A longitudinal deceleration identification generation method and system based on perception data

Abstract

The application relates to a longitudinal deceleration mark generation method and system based on perception data, lane line data of a deceleration section is obtained; a plurality of anchor points are arranged on the lane line data of the deceleration section, the arrangement direction of the anchor points is the same as the advancing direction of the lane line data; the coordinates of a plurality of contour points of deceleration marks are obtained based on the coordinate data of the anchor points, and then the deceleration marks are drawn, the deceleration marks correspond to the anchor points one by one; and the drawn deceleration marks are displayed on a display screen of a vehicle-mounted entertainment system. The application provides an input source of perception data (original data), uses basic theories of graphics to assemble the original data into description data, and finally restores the longitudinal deceleration marks in the vehicle-mounted entertainment system, so that the driving conditions of the vehicle are fed back in real time, and accurate and real-time road surface information is provided for drivers and passengers.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, specifically to high-precision map / 3D scene reconstruction technology based on a cockpit platform. Background Technology

[0002] With the continuous diversification of in-vehicle entertainment systems, digital reconstruction of the vehicle's surrounding environment has gradually become a necessary means of providing visualization services. Longitudinal deceleration markers are white markings placed at toll booths, exit ramps, or other road sections requiring vehicle deceleration. They are typically a group of diamond-shaped dashed blocks parallel to both sides of the lane lines, creating a sense of pressure by visually narrowing the lane, prompting drivers to automatically slow down at curves, slopes, and other locations where deceleration is required.

[0003] Longitudinal deceleration markings extend along the lane lines, so the perceived lane line data will carry information about the longitudinal deceleration markings. Lane line data does not carry the specific physical data of the longitudinal deceleration markings; it only identifies the longitudinal deceleration markings for that lane in the lane line type information. This method is not obvious enough, resulting in the inability to provide drivers, passengers, and other users with accurate and real-time road information. Summary of the Invention

[0004] One of the objectives of this invention is to provide a method for generating longitudinal deceleration markers based on perception data, in order to solve the problem that existing technologies cannot provide accurate and real-time road surface information.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for generating longitudinal deceleration markers based on perception data.

[0007] Obtain lane line data for deceleration sections;

[0008] Several anchor points are set on the lane line data of the deceleration section, and the direction of the anchor points is the same as the forward direction of the lane line data.

[0009] Based on the coordinate data of the anchor points, the coordinates of the contour points of several deceleration markers are obtained, and then the deceleration markers are drawn. Each deceleration marker corresponds to one anchor point.

[0010] The completed deceleration sign will be displayed on the screen.

[0011] Based on the above technical means, after obtaining the original lane line data, the coordinates of several deceleration sign outline points are calculated by calibrating anchor points. Then, deceleration signs are drawn based on the coordinates of the outline points and displayed on the display screen, realizing the visualization of deceleration signs. In this way, when drivers enter the deceleration section and observe the lane-level navigation data on the display screen, the drawn deceleration sign pattern will be displayed on the lane. Compared with the existing technology, the salience of deceleration signs in navigation data is improved, providing drivers and passengers with accurate and real-time road information.

[0012] Meanwhile, since the deceleration indicator is generated by drawing in the above-mentioned technical means, this method is not affected by scene scaling or resolution changes, and is more flexible if certain scenes (such as skinning) require customized display.

[0013] Furthermore, the step of obtaining the coordinates of several deceleration marker contour points based on the anchor point coordinate data is executed on the CPU, while the step of drawing the deceleration markers is executed on the GPU.

[0014] Based on the above technical means, the process of creating depiction data and the process of rendering depiction data are made independent of each other and do not affect each other.

[0015] Furthermore, after obtaining the coordinates of several deceleration marker contour points based on the anchor point coordinate data, all the contour point data is rendered and converted into data supported by the graphics API. Then, instructions are sent to the GPU to draw the deceleration markers.

[0016] Furthermore, the coordinates of the contour points are points on a two-dimensional plane, and the plurality of contour points include a first contour point, a second contour point, a third contour point, and a fourth contour point. The method for obtaining the coordinates of the first contour point, the second contour point, the third contour point, and the fourth contour point is as follows:

[0017] On a two-dimensional plane, the coordinates of the anchor point are moved a first preset distance in a direction perpendicular to the lane line to obtain the coordinates of the midpoint;

[0018] The midpoint is moved a second preset distance towards both the forward and backward directions of the lane line to obtain the coordinates of the first contour point and the second contour point.

[0019] Move the coordinates of the first contour point and the second contour point in the preset direction by a third preset distance to obtain the coordinates of the third contour point and the fourth contour point, respectively.

[0020] Furthermore, after obtaining the lane line data of the deceleration section, the length of the lane line is calculated, and the position of the anchor point is set according to the preset spacing to obtain the coordinates of the anchor point.

[0021] A longitudinal deceleration marker generation system based on the aforementioned longitudinal deceleration marker generation method based on sensing data, comprising:

[0022] The acquisition unit is configured to obtain lane line data for deceleration sections;

[0023] The scene reconstruction unit is configured to set several anchor points on the lane line data of the deceleration section, the direction of the anchor points being the same as the forward direction of the lane line data, and to obtain the coordinates of the outline points of several deceleration markers based on the coordinate data of the anchor points.

[0024] The drawing unit is configured to draw a deceleration indicator based on the coordinates of the contour points;

[0025] The display unit is configured to display the completed deceleration indicator on the display screen.

[0026] Furthermore, the step of obtaining the coordinates of several deceleration marker contour points based on the anchor point coordinate data is executed on the CPU, the step of drawing the deceleration markers is executed on the GPU, the scene reconstruction unit is integrated on the CPU, and the drawing unit is integrated on the GPU.

[0027] Furthermore, the scene reconstruction unit includes a data creation module and a data rendering module;

[0028] The data creation module is configured to obtain the coordinates of several deceleration marker contour points based on the coordinate data of the anchor points, and then send the coordinates of the contour points to the data rendering module.

[0029] The data rendering module is configured to render all the contour point data, convert it into data supported by the graphics API, and then send instructions to the GPU.

[0030] Furthermore, the coordinates of the contour points are points on a two-dimensional plane, and the plurality of contour points include a first contour point, a second contour point, a third contour point, and a fourth contour point. The method for obtaining the coordinates of the first contour point, the second contour point, the third contour point, and the fourth contour point is as follows:

[0031] On a two-dimensional plane, the coordinates of the anchor point are moved a first preset distance in a direction perpendicular to the lane line to obtain the coordinates of the midpoint;

[0032] The midpoint is moved a second preset distance towards both the forward and backward directions of the lane line to obtain the coordinates of the first contour point and the second contour point.

[0033] Move the coordinates of the first contour point and the second contour point in the preset direction by a third preset distance to obtain the coordinates of the third contour point and the fourth contour point, respectively.

[0034] Furthermore, after obtaining the lane line data of the deceleration section, the length of the lane line is calculated, and the position of the anchor point is set according to the preset spacing to obtain the coordinates of the anchor point.

[0035] The beneficial effects of this invention are:

[0036] This invention proposes to use perceived data (raw data) as the input source, assemble the raw data into depiction data using basic computer graphics theory, and finally restore the longitudinal deceleration markers in the in-vehicle entertainment system, thereby providing real-time feedback on the vehicle's driving status and providing accurate and real-time road information for the driver and passengers.

[0037] This invention utilizes radar perception and visual recognition lane marking schemes based on computer graphics principles to rapidly generate drawing data on the CPU. This line data is then used as the raw input, enabling the in-vehicle entertainment terminal to reproduce the vehicle scene as accurately as possible and display longitudinal deceleration signs, thereby serving as a warning to remind users to slow down.

[0038] This invention utilizes the high concurrency capabilities of GPUs to accurately and in real-time display longitudinal deceleration markers on the lane markings on the screen. While maintaining speed and accuracy, this solution is unaffected by scene scaling or resolution changes. It also offers greater flexibility for scenarios requiring customized display (such as skinning). Attached Figure Description

[0039] Figure 1 This is a flowchart of Embodiment 1 of the present invention;

[0040] Figure 2 A summary of the original data flow process;

[0041] Figure 3 Generate a schematic diagram of longitudinal deceleration markings for lane lines;

[0042] Figure 4 The principle behind creating longitudinal deceleration marker data;

[0043] Figure 5 This is a structural diagram of Embodiment 2 of the present invention;

[0044] Figure 6 This is a schematic diagram of the data flow in Example 2;

[0045] Figure 7 This is a schematic diagram of the data flow in Example 2.

[0046] Among them, 1-Acquisition unit; 2-Scene reconstruction unit; 21-Data creation module; 22-Data rendering module; 3-Drawing unit; 4-Display unit. Detailed Implementation

[0047] The following description, with reference to the accompanying drawings and preferred embodiments, illustrates the implementation of the technical solution of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0048] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0049] Example 1

[0050] This embodiment proposes a method for generating longitudinal deceleration markers based on perception data, such as... Figure 1 As shown, the specific method is as follows:

[0051] S1: Obtain lane line data for the deceleration section.

[0052] In this step, such as Figure 2 As shown, based on existing cameras and radar, the original information of lane-level data is obtained in real time. When the lane-level data corresponding to the vehicle position is identified by radar perception and visual recognition and contains deceleration indicators, namely the diamond-shaped virtual blocks set along the direction of lane line movement mentioned in the background technology, it is considered that the vehicle has entered the deceleration section. At this time, the lane line data of the deceleration section is collected to obtain the lane line.

[0053] S2: As Figure 3 As shown, several anchor points are set on the lane line data of the deceleration section, and the direction of the anchor points is the same as the forward direction of the lane line data.

[0054] S3: Based on the coordinate data of the anchor points, obtain the coordinates of the outline points of several deceleration markers, and then draw them into deceleration markers.

[0055] This implementation example Figure 4 As shown, the method for drawing deceleration indicators is as follows:

[0056] First, using the point data of the lane line as a reference, calculate the length d of the lane line segment;

[0057] In the direction of travel of the lane line, an interpolation is performed according to the lane line length d and a certain density to obtain a series of anchor points on the lane line. These anchor points are the reference points for making longitudinal deceleration markings. In this embodiment, the density is represented by the preset spacing between adjacent deceleration markings.

[0058] For each anchor point, a longitudinal deceleration indicator is generated on both the left and right sides of the lane line. In summary, this yields a lane line with longitudinal deceleration indicators, as shown in the attached diagram. Figure 4 .

[0059] The method for creating longitudinal deceleration markings is as follows:

[0060] Given the coordinates of the current anchor point and lane lines forward direction (Obtained by subtracting adjacent anchor points). Since lane lines and longitudinal deceleration markings are on the lane surface, the height dimension can be ignored. and .

[0061] Based on the direction of travel along the lane lines, and the fact that the product of the slopes of mutually perpendicular vectors is -1, we can derive two directions perpendicular to the lane lines, located to the right of the lane lines. and left side .

[0062] The following explanation uses the production of a longitudinal deceleration sign on one side as an example:

[0063] Extend the anchor point a first preset distance perpendicular to the lane line to obtain the midpoint on one side of the longitudinal deceleration marker: ,

[0064] At point m, extend a second preset distance along the direction of travel of the lane line and in the opposite direction of travel of the lane line, respectively, to obtain two points: and .

[0065] Rotate the lane line's forward direction by 45° clockwise and counterclockwise around the Z-axis respectively to obtain the hypotenuse directions dl and dr: , where m_rot(45) is a rotation matrix that rotates 45° around the Z-axis (graphics theory) to form a preset direction.

[0066] Finally in and Press up Extending the r direction by a third preset distance (the length of the hypotenuse) yields... and point.

[0067] , .

[0068] In conclusion, , , and The four points serve as the outline points of the deceleration marker in this embodiment, according to... and The distribution forms two triangles, which serve as descriptive data for constructing longitudinal deceleration markers.

[0069] S4: Display the completed deceleration markings on the in-vehicle entertainment system's screen.

[0070] In S4, the generated deceleration indicator data is generated by and The four points are all data with coordinates. Therefore, when the deceleration signs drawn from these four data points are displayed on the screen, the coordinates of the four points correspond to the coordinates of the lane lines. This ensures that the density of the deceleration signs displayed on the screen and their relative positions to the lane lines are the same as the designed relative positions.

[0071] The overall approach includes data creation and rendering, ensuring that the data creation and rendering processes are independent and do not interfere with each other. Upon receiving lane line data with longitudinal deceleration indicators, the data is sent to the data creation thread to begin creating the rendering data. Once data creation is complete, the rendering thread is notified to send instructions to the GPU to begin drawing.

[0072] This embodiment utilizes radar perception and visual recognition of lanes. Based on computer graphics principles, the solution rapidly generates drawing data on the CPU and uses the line data as the raw input. This enables the vehicle terminal to recreate the vehicle scene as accurately as possible and display longitudinal deceleration signs, thereby reminding users to slow down.

[0073] Leveraging the high concurrency capabilities of GPUs, vertical deceleration markers are accurately and in real-time displayed on the lane markings on the screen. Compared to texture mapping solutions, this approach maintains speed and accuracy while remaining unaffected by scene scaling or resolution changes. It also offers greater flexibility for customized displays in certain scenarios (such as skinning).

[0074] In this embodiment, after collecting lane line data for the deceleration section in S1, the data is sent to the CPU. The CPU then calculates the drawing data points using methods S2-S3. , , and After obtaining the coordinates of the four contour points, the coordinate data is converted into data supported by the graphics API through data rendering. Then, instructions are sent to the GPU, which has a graphics pipeline, to draw the deceleration indicator.

[0075] In this embodiment, the deceleration indicator and lane lines are displayed together on the display screen of the in-vehicle entertainment system. This is because the lane lines themselves can be displayed on the display screen of the in-vehicle entertainment system, which is possible with existing technology. The deceleration indicator is made on both sides of the lane lines. In S3, the outline points of the deceleration indicator are represented in the form of coordinates. The coordinates are related to the position and spacing of the lane lines, so it can be ensured that the position of the calculated outline points coincides with the display position.

[0076] Example 2

[0077] This embodiment, based on Embodiment 1, proposes a longitudinal deceleration marker generation system based on perception data, such as... Figure 5 As shown, it includes:

[0078] Acquisition Unit 1 is configured to obtain lane line data for deceleration sections;

[0079] Scene reconstruction unit 2 is configured to set several anchor points on the lane line data of the deceleration section. The direction of the anchor points is the same as the forward direction of the lane line data. Based on the coordinate data of the anchor points, the coordinates of the outline points of several deceleration signs are obtained.

[0080] Drawing unit 3 is configured to draw a deceleration indicator based on the coordinates of contour points;

[0081] Display unit 4 is configured to display the completed deceleration sign on the navigation data of the in-vehicle entertainment system's display screen.

[0082] In this embodiment, the scene reconstruction unit is integrated on the CPU; the rendering unit 3 is integrated on the GPU.

[0083] In this embodiment, the scene reconstruction unit 2 includes a data creation module 21 and a data rendering module 22;

[0084] The drawing data module 21 is configured to obtain the coordinates of several deceleration marker contour points based on the coordinate data of the anchor points, and then send the coordinates of the contour points to the data rendering module 22.

[0085] The data rendering module 22 is configured to render all the contour point data, convert it into data supported by the graphics API, and then send instructions to the GPU. After receiving the instructions, the drawing unit 3 draws the deceleration mark based on the drawing data.

[0086] In this embodiment, as Figure 7 As shown, the autonomous driving system publishes the collected lane marking data via the DDS protocol. The cockpit system subscribes to this message and retrieves the data from the data bus. After the data arrives inside the cockpit system, it is then transmitted a second time through acquisition unit 1 before being published to scene reconstruction unit 2.

[0087] like Figure 6 As shown, after receiving the data, the scene reconstruction unit 2 calculates the data based on S2-S3 of embodiment 1 in the data creation module 21. , , and Then, a drawing request is sent to the data rendering module 22. The data rendering module 22 renders the calculated data into data that the graphics API can understand, and then sends the instructions to the platform GPU to perform drawing.

[0088] The above embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention.

Claims

1. A method for generating longitudinal deceleration markers based on sensing data, characterized in that: Obtain lane line data for deceleration sections; Several anchor points are set on the lane line data of the deceleration section, and the direction of the anchor points is the same as the forward direction of the lane line data. Based on the coordinate data of the anchor points, the coordinates of the contour points of several deceleration markers are obtained, and then the deceleration markers are drawn. Each deceleration marker corresponds to one anchor point. Display the completed deceleration indicator on the screen; The coordinates of the contour points are points on a two-dimensional plane. The plurality of contour points includes a first contour point, a second contour point, a third contour point, and a fourth contour point. The method for obtaining the coordinates of the first contour point, the second contour point, the third contour point, and the fourth contour point is as follows: On a two-dimensional plane, the coordinates of the anchor point are moved a first preset distance in a direction perpendicular to the lane line to obtain the coordinates of the midpoint; The midpoint is moved a second preset distance towards both the forward and backward directions of the lane line to obtain the coordinates of the first contour point and the second contour point. Move the coordinates of the first contour point and the second contour point in the preset direction by a third preset distance to obtain the coordinates of the third contour point and the fourth contour point, respectively.

2. The longitudinal deceleration marker generation method based on sensing data according to claim 1, characterized in that: The step of obtaining the coordinates of several deceleration marker contour points based on the coordinate data of the anchor points is executed on the CPU, while the step of drawing the deceleration markers is executed on the GPU.

3. The longitudinal deceleration marker generation method based on sensing data according to claim 2, characterized in that: After obtaining the coordinates of several deceleration marker contour points based on the anchor point coordinate data, all the contour point data is rendered and converted into data supported by the graphics API. Then, instructions are sent to the GPU to draw the deceleration markers.

4. The longitudinal deceleration marker generation method based on sensing data according to claim 3, characterized in that: After obtaining the lane line data of the deceleration section, the length of the lane line is calculated, and the position of the anchor point is set according to the preset spacing to obtain the coordinates of the anchor point.

5. A longitudinal deceleration marker generation system based on sensing data, based on the longitudinal deceleration marker generation method based on sensing data according to any one of claims 1-4, characterized in that: include: The acquisition unit is configured to obtain lane line data for deceleration sections; The scene reconstruction unit is configured to set several anchor points on the lane line data of the deceleration section, the direction of the anchor points being the same as the forward direction of the lane line data, and to obtain the coordinates of the outline points of several deceleration markers based on the coordinate data of the anchor points. The drawing unit is configured to draw deceleration markers based on the coordinates of the contour points, with each deceleration marker corresponding to an anchor point. The display unit is configured to display the completed deceleration indicator on the display screen.

6. The longitudinal deceleration marker generation system based on perception data according to claim 5, characterized in that: The step of obtaining the coordinates of several deceleration marker contour points based on the coordinate data of the anchor points is executed on the CPU, the step of drawing the deceleration markers is executed on the GPU, the scene reconstruction unit is integrated on the CPU, and the drawing unit is integrated on the GPU.

7. The longitudinal deceleration marker generation system based on perception data according to claim 6, characterized in that: The scene reconstruction unit includes a data creation module and a data rendering module; The data creation module is configured to obtain the coordinates of several deceleration marker contour points based on the coordinate data of the anchor points, and then send the coordinates of the contour points to the data rendering module. The data rendering module is configured to render all the contour point data, convert it into data supported by the graphics API, and then send instructions to the GPU.

8. The longitudinal deceleration marker generation system based on perception data according to claim 6, characterized in that: The coordinates of the contour points are points on a two-dimensional plane. The plurality of contour points includes a first contour point, a second contour point, a third contour point, and a fourth contour point. The method for obtaining the coordinates of the first contour point, the second contour point, the third contour point, and the fourth contour point is as follows: On a two-dimensional plane, the coordinates of the anchor point are moved a first preset distance in a direction perpendicular to the lane line to obtain the coordinates of the midpoint; The midpoint is moved a second preset distance towards both the forward and backward directions of the lane line to obtain the coordinates of the first contour point and the second contour point. Move the coordinates of the first contour point and the second contour point in the preset direction by a third preset distance to obtain the coordinates of the third contour point and the fourth contour point, respectively.

9. The longitudinal deceleration marker generation system based on perception data according to claim 8, characterized in that: After obtaining the lane line data of the deceleration section, the length of the lane line is calculated, and the position of the anchor point is set according to the preset spacing to obtain the coordinates of the anchor point.

Images