right side of the vehicle 1. At this time, the first intersection point B is the intersection of the right horizontal long line and the right outer edge of the right inner wheel difference area 2 in the right turn state;

[0052] When the vehicle turns left, the calculated inner wheel difference area 2 is located on the left side of the vehicle 1. At this time, the first intersection point B is the intersection of the left horizontal long line and the left outer edge of the left inner wheel difference area 2 in the left turn state.

[0053] Step 302: Determine a blind spot detection area (ROI) 5 within the detection range of the obstacle detection sensor 8 based on the first intersection point B, the front position point A of the vehicle 1, and the center coordinate point C of the field of view coordinate system of the obstacle detection sensor 8. The front position point A is the front edge of the vehicle on the side toward which the vehicle is turning.

[0054] Furthermore, the real-time driving information of vehicle 1 includes information such as vehicle speed, acceleration, and steering angle. This information can be obtained through an IMU unit or collected by installing corresponding sensors on vehicle 1.

[0055] Furthermore, the obstacle detection sensor 8 is installed at the rear left and / or rear right of the rear of the vehicle 1 .

[0056] Furthermore, in step S5, the side blind spot collision warning is mainly implemented through a warning display, and the warning display is connected to the obstacle detection sensor 8 through a data processing device.

[0057] It should be noted that the calculation of the inner wheel difference of the vehicle 1 in this embodiment is a prior art and the calculation process will not be discussed in detail.

[0058] It should be noted that the origin of the coordinate system in this embodiment is the vertical point of the obstacle detection sensor 8 on the ground. The X-axis direction is perpendicular to the vehicle body, the Z-axis direction is parallel to the vehicle body and forward, the Y-axis direction is perpendicular to the ground and downward, and the negative Y-axis direction is upward. The origin of the coordinate system is also the center coordinate point C of the obstacle detection sensor 8's field of view coordinate system.

[0059] After calibration, the obstacle detection sensor 8 constructs a three-dimensional real-world space and maps the calculated inner wheel difference blind spot into this three-dimensional space. When an obstacle is detected within the inner wheel difference area 2, the system triggers an alert. The warning display sounds an alarm to alert the driver and displays a real-time image for the driver to view.

[0060] This embodiment shows a vehicle side blind spot collision warning method, which establishes a world coordinate system through the obstacle detection sensor 8 to form an active safety system, which can achieve the technical effect of monitoring the blind spot. For the effect, please refer to Figure 3 , where A represents the vehicle head position point, B represents the first intersection point, and C represents the center coordinate point.

[0061] This application determines the first intersection B of the horizontal vertical line in the direction of the vehicle's travel speed, or the long line along the horizontal direction of the front of the vehicle, and the inner wheel difference area 2 based on the vehicle length and the current steering information; and determines the blind spot detection area ROI 5 within the detection range 9 of the obstacle detection sensor based on the connection line between the first intersection B and the obstacle detection sensor 8, and can update the ROI area in real time to achieve the technical effect of safety monitoring.

[0062] Specific effects such as Figure 7 As shown, the warning display shows the blind spot image in real time, which can ensure real-time monitoring of the position of people in the blind spot. When the person does not enter the blind spot detection area ROI 5, the warning system does not make a sound. When the person enters the blind spot detection area ROI5, the system makes a sound to remind the driver to avoid it.

[0063] Example 2

[0064] Based on the above-mentioned embodiment 1, this embodiment mainly introduces a vehicle side blind spot collision avoidance warning method for the A-pillar. For large vehicles, such as trucks and buses, due to the influence of the vehicle cab height and the vehicle A-pillar, the side blind spot during the vehicle turning process is increased on top of the blind spot caused by the inner wheel difference. Therefore, this embodiment shows that the vehicle side blind spot collision avoidance warning method for the A-pillar includes the following steps:

[0065] Step S1: Acquire real-time driving information of vehicle 1. This real-time driving information can be obtained by an IMU unit installed on the vehicle, or by other sensors on the vehicle, such as a gyroscope and a speed sensor. This real-time driving information includes at least speed information, acceleration information, and steering information.

[0066] Step S2: when it is determined that the vehicle 1 is turning, the inner wheel difference area 2 of the vehicle 1 in the current state is calculated in real time based on the real-time driving information;

[0067] Step S3: determining a blind spot detection area ROI 5 based on the inner wheel difference area 2 and the detection range 9 of the obstacle detection sensor 8;

[0068] Step S4: performing obstacle detection according to the determined blind spot detection area ROI 5;

[0069] Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI 5, a side blind spot collision avoidance warning in the vehicle turning direction is performed.

[0070] Furthermore, the real-time driving information of vehicle 1 includes information such as vehicle speed, acceleration, and steering angle. This information can be obtained through an IMU unit or collected by installing corresponding sensors on vehicle 1.

[0071] Furthermore, the obstacle detection sensor 8 is installed at the rear left and / or rear right of the rear of the vehicle 1 .

[0072] Furthermore, in step S5, the side blind spot collision warning is mainly implemented through a warning display, and the warning display is connected to the obstacle detection sensor 8 through a data processing device.

[0073] Furthermore, the method for determining the blind spot detection area ROI 5 in step S3 includes:

[0074] Step 301: Determine a first intersection point Q between a first longitudinal line and an outer edge of the inner wheel difference region 2 when the vehicle 1 is in a turning state, wherein the first longitudinal line is a horizontal longitudinal line perpendicular to the direction of the vehicle 1's travel speed or a horizontal longitudinal line in front of the vehicle's front end;

[0075] When the vehicle turns right, the calculated inner wheel difference area 2 is located on the right side of the vehicle 1. At this time, the first intersection point Q is the intersection of the right horizontal long line and the right outer edge of the right inner wheel difference area 2 in the right turn state;

[0076] When the vehicle turns left, the calculated inner wheel difference area 2 is located on the left side of the vehicle 1. At this time, the first intersection point Q is the intersection of the left horizontal long line and the left outer edge of the left inner wheel difference area 2 in the left turn state.

[0077] Step 302 : Determine a blind spot detection area ROI 5 within the detection range of the obstacle detection sensor 8 based on the first intersection point Q, the front position point E of the vehicle 1 , and the center coordinate point H of the field of view coordinate system of the obstacle detection sensor 8 .

[0078] Furthermore, in step S3, if the vehicle 1 is a large vehicle, the vehicle length revision value L3 needs to be increased. For details, please refer to Figure 6 , where point E is the vehicle head position, point Q is the first intersection point, point M is the endpoint of the L3 extension line, and point H is the center coordinate point; the vehicle length revision value L3 can be manually set based on information such as vehicle body length, vehicle type, and steering wheel angle to re-determine the second intersection point with the inner wheel difference area 2. At this time, after increasing the vehicle length revision value L3, a new blind spot detection area ROI 5 can be obtained more accurately, the range of the blind spot detection area ROI 5 can be increased, and the accuracy of blind spot collision detection can be improved.

[0079] In the above step 3, the method for determining the blind spot detection area ROI 5 further includes:

[0080] According to the vehicle length revision value L3, the second along-line and the revised front position point of the vehicle 1 in the turning state are determined, wherein the revised front position point is the position point in front of the vehicle 1 at a distance from the front position point E by the vehicle length revision value L3, and the second along-line is the along-line passing through the revised front position point and parallel to the first along-line.

[0081] Taking the vehicle turning right as an example, the revised front position point is the position point on the right side of the vehicle at a distance from the front end of the front position point E by the revised vehicle length value L3, and this point is on the second long line.

[0082] Determine a second intersection point of the second longitudinal line and the outer edge of the inner wheel difference area 2 when the vehicle 1 is in a turning state;

[0083] According to the second intersection point, the revised vehicle head position point, and the center coordinate point H of the field of view coordinate system of the obstacle detection sensor 8 , a blind spot detection area ROI 5 is determined within the detection range of the obstacle detection sensor 8 .

[0084] In step 3 above, by increasing the vehicle length revision value L3 and re-determining the second intersection point with the inner wheel difference area 2, the range of the blind spot detection area ROI 5 is increased, which helps to improve the reliability of the side blind spot collision avoidance warning.

[0085] Example 3

[0086] Based on the above embodiments 1-2, this embodiment mainly introduces a vehicle side blind spot collision warning method for a semi-trailer. For a semi-trailer, it includes at least a tractor 6 and a trailer 7. After the obstacle detection sensor 8 is installed at the rear of the trailer, when the large semi-trailer is turning, such as Figure 4 As shown, part of the head area of ​​the tractor 6 (such as the shaded area) will appear within the sensor detection range 9, and there is a possibility that the head of the tractor 6 within the detection range will be identified as an obstacle (such as a vehicle in front).

[0087] Therefore, this embodiment designs a vehicle side blind spot collision warning method for a semi-trailer, including the following steps:

[0088] Step S1: Acquire real-time driving information of vehicle 1. This real-time driving information can be obtained by an IMU unit installed on the vehicle, or by other sensors on the vehicle, such as a gyroscope and a speed sensor. This real-time driving information includes at least speed information, acceleration information, and steering information.

[0089] Step S2: When it is determined that the vehicle 1 (tractor 6) is turning, the inner wheel difference area 2 of the vehicle 1 in the current state is calculated in real time based on the real-time driving information;

[0090] Step S3: determining a blind spot detection area ROI 5 based on the inner wheel difference area 2 and the detection range 9 of the obstacle detection sensor 8;

[0091] Step S4: performing obstacle detection according to the determined blind spot detection area ROI 5;

[0092] Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI 5, a side blind spot collision avoidance warning in the vehicle turning direction is performed.

[0093] Furthermore, the real-time driving information of vehicle 1 includes information such as vehicle speed, acceleration, and steering angle. This information can be obtained through an IMU unit or collected by installing corresponding sensors on vehicle 1.

[0094] Furthermore, the obstacle detection sensor 8 is installed at the rear left and / or rear right of the rear of the vehicle 1 .

[0095] Furthermore, in step S5, the side blind spot collision warning is mainly implemented through a warning display, and the warning display is connected to the obstacle detection sensor 8 through a data processing device.

[0096] Furthermore, step S3 further includes the following steps:

[0097] Step 311: The center coordinate point of the field of view coordinate system of the obstacle detection sensor 8 at the rear of the trailer 7 is recorded as point P;

[0098] Step 312: Calculate the intersection point between the tractor 6 and the trailer 7 based on the tractor length, the trailer length, and the angle β between the tractor 6 and the trailer 7 through trigonometric relationships, and record it as point N.

[0099] Taking a semitrailer turning right as an example, since the tractor 6 and trailer 7 are connected by a non-rigid body, the right side of the tractor 6 will overlap with the right side of the trailer 7. The angle between the tractor 6 and trailer 7 is set to β. To ensure the accuracy of the blind spot detection area ROI 5 during right turns and avoid false detections, the middle connection between the tractor 6 and trailer 7 is used as the rotation midpoint. Based on the semitrailer vehicle parameters and the angle β, the intersection point between the right edge of the tractor 6 and the right edge of the trailer 7 can be calculated, recorded as point N.

[0100] Step 313: Calculate point Z based on the tractor length, the coordinates of point N, and the revised vehicle length value L3 through triangular geometry.

[0101] Specifically, for a semitrailer turning right, a vehicle length revision value L3 is set at the right front end of tractor 6. An extension line is drawn from the right edge of the front of tractor 6, starting at zero at the front end of tractor 6. When the extension length is L3, the endpoint of the extension line is designated as point Z. Given the known lengths PN and NZ, as well as the angle β between PN and NZ, the coordinates of point Z (Xz, Zz) can be calculated using trigonometric relationships.

[0102] Step 314: Determine a third long line at point Z, and record the intersection of the third long line and the outer edge of the inner wheel difference area 2 as point W, wherein the third long line is a long line passing through point Z and parallel to the first long line.

[0103] Step 315 : Determine the blind spot detection area ROI 5 according to point P, point N, point Z, point W, and the detection range 9 of the obstacle detection sensor 8 .

[0104] The obstacle detection sensor 8 is installed at the left rear of the vehicle, N is the junction of the tractor 6 and the trailer 7, and L3 is a manual setting used to include parameters of the A-pillar blind spot.

[0105] By setting the known inner wheel difference m' and the angle β between the tractor 6 and the trailer 7, and combining the vehicle body information, the steering wheel angle, and the calculated area PNZW (blind spot detection area ROI 5), active filtering of the tractor 6 body part is achieved.

[0106] When calculating the area PNZW, based on the above known parameters, geometric operations can be used to first calculate the PN length Zn and the NZ length, and then the vertical distance Xz from point Z to the Z axis, the vertical distance Zz from point Z to the X axis, and the vertical distance Xw from point W to the Z axis, the vertical distance Zw from point W to the X axis are calculated. Therefore, the coordinates of point P are (0, 0), the coordinates of point N are (0, Zn), the coordinates of point Z are (Xz, Zz), and the coordinates of point W are (Xw, Zw).

[0107] It should be noted that in step S312, the determination of the N points also includes identifying the continuous changes of the head portion of the tractor in the detected image, which is obtained by parallax calculation.

[0108] Furthermore, in order to further refine the ROI area of ​​the semi-trailer's side blind spot, after determining the coordinates of point W,

[0109] The curve WP is determined based on point P, point W, and the outer edge of the inner wheel difference region 2 corresponding to the current driving state. The curve WP is calculated in the same manner as the inner wheel difference, and will not be described in detail.

[0110] The blind spot detection area ROI5 is determined according to the curve WP, point N, point Z and the detection range of the obstacle detection sensor 8 .

[0111] Furthermore, the obstacle detection sensor 8 adopts a binocular camera, or other equivalent sensors. This embodiment adopts a binocular camera as an example.

[0112] Please refer to the effect Figure 5 , where point Z is the endpoint of the extended line L3, W represents the intersection of the third longitudinal line and the outer edge of the inner wheel difference area 2, point N represents the intersection between the right edge of the tractor 6 and the right edge of the trailer 7, and P represents the center coordinate point. This embodiment determines the curve WP by using a curve fitting method based on the inner wheel difference corresponding to the current driving state, thereby avoiding blind spot detection in areas outside the curve WP, which would waste resources and cause false detection outside the blind spot.

[0113] The above are merely preferred embodiments of the present application and are not intended to limit the scope of protection of the present application. Persons skilled in the art will readily appreciate that the present application may be modified and altered in various ways. Any changes, modifications, substitutions, integrations, or parameter changes to these embodiments, which fall within the spirit and principles of the present application and achieve the same functionality through conventional substitutions, without departing from the principles and spirit of the present application, fall within the scope of protection of the present application.

Claims

1. A vehicle side blind spot collision warning method, characterized in that: The following steps are involved: Step S1, obtaining real-time driving information of vehicle (1); Step S2: when it is determined that the vehicle (1) is turning, calculating the inner wheel difference area (2) of the vehicle (1) in the current state in real time based on the real-time driving information; Step S3, determining a blind spot detection area ROI (5) according to the inner wheel difference area (2) and the detection range of the obstacle detection sensor (8); Step S4, performing obstacle detection according to the determined blind spot detection area ROI (5); Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI (5), a side blind spot collision avoidance warning in the vehicle steering direction is performed; In step S3, the method for determining the blind spot detection area ROI (5) includes: Step 301: determining a first intersection point between a first longitudinal line and an outer edge of the inner wheel difference area (2) when the vehicle (1) is in a turning state, wherein the first longitudinal line is a horizontal longitudinal line perpendicular to the direction of the vehicle (1) traveling speed or a horizontal longitudinal line in front of the vehicle head; Step 302: Determine the blind spot detection area ROI (5) within the detection range of the obstacle detection sensor (8) based on the first intersection point, the front position point of the vehicle (1), and the center coordinate point of the field of view coordinate system of the obstacle detection sensor (8).

2. The vehicle side blind spot collision warning method according to claim 1, characterized in that: The vehicle head position point is the front edge point of the vehicle head on the side to which the vehicle (1) turns.

3. A vehicle side blind spot collision warning method, characterized in that: The following steps are involved: Step S1, obtaining real-time driving information of vehicle (1); Step S2: when it is determined that the vehicle (1) is turning, calculating the inner wheel difference area (2) of the vehicle (1) in the current state in real time based on the real-time driving information; Step S3, determining a blind spot detection area ROI (5) according to the inner wheel difference area (2) and the detection range of the obstacle detection sensor (8); Step S4, performing obstacle detection according to the determined blind spot detection area ROI (5); Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI (5), a side blind spot collision avoidance warning in the vehicle steering direction is performed; The vehicle (1) is a large vehicle, and the method for determining the blind spot detection area ROI (5) in step S3 includes: According to the revised vehicle length value L3, a second longitudinal line and a revised vehicle head position point are determined when the vehicle (1) is in a turning state; wherein the second longitudinal line is a longitudinal line passing through the revised vehicle head position point and parallel to the first longitudinal line, and the first longitudinal line is a horizontal longitudinal line perpendicular to the direction of the vehicle (1) traveling speed or a horizontal longitudinal line in front of the vehicle head; Determining a second intersection point between the second longitudinal line and the outer edge of the inner wheel difference area (2) when the vehicle (1) is in a turning state; The blind spot detection area ROI (5) is determined within the detection range of the obstacle detection sensor (8) based on the second intersection point, the revised vehicle head position point, and the center coordinate point of the field of view coordinate system of the obstacle detection sensor (8).

4. The vehicle side blind spot collision warning method according to claim 3, characterized in that: The revised vehicle head position point is a position point in front of the vehicle (1) at a distance from the vehicle head position point by the vehicle length revision value L3, and the vehicle head position point is a front edge point of the vehicle head on the turning side of the vehicle (1).

5. A vehicle side blind spot collision warning method, characterized in that: The following steps are involved: Step S1, obtaining real-time driving information of vehicle (1); Step S2: when it is determined that the vehicle (1) is turning, calculating the inner wheel difference area (2) of the vehicle (1) in the current state in real time based on the real-time driving information; Step S3, determining a blind spot detection area ROI (5) according to the inner wheel difference area (2) and the detection range of the obstacle detection sensor (8); Step S4, performing obstacle detection according to the determined blind spot detection area ROI (5); Step S5: When it is determined that there is an obstacle in the blind spot detection area ROI (5), a side blind spot collision avoidance warning in the vehicle steering direction is performed; The vehicle (1) is a semitrailer, which comprises at least a tractor (6) and a trailer (7). In step S3, the method for determining the blind spot detection area ROI (5) includes: Step 311: The center coordinate point of the field of view coordinate system of the obstacle detection sensor (8) at the rear of the trailer (7) is recorded as point P; Step 312: Calculate the intersection point between the tractor (6) and the trailer (7) based on the length of the tractor, the length of the trailer, and the angle β between the tractor (6) and the trailer (7) through trigonometric relationships, and record it as point N. Step 313: Calculate point Z based on the tractor length, the coordinates of point N, and the revised vehicle length value L3 through triangular geometry. Step 314: determining a third longitudinal line at the Z point, and recording the intersection of the third longitudinal line and the outer edge of the inner wheel difference area (2) as point W, wherein the third longitudinal line is a longitudinal line passing through the Z point and parallel to the first longitudinal line; the first longitudinal line is a horizontal longitudinal line perpendicular to the direction of the vehicle (1) traveling speed or a horizontal longitudinal line in front of the vehicle head; Step 315: Determine the blind spot detection area ROI (5) based on the point P, the point N, the point Z, the point W, and the detection range of the obstacle detection sensor (8).

6. The vehicle side blind spot collision warning method according to claim 5, characterized in that: The step S315 further includes: Determine the curve WP by using a curve fitting method based on the point P, the point W, and the outer edge of the inner wheel difference area (2) corresponding to the current driving state; The blind spot detection area ROI (5) is determined based on the curve WP, the point N, the point Z and the detection range of the obstacle detection sensor (8).

7. A vehicle side blind spot collision warning method according to claim 1, 3 or 5, characterized in that: In step S1, the real-time driving information of the vehicle (1) is obtained through an IMU unit, or through a gyroscope and a speed sensor.

8. A vehicle side blind spot collision warning method according to claim 1, 3 or 5, characterized in that: The obstacle detection sensor (8) is installed at the left rear and / or right rear of the rear of the vehicle (1).

9. A vehicle side blind spot collision warning method according to claim 1, 3 or 5, characterized in that: In step S5, the side blind spot collision warning is implemented through a warning display, and the warning display is connected to the obstacle detection sensor (8) through a data processing device.

Images