Suspension data acquisition method based on positioning of luminous pyramid

Abstract

The invention discloses a suspension data acquisition method based on positioning of a luminous pyramid. The method comprises the following steps: S10, mounting suspension detection equipment for positioning the luminous pyramid at a suspension position of a vehicle; s20, the relative relation between the ground and the claw is confirmed; and S30, performing digital twinning processing to obtain suspension data acquisition data. According to the invention, detection can be carried out along with on-vehicle on-road driving, and suspension data acquisition can be efficiently and accurately carried out.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle suspension, and particularly relates to a suspension data acquisition method based on a positionable light-emitting pyramid. Background Art

[0002] For current mainstream vehicle attitude sensing solutions, their positioning mainly relies on multiple sets of angle sensors and distance sensors connected to various components of the suspension. Through programming by the user for the entire suspension system and multiple sensors (such as rocker-type angle sensors fixed on the lower control arm of the chassis and distance sensors fixed on the suspension and the chassis), real-time monitoring of the vehicle suspension is achieved. Since the wheels are rigidly fixed to the vehicle suspension structural components, the system can relatively easily calculate the running conditions of the wheels themselves.

[0003] Since only the rigid components of the suspension are measured, the deformation of flexible components (such as tires, bushings, etc.) in the suspension is not considered during the data reduction stage, resulting in not only inherent deviations in the measurement of the tire running state but also a high dependence on professionals for programming in data derivation.

[0004] Moreover, the detection equipment is bulky and requires high-level debugging. The vehicle needs to be fixed on a stationary machine platform in a closed and dry environment, which not only has high costs and fixed positions and is completely unable to be used on a moving vehicle but also highly depends on professionals for debugging work. And due to the interference of flexible components (tires, bushings) in the suspension, accurate data cannot be obtained, and only relatively accurate indirect measurements of suspension characteristics can be performed. The measured suspension motion data itself cannot guarantee to be within a reasonable confidence interval. Summary of the Invention

[0005] To solve the above problems, the present invention proposes a suspension data acquisition method based on a positionable light-emitting pyramid, which can be used for detection while the vehicle is on the road and can efficiently and accurately collect suspension data.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a suspension data acquisition method based on a positionable light-emitting pyramid, including the steps of:

[0007] S10, installing a suspension detection device with a positionable light-emitting pyramid at the vehicle suspension;

[0008] S20, confirming the relative relationship between the ground and the knuckle;

[0009] S30, performing digital twin processing to obtain suspension data acquisition data.

[0010] Furthermore, installing a positionable light-emitting pyramid at the vehicle suspension includes the steps of:

[0011] S101. Place the device flat on the ground and use the built-in gyroscope of the fuselage to determine the state of the ground at this time;

[0012] S102. Place the device at the bracket inside the wheel arch;

[0013] S103. Calculate the relative attitude between the position of the device in the wheel arch and the ground according to the gyroscope data;

[0014] S104. Use the point cloud data obtained by the ToF sensor to calculate the relative attitude relationship between the ground and the device obtained according to the point cloud data;

[0015] S105. Repeat this process and use the two data to correct each other to obtain the relative attitude relationship between the ground and the device, and install the suspension detection device based on the positioning luminous pyramid.

[0016] Furthermore, confirm the relative relationship between the ground and the knuckle, including the steps:

[0017] S201. Calculate the relative position relationship between the positioning block and the device by using the feature points on the positioning block;

[0018] S202. For different vehicle models, pre-confirm the accurate attitude relative relationship between the positioning block on the knuckle, the positioning pattern on the swing arm, and the suspension components and the ground through tests in the actual environment, and test the movement trajectory of the positioning block;

[0019] S203. Turn the steering wheel and calibrate the difference between the actual position and the initial design position of the device in the wheel arch bracket by tracking the position change relationship of the pattern and the trajectory difference in the original tracking detection design through the camera;

[0020] S204. After completing the detection and calibration process, generate a data acquisition model for confirming the suspension components through the visual feature positioning points.

[0021] Furthermore, calculate the relative position relationship between the positioning block and the device by using the feature points on the positioning block, including the steps:

[0022] S2011. Patterns with area differences are set on each surface of the polyhedron positioning block, and coatings with different reflectivities are attached to each surface, so that very obvious edge features are generated in image processing, and the existing pattern boundaries are shown by the sharp change of pixel intensity;

[0023] S2012. The camera finds the corners through the corner detection algorithm by obtaining the corner points on each plane and marks them as feature points;

[0024] S2013. After multiple corner points are recognized and matched, use these features to confirm the edges and shapes of the object;

[0025] In 2014, match the feature points enclosing the shape in the current frame with the feature points in subsequent frames;

[0026] In S2015, after confirming the boundaries and feature points of the tracking pattern, calculate the pose of the plane where the pattern is located.

[0027] Furthermore, perform digital twin processing to obtain suspension data acquisition data, including the steps of:

[0028] S301, pre-scan and model the suspension structure;

[0029] S302, bind the spatial position information to the digital model of the chassis structure to track rigid components in real time;

[0030] S303, track non-rigid components in real time through sensors;

[0031] S304, output in real time the digital model including the rigid component part and the non-rigid component part; combine the outputs of S302 and S303 to show the position changes of the rigid components and the morphological changes of the non-rigid components in the digital model in real time.

[0032] Furthermore, pre-scan and model the suspension structure, including:

[0033] S3011, use high-precision scanning for the scanning method of the suspension structure, and pre-construct a high-precision digital model of the chassis structure including a tracking and positioning pyramid system;

[0034] S3012, select the key motion feature points showing the system motion in the suspension structure.

[0035] Furthermore, bind the spatial position information to the digital model of the chassis structure to track rigid components in real time, including the steps of:

[0036] S3021, through the positional relationship between the known key motion feature points and the positioning points of the positioning and tracking system, construct a relationship equation between the key motion feature points and the positioning points of the positioning and tracking system, determine the relative positions between the key motion feature points and the positioning points of the positioning and tracking system, and bind the relationship between the key motion feature points and the positioning points of the positioning and tracking system;

[0037] S3022, the camera continuously tracks the luminous pyramid and the wheel-side luminous structure, shows the real-time motion of the tracking points of the positioning system, and obtains the real-time spatial positions of the key motion feature points through calculation;

[0038] S3023, by binding the key motion feature points in the digital model to the key motion feature points calculated through the equation, bind the positional relationship between the positioning and tracking feature points in the digital model and the positioning and tracking feature points obtained by the camera;

[0039] S3024, Output the position changes of the suspension system in real time and present the data through a digital model in real time.

[0040] Furthermore, the non-rigid components are tracked in real time through sensors, including the steps of:

[0041] S3031, Scan the non-rigid components in the suspension structure in real time through a sensor with three-dimensional imaging capabilities to form an updated point cloud, and perform modeling operations on the point cloud data obtained by the sensor in real time;

[0042] S3032, Output the model of the non-rigid components and present the changes through the digital model in real time.

[0043] Beneficial effects of adopting this technical solution:

[0044] The present invention is applicable to a moving vehicle, and the installation method of the device itself will not cause any burden to the normal use of the user. The present invention adopts a non-invasive fixing method, which greatly reduces the workload, professional knowledge cost and obstacles of the user when using any similar products or analogous products.

[0045] Compared with the existing method for obtaining the position data of the rigid components of the suspension, the present invention can display the position state of the suspension system in real time, enabling the user to more intuitively understand the movement relationship of the suspension.

[0046] Compared with the existing method for obtaining the state changes of the non-rigid components of the suspension, the present invention can be based on real-time observation, and the obtained data is more intuitive and accurate.

[0047] Compared with the existing method for obtaining data of the chassis control algorithm, the present invention can not only serve as a data basis, but also directly present the data changes to the user.

[0048] By using the present invention, the user can display the position state of the rigid components and the real-time state of the non-rigid components in the suspension system in real time, enabling the user to more intuitively understand the movement relationship of the suspension and the current state.

[0049] By connecting the data obtained by the present invention to the chassis control algorithm, the state of the vehicle can be adjusted in real time using the control algorithm without using additional invasive devices.

[0050] By using the real-time non-rigid component model obtained by the present invention, the user can conduct a deeper study on the material properties of the non-rigid components. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 Schematic flow diagram of a suspension data acquisition method based on a positionable light-emitting pyramid according to the present invention;

[0052] Figure 2 It is a schematic flow chart of the installation stage in the embodiment of the present invention;

[0053] Figure 3 It is a schematic flow chart of confirming the relative relationship between the ground and the knuckle in the embodiment of the present invention;

[0054] Figure 4 It is a schematic flow chart of high-precision scanning and modeling of the suspension structure in the embodiment of the present invention;

[0055] Figure 5 It is a schematic flow chart of binding the spatial position information with the digital model of the chassis structure to track the rigid components in real time in the embodiment of the present invention;

[0056] Figure 6 It is a schematic flow chart of tracking non-rigid components in real time through sensors in the embodiment of the present invention. Detailed implementation manners

[0057] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described below with reference to the accompanying drawings.

[0058] In this embodiment, referring to Figure 1 as shown, the present invention proposes a suspension data acquisition method based on a positionable light-emitting pyramid, including the steps of:

[0059] S10. Install a suspension detection device with a positionable light-emitting pyramid at the vehicle suspension;

[0060] S20. Confirm the relative relationship between the ground and the knuckle;

[0061] S30. Perform digital twin processing to obtain suspension data acquisition data.

[0062] As an optimized solution of the above embodiment, as Figure 2 shown, installing a positionable light-emitting pyramid at the vehicle suspension includes the steps of:

[0063] S101. Place the device flat on the ground and use the built-in gyroscope of the fuselage to determine the state of the ground at this time;

[0064] S102. Place the device at the bracket inside the wheel arch;

[0065] S103. Calculate the relative attitude of the device in the wheel arch with respect to the ground according to the gyroscope data;

[0066] S104. Use the point cloud data obtained by the ToF sensor to calculate the relative attitude relationship between the ground and the device obtained from the point cloud data;

[0067] S105, repeat this process and use the two data to calibrate each other, obtain the relative posture relationship between the ground and the equipment, and install the suspension detection equipment based on the positioning luminous pyramid.

[0068] As an optimization solution of the above embodiment, Figure 3 As shown, confirming the relative relationship between the ground and the horns includes the following steps:

[0069] S201, calculating the relative position relationship between the positioning block and the device using the feature points on the positioning block;

[0070] S202, confirming the accurate relative posture relationship between the positioning block on the horn, the positioning pattern on the swing arm, and the suspension components and the ground through tests in the actual environment for different vehicle models, and testing the motion trajectory of the positioning block;

[0071] S203, turning the steering wheel, and calibrating the difference between the actual position of the bracket in the wheel arch and the originally designed position by using the position change relationship of the camera tracking pattern and the trajectory difference in the original tracking detection design;

[0072] S204, after completing the detection and calibration process, a data acquisition model of the suspension components confirmed by visual feature positioning points is generated.

[0073] Among them, Figure 3 As shown, the relative position relationship between the positioning block and the device is calculated using the feature points on the positioning block, including the steps of:

[0074] S2011, patterns with different regions are set on each surface of the polyhedral positioning block, and coatings with different reflectivity are attached to each surface, so that very obvious edge features are produced in image processing, and the existing pattern boundaries are displayed by using the sharp change of pixel intensity. Some common CV algorithms, such as Canny edge detection, can be used to detect the edge in this part;

[0075] In S2012, the camera obtains the corner points on each plane. The corner points are the points where the direction in the image changes significantly and usually appear at the corners of the image edge. The corner detection algorithm is used to find the corners and mark them as feature points. Some mature algorithms can be used to calculate the positions of the corner points. For example, in the FAST method, based on the pixel positions of the edge features obtained in the previous step, using the FAST corner detection algorithm, the pixel point p to be detected and its surrounding 16 pixels (forming a circle with a radius of 3, numbered from 1 to 16 in sequence) are selected. A brightness threshold t is set. If the pixel p is a corner point, there should be a set of continuous pixels in the circular neighborhood, whose brightness is either higher than the brightness of p plus the threshold t or lower than the brightness of p minus t. Assuming there are at least 12 continuous pixels, then we can consider that this pixel point meets this condition and determine that the pixel p is a corner point.

[0076] In S2013, after multiple corner points are recognized and matched, these features are used to confirm the edges and shapes of the object. The shape matching algorithm (such as matchShapes or DPM in OpenCV, and training a CNN model with a large amount of data) determines the corresponding shape based on the characteristics of different contours. Since we use different patterns on different planes, the plane numbers can also be determined simultaneously.

[0077] In S2014, the feature points that enclose the shape in the current frame are matched with the feature points in the subsequent frames. Feature point matching can be achieved using descriptors (such as ORB or SIFT descriptors). This method can continuously track the positions and shapes in different frames.

[0078] In S2015, after confirming the boundaries and feature points of the tracking pattern, the pose of the plane where the pattern is located is calculated. The pose calculation can be completed through the PnP (Perspective-n-Point) algorithm: 3D points correspond to 2D points. If the 3D coordinates of the tracking pattern in space are known (for example, by pre-calibrating the physical size of the tracking pattern), these 3D coordinates can be corresponded to the 2D feature points in the image. Using this set of 3D and 2D points, the rotation and translation of the camera relative to the plane of the tracking pattern are calculated through the PnP algorithm. These rotation and translation parameters represent the pose of the plane, that is, the position and angle between the camera and the tracking pattern. For errors, we can filter the noise points through RANSAC (Random Sample Consensus algorithm) to ensure that the pose estimation of the tracking pattern is more stable. The pose parameters can be further optimized using Bundle Adjustment to minimize the projection error.

[0079] As an optimized solution of the above embodiment, digital twin processing is performed to obtain the suspension data acquisition data, including the steps:

[0080] S301, Pre-scan and model the suspension structure in advance;

[0081] S302, Bind the spatial position information to the digital model of the chassis structure to track the rigid components in real time;

[0082] S303, Track the non-rigid components in real time through sensors;

[0083] S304, Output the digital model including the rigid component part and the non-rigid component part in real time; Combine the outputs of S302 and S303 to show the position changes of the rigid components and the morphological changes of the non-rigid components in the digital model in real time.

[0084] Among them, as Figure 4 shown, pre-scanning and modeling the suspension structure in advance includes:

[0085] S3011, Use high-precision scanning for the scanning method of the suspension structure, and pre-construct a high-precision digital model of the chassis structure including a tracking and positioning pyramid system;

[0086] S3012, Select the key motion feature points showing the system motion in the suspension structure, such as the positions of the suspension component connection points.

[0087] Among them, as Figure 5 shown, binding the spatial position information to the digital model of the chassis structure to track the rigid components in real time includes the steps:

[0088] S3021, Through the positional relationship between the known key motion feature points and the positioning points of the positioning and tracking system, construct the relationship equation between the key motion feature points and the positioning points of the positioning and tracking system, determine the relative positions between the key motion feature points and the positioning points of the positioning and tracking system, and bind the relationship between the key motion feature points and the positioning points of the positioning and tracking system;

[0089] S3022, The camera continuously tracks the light-emitting pyramid and the wheel-side light-emitting structure, shows the real-time motion of the tracking points of the positioning system, and obtains the real-time spatial positions of the key motion feature points through calculation;

[0090] S3023, By binding the key motion feature points in the digital model to the key motion feature points calculated by the equation, bind the positional relationship between the positioning and tracking feature points in the digital model and the positioning and tracking feature points obtained by the camera;

[0091] S3024, Output the position changes of the suspension system in real time and present the data in the digital model in real time.

[0092] Among them, as Figure 6 shown, tracking the non-rigid components in real time through sensors includes the steps:

[0093] S3031, Use a sensor with three-dimensional imaging capabilities to scan non-rigid components in the suspension structure in real time, such as rubber bushings and tires, etc., to form an ever-updating point cloud, and perform real-time modeling operations on the point cloud data obtained by the sensor;

[0094] S3032, Output the model of the non-rigid component and present the changes in real time through the digital model.

[0095] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A suspension data acquisition method based on a positionable and light-emitting pyramid, characterized in that Including the steps: S10, Install a suspension detection device for positioning a glow-in-the-dark pyramid at the vehicle suspension; S20, Confirm the relative relationship between the ground and the knuckle; S30, Perform digital twin processing to obtain suspension data acquisition data.

2. The suspension data acquisition method based on a positionable light-emitting pyramid according to claim 1, wherein, Installing a glow-in-the-dark pyramid for positioning at the vehicle suspension includes the steps: S101, Place the device flat on the ground and use the built-in gyroscope of the fuselage to determine the state of the ground at this time; S102, Place the device at the bracket inside the wheel arch; S103, Calculate the relative attitude between the position of the device in the wheel arch and the ground according to the gyroscope data; S104, Use the point cloud data obtained by the ToF sensor to calculate the relative attitude relationship between the ground and the device based on the point cloud data; S105, Repeat this process and use the two data to correct each other to obtain the relative attitude relationship between the ground and the device, and install the suspension detection device based on the glow-in-the-dark pyramid for positioning.

3. The suspension data acquisition method based on a positionable light-emitting pyramid according to claim 1, wherein, Confirming the relative relationship between the ground and the knuckle includes the steps: S201, Calculate the relative position relationship between the positioning block and the device using the feature points on the positioning block; S202, For different vehicle models, pre-confirm the accurate attitude relative relationship between the positioning block on the knuckle, the positioning pattern on the swing arm, and the suspension components and the ground through tests in the actual environment, and test the movement trajectory of the positioning block; S203, Turn the steering wheel and calibrate the difference between the actual position and the initial design position of the device in the wheel arch bracket by tracking the position change relationship of the pattern and the trajectory difference in the original tracking detection design through the camera; S204, After completing the detection and calibration process, generate a data acquisition model for the suspension components by confirming the positioning points through visual features.

4. A suspension data acquisition method based on a positionable and light-emitting pyramid according to claim 3, wherein, Calculating the relative position relationship between the positioning block and the device using the feature points on the positioning block includes the steps: S2011, Patterns with regional differences are set on each surface of the polyhedron positioning block, and coatings with different reflectivities are attached to each surface, so that very obvious edge features are generated in image processing, and the existing pattern boundaries are shown using the sharp change in pixel intensity; S2012, The camera finds the corners through the corner detection algorithm using the corner points on each plane obtained and marks them as feature points; S2013, When multiple corner points are recognized and matched, use these features to confirm the edges and shape of the object; S2014, Match the feature points that enclose the shape in the current frame with the feature points in the subsequent frame; S2015, After confirming the boundaries and feature points of the tracking pattern, calculate the attitude of the plane where the pattern is located.

5. A suspension data acquisition method based on a positionable light-emitting pyramid according to claim 1, characterized in that, Performing digital twin processing to obtain suspension data acquisition data includes the steps: S301, Pre-scan and model the suspension structure; S302, Bind the spatial position information to the digital model of the chassis structure to track the rigid components in real time; S303, Real-time track non-rigid components through sensors; S304, Output a digital model including the rigid component part and the non-rigid component part in real time; Combine the outputs of S302 and S303 to show the position change of the rigid components and the morphological change of the non-rigid components in the digital model in real time.

6. A suspension data acquisition method based on a positionable luminous pyramid according to claim 5, characterized in that, Pre-scanning and modeling the suspension structure includes: S3011. Use high-precision scanning for the scanning method of the suspension structure, and pre-construct a high-precision digital model of the chassis structure including a tracking and positioning pyramid system; S3012. Select the key motion feature points that show the system motion in the suspension structure.

7. A suspension data acquisition method based on a positionable and luminous pyramid according to claim 5, characterized in that, Bind the spatial position information to the digital model of the chassis structure to track the rigid components in real time, including the steps: S3021. Construct a relationship equation between the key motion feature points and the positioning points of the positioning and tracking system through the position relationship between the known key motion feature points and the positioning points of the positioning and tracking system, determine the relative position between the key motion feature points and the positioning points of the positioning and tracking system, and bind the relationship between the key motion feature points and the positioning points of the positioning and tracking system; S3022. The camera continuously tracks the light-emitting pyramid and the wheel-side light-emitting structure, shows the real-time motion of the tracking points of the positioning system, and obtains the real-time spatial position of the key motion feature points through calculation; S3023. Bind the key motion feature points in the digital model to the key motion feature points calculated by the equation, and bind the position relationship between the positioning and tracking feature points in the digital model and the positioning and tracking feature points obtained by the camera; S3024. Output the position change of the suspension system in real time, and present the data in real time through the digital model.

8. A suspension data acquisition method based on a positionable and light-emitting pyramid according to claim 5, characterized in that Track the non-rigid components in real time through sensors, including the steps: S3031. Use a sensor with three-dimensional imaging ability to scan the non-rigid components in the suspension structure in real time to form an updated point cloud, and perform real-time modeling operations on the point cloud data obtained by the sensor; S3032. Output the model of the non-rigid components, and present the changes in real time through the digital model.

Images