A calibration system for geometric parameters of a vehicle under test in a full vehicle in-the-loop test

Abstract

The present invention discloses a calibration system for geometric parameters of a vehicle under test in a vehicle-in-the-loop test. The controller controls two industrial cameras of a parameter calibrator to capture the steering videos of the left front wheel and the right front wheel. Based on the target tracking algorithm, the steering videos of the left front wheel and the right front wheel are respectively tracked and processed to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation and the second distance from the center of the right front wheel to the center of the right front wheel's rotation. According to the first distance, the second distance, the first extension distance of the industrial camera corresponding to the first servo cylinder, and the second extension distance of the industrial camera corresponding to the second servo cylinder, the kingpin offset of the vehicle under test is calculated. The current steering wheel angle of the vehicle under test sent by the ECU is received, and the current vehicle front wheel angle of the vehicle under test is determined. According to the current vehicle front wheel angle and the current steering wheel angle, the ratio of the vehicle front wheel angle to the steering wheel angle is calculated, improving the accuracy of the test results.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-in-the-loop testing, and in particular, to a calibration system for geometric parameters of a vehicle under test in vehicle-in-the-loop testing. Background Art

[0002] Intelligent vehicles integrated with artificial intelligence technology are one of the main development directions of future vehicles. Compared with traditional vehicles, intelligent vehicles have more sensors, powerful data processing and analysis devices to perceive the surrounding environment and traffic conditions in real time, and make corresponding decisions and operations to assist or replace human drivers to complete vehicle driving work. Therefore, the safety and reliability of intelligent vehicles are the premise for ensuring road traffic safety. In order to ensure that intelligent vehicles can operate safely, reliably and efficiently under various traffic conditions, meteorological environments and usage scenarios on the road, a large number of test and trials are required during the R & D process of intelligent vehicles. Moreover, to ensure the performance of each mass-produced intelligent vehicle, relevant test and trials need to be carried out at different stages of the intelligent vehicle production process to provide quality assurance for the vehicles off the production line. In addition, for intelligent vehicles in service, the performance of some sensors or data processing devices decays with the extension of their service time. To ensure the driving safety and reliability of intelligent vehicles in service, it is necessary to regularly detect the performance of various parts of the vehicles in service.

[0003] According to different test stages and test environments, the test methods of intelligent vehicles can be divided into virtual simulation testing, natural scene testing, hardware-in-the-loop testing and vehicle-in-the-loop testing. Among them, vehicle-in-the-loop testing is a process of comprehensively testing and verifying the whole vehicle on a test bench by means of digital twin technology to simulate various real road environments, such as different road conditions, weather and traffic conditions including urban roads, highways, rural roads, etc. This kind of testing aims to evaluate the performance, safety and reliability of the vehicle under different road and traffic conditions. It is an important means to verify whether the design and engineering of the vehicle meet the relevant regulations and standard requirements. At the same time, the fuel economy, noise, vibration level, emission performance, etc. of the vehicle will also be examined in vehicle-in-the-loop testing. It can help manufacturers discover and solve potential problems to ensure the safety and reliability of the vehicle in various situations. To ensure that their vehicles can achieve high standards of performance and safety in the actual road environment.

[0004] At present, the drum test bench is one of the important devices for realizing vehicle-in-the-loop testing. Among them, the drum test bench with front-wheel steering function can rotate with the follow-up of the vehicle wheel angle, and can detect the autonomous driving function of intelligent vehicles. However, before carrying out the vehicle-in-the-loop test and trial, it is necessary to know in advance the kingpin distance of the front wheels of the vehicle under test and the ratio of the vehicle front-wheel angle to the steering wheel angle to adjust the front-wheel drum distance and set the relevant control parameters of the drum test bench.

[0005] Currently, in vehicle-in-the-loop testing, the kingpin offset of the front wheels and the ratio of the front-wheel steering angle to the steering-wheel angle of a vehicle can be obtained through the following two methods:

[0006] The first method: Refer to vehicle technical documents, such as vehicle manuals, design specifications, or engineering drawings, etc.;

[0007] The second method: Conduct on-site measurement, such as using a tape measure, an angle measuring instrument, etc.

[0008] However, due to the large number of vehicle brands and models, and many parameters being vehicle design parameters, vehicle technical documents are not publicly available to users and are difficult to obtain. On-site measurement is relatively inconvenient and prone to manual measurement errors, resulting in inaccurate parameters obtained, and further leading to inaccurate test results.

[0009] In addition, some parameters may evolve over time as the vehicle is in service. Using inaccurate parameters for vehicle-in-the-loop testing will result in inaccurate test results, or adjustment tests need to be carried out before vehicle-in-the-loop testing, affecting the test progress. Summary of the Invention

[0010] The present invention provides a calibration system for geometric parameters of a vehicle under test in vehicle-in-the-loop testing, which can improve the accuracy of test results and avoid affecting the test progress. The specific technical solution is as follows.

[0011] In a first aspect, the present invention provides a calibration system for geometric parameters of a vehicle under test in vehicle-in-the-loop testing, including a controller, two pit gantries, and a parameter calibrator disposed in a pit. The vehicle under test is placed directly above the pit. The two pit gantries are arranged in parallel. The controller is fixedly installed on the parameter calibrator, and the controller is communicatively connected to the electronic control unit ECU of the vehicle under test;

[0012] Each pit gantry includes a gantry support and a scissor lift. The scissor lift is fixedly installed at the rear side of the gantry support. The two gantry supports are respectively disposed directly below the two front wheels of the vehicle under test, and the parameter calibrator is disposed between the two gantry supports;

[0013] The parameter calibrator includes at least two servo cylinders and two industrial cameras. The two servo cylinders are both signal-connected to the controller, and each servo cylinder is drivingly connected to the corresponding industrial camera. The two scissor lifts lift the vehicle under test;

[0014] When the controller receives the alignment instruction, it controls the first servo cylinder to drive the corresponding industrial camera to move its camera center to align with the center of the left front wheel, and controls the second servo cylinder to drive the corresponding industrial camera to move its camera center to align with the center of the right front wheel. Herein, the first servo cylinder is the servo cylinder close to the left front wheel of the vehicle to be measured, and the second servo cylinder is the servo cylinder close to the right front wheel of the vehicle to be measured;

[0015] After the camera centers of the two industrial cameras have been aligned with the centers of the corresponding wheels, control the two industrial cameras to respectively capture the steering videos of the left front wheel and the right front wheel;

[0016] Receive the steering videos of the left front wheel and the right front wheel captured by the two industrial cameras, respectively perform tracking processing on the steering videos of the left front wheel and the right front wheel based on the target tracking algorithm to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation center and the second distance from the center of the right front wheel to the center of the right front wheel's rotation center, obtain the first extended distance of the industrial camera corresponding to the first servo cylinder and the second extended distance of the industrial camera corresponding to the second servo cylinder, and calculate the kingpin offset of the vehicle to be measured according to the first distance, the second distance, the first extended distance and the second extended distance. Herein, the extended distance is the distance between the center of each industrial camera and the installation position of the corresponding servo cylinder when the camera center of each industrial camera is aligned with the center of the corresponding wheel;

[0017] Receive the current steering wheel angle of the vehicle to be measured sent by the ECU, determine the current front wheel angle of the vehicle to be measured, and calculate the ratio of the front wheel angle to the steering wheel angle of the vehicle to be measured according to the current front wheel angle and the current steering wheel angle.

[0018] Optionally, the parameter calibrator further includes an electric cylinder bottom plate;

[0019] The two servo cylinders are respectively fixedly connected to the vertical center line on the upper surface of the electric cylinder bottom plate and are respectively located on both sides of the horizontal center line on the upper surface of the electric cylinder bottom plate. The two industrial cameras are respectively fixedly installed at the ends of the two servo cylinders far from the electric cylinder bottom plate.

[0020] Optionally, the parameter calibrator further includes a parameter calibrator bracket, a parameter calibrator bottom plate, a motor, two front-back moving guide rails, a slider, a turntable bottom plate and an electric rotating turntable. The motor and the electric rotating turntable are both signal-connected to the controller, and the controller is fixedly installed on the upper surface of the parameter calibrator bottom plate;

[0021] The parameter calibrator bracket is fixedly installed in the pit. The parameter calibrator bottom plate is fixedly connected above the parameter calibrator bracket. The two front-back moving guide rails are fixedly connected to the upper surface of the parameter calibrator bottom plate and are symmetrically arranged with respect to the vertical center line of the upper surface of the parameter calibrator bottom plate. The slider is slidably connected above the two front-back moving guide rails. The motor is fixedly connected to the upper surface of the parameter calibrator bottom plate, and the output end of the motor is drivingly connected to the slider;

[0022] The turntable bottom plate is fixedly connected above the slider. The electric rotary turntable is fixedly connected to the center position of the upper surface of the turntable bottom plate. The electric cylinder bottom plate is fixedly installed at the center position of the upper surface of the electric rotary turntable;

[0023] When the controller receives a movement instruction, it controls the motor to drive the slider to move a first target distance along the two front-back moving guide rails in a first target direction, where the movement instruction includes the first target direction and the first target distance;

[0024] When the controller receives a rotation instruction, it controls the electric rotary turntable to rotate a target angle in a second target direction, where the rotation instruction includes the second target direction and the target angle.

[0025] Optionally, when the controller receives an alignment instruction, it takes the moving distance of the industrial camera corresponding to the first servo cylinder included in the alignment instruction as the first current target distance, and takes the moving distance of the industrial camera corresponding to the second servo cylinder included in the alignment instruction as the second current target distance;

[0026] After controlling the first servo cylinder to drive the corresponding industrial camera to move the first current target distance, it controls the industrial camera corresponding to the first servo cylinder to take a picture of the left front wheel to obtain a first image, performs image processing on the first image to obtain a second target distance between the camera center of the industrial camera corresponding to the first servo cylinder and the center of the left front wheel, and determines whether the second target distance is 0. If not, it takes the second target distance as the first current target distance, and returns to execute the step of controlling the first servo cylinder to drive the corresponding industrial camera to move the first current target distance until the second target distance is 0. Here, the second target distance is the alignment distance. When the alignment distance is 0, the camera center of the industrial camera corresponding to the first servo cylinder is aligned with the center of the left front wheel;

[0027] After controlling the second servo cylinder to drive the corresponding industrial camera to move the second current target distance, control the industrial camera corresponding to the second servo cylinder to take a photo of the right front wheel to obtain a second image, perform image processing on the second image to obtain a third target distance between the camera center of the industrial camera corresponding to the second servo cylinder and the center of the right front wheel, and determine whether the third target distance is 0. If not, use the third target distance as the second current target distance, and return to execute the step of controlling the second servo cylinder to drive the corresponding industrial camera to move the second current target distance until the third target distance is 0. Wherein, the third target distance is the alignment distance, and when the alignment distance is 0, the camera center of the industrial camera corresponding to the second servo cylinder is aligned with the center of the right front wheel.

[0028] Optionally, for each video frame image in the steering video of the left front wheel, the controller performs a correlation value operation on the video frame image and a preset filtering template based on the correlation filtering method to obtain the coordinate of the motion trajectory point of the tracking target in the video frame image. Wherein, the preset filtering template is the filtering template that obtains the maximum response when acting on the tracking target in the video frame image, and the tracking target is any point on the wheel.

[0029] Fit all the coordinate of the motion trajectory points of the tracking target in the steering video of the left front wheel by the least square method to obtain a fitted circular curve.

[0030] Solve the circular curve to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation.

[0031] Optionally, the controller calculates the first sum between the first extension distance and the second extension distance, calculates the second sum between the first distance and the second distance, calculates the difference between the first sum and the second sum, and uses the difference as the kingpin inclination of the front wheels of the vehicle under test.

[0032] Optionally, the controller generates a trajectory point arc image based on all the motion trajectory points of the tracking target in the steering video of the left front wheel.

[0033] Determine the coordinates of the starting point, the ending point, and the center of the circle of the arc formed by all the motion trajectory points in the steering video of the left front wheel in the trajectory point arc image.

[0034] Calculate the first angle between the straight line formed by the center of the circle and the starting point and the horizontal axis of the trajectory point arc image.

[0035] Calculate the second angle between the straight line formed by the center of the circle and the ending point and the horizontal axis of the trajectory point arc image.

[0036] Calculate the current front-wheel steering angle of the vehicle under test based on the first included angle and the second included angle.

[0037] Optionally, the controller determines the current front-wheel steering angle of the vehicle under test based on a pre-established target deep learning neural network model and the current steering wheel angle, where the target deep learning neural network model is used to correlate the steering wheel angle sample data with the corresponding front-wheel steering angle sample data.

[0038] Optionally, the training process of the target deep learning neural network model is as follows:

[0039] Obtain the steering wheel angle sample data and the corresponding front-wheel steering angle sample data;

[0040] Use the steering wheel angle sample data and the corresponding front-wheel steering angle sample data as model training data to train an initial deep learning neural network model to obtain a target deep learning neural network model.

[0041] Optionally, the target deep learning neural network model is a backpropagation BP neural network model.

[0042] As can be seen from the above, a geometric parameter calibration system for a vehicle under test in a vehicle-in-the-loop test provided by an embodiment of the present invention includes a controller, two pit gantries and a parameter calibrator disposed in a pit. The vehicle under test is placed directly above the pit. The two pit gantries are arranged in parallel. The controller is fixedly installed on the parameter calibrator and is communicatively connected to the electronic control unit (ECU) of the vehicle under test. Each pit gantry includes a gantry bracket and a scissor lift. The scissor lift is fixedly installed at the rear side of the gantry bracket. The two gantry brackets are respectively disposed directly below the two front wheels of the vehicle under test. The parameter calibrator is disposed between the two gantry brackets. The parameter calibrator includes at least two servo cylinders and two industrial cameras. Both servo cylinders are signal-connected to the controller. Each servo cylinder is drivingly connected to the corresponding industrial camera. The two scissor lifts lift the vehicle under test. When the controller receives an alignment instruction, it controls the first servo cylinder to drive the corresponding industrial camera to move until the camera center of itself is aligned with the center of the left front wheel, and controls the second servo cylinder to drive the corresponding industrial camera to move until the camera center of itself is aligned with the center of the right front wheel. Herein, the first servo cylinder is the servo cylinder close to the left front wheel of the vehicle under test, and the second servo cylinder is the servo cylinder close to the right front wheel of the vehicle under test. After the camera centers of the two industrial cameras are aligned with the centers of the corresponding wheels, it controls the two industrial cameras to respectively capture the steering videos of the left front wheel and the right front wheel. It receives the steering videos of the left front wheel and the right front wheel captured by the two industrial cameras, and respectively performs tracking processing on the steering videos of the left front wheel and the right front wheel based on a target tracking algorithm to obtain a first distance from the center of the left front wheel to the center of the left front wheel's rotation center and a second distance from the center of the right front wheel to the center of the right front wheel's rotation center. It obtains a first extended distance of the industrial camera corresponding to the first servo cylinder and a second extended distance of the industrial camera corresponding to the second servo cylinder, and calculates the kingpin offset of the front wheels of the vehicle under test according to the first distance, the second distance, the first extended distance and the second extended distance. Herein, the extended distance is the distance between the center of each industrial camera and the installation position of the corresponding servo cylinder when the camera center of each industrial camera is aligned with the center of the corresponding wheel. It receives the current steering wheel angle of the vehicle under test sent by the ECU, determines the current front wheel angle of the vehicle under test, and calculates the ratio of the front wheel angle to the steering wheel angle of the vehicle under test according to the current front wheel angle and the current steering wheel angle.Thus, by setting up a controller, two pit gantries, and a parameter calibrator, the controller controls two industrial cameras of the parameter calibrator to capture the steering videos of the left front wheel and the right front wheel. Then, based on the object tracking algorithm, the steering videos of the left front wheel and the right front wheel are respectively tracked and processed to obtain the first distance from the center of the left front wheel to the center of rotation of the left front wheel and the second distance from the center of the right front wheel to the center of rotation of the right front wheel. Then, the first extended distance of the industrial camera corresponding to the first servo cylinder and the second extended distance of the industrial camera corresponding to the second servo cylinder are obtained. The kingpin offset of the vehicle under test is calculated based on the first distance, the second distance, the first extended distance, and the second extended distance. And the current steering wheel angle of the vehicle under test sent by the ECU is received to determine the current front wheel angle of the vehicle under test. The ratio of the front wheel angle to the steering wheel angle of the vehicle under test is calculated based on the current front wheel angle and the current steering wheel angle, achieving the purpose of automatically obtaining the kingpin offset and the ratio of the front wheel angle to the steering wheel angle in real time, without the need for on-site measurement, without manual measurement error, improving the accuracy of the obtained parameters, further improving the accuracy of the test results, and avoiding affecting the test progress.

[0043] The innovation points of the embodiments of the present invention include:

[0044] 1. By setting up a controller, two pit gantries, and a parameter calibrator, the controller controls two industrial cameras of the parameter calibrator to capture the steering videos of the left front wheel and the right front wheel. Then, based on the object tracking algorithm, the steering videos of the left front wheel and the right front wheel are respectively tracked and processed to obtain the first distance from the center of the left front wheel to the center of rotation of the left front wheel and the second distance from the center of the right front wheel to the center of rotation of the right front wheel. Then, the first extended distance of the industrial camera corresponding to the first servo cylinder and the second extended distance of the industrial camera corresponding to the second servo cylinder are obtained. The kingpin offset of the vehicle under test is calculated based on the first distance, the second distance, the first extended distance, and the second extended distance. And the current steering wheel angle of the vehicle under test sent by the ECU is received to determine the current front wheel angle of the vehicle under test. The ratio of the front wheel angle to the steering wheel angle of the vehicle under test is calculated based on the current front wheel angle and the current steering wheel angle, achieving the purpose of automatically obtaining the kingpin offset and the ratio of the front wheel angle to the steering wheel angle in real time, without the need for on-site measurement, without manual measurement error, improving the accuracy of the obtained parameters, further improving the accuracy of the test results, and avoiding affecting the test progress.

[0045] 2. Since the embodiments of the present invention can obtain the kingpin offset and the ratio of the front wheel angle to the steering wheel angle in real time, it greatly reduces the preparation time for the vehicle-in-the-loop test, improves the test efficiency, and also ensures the reliability of the test results.

[0046] 3. The controller controls two servo cylinders to drive the corresponding industrial cameras to move respectively so that the camera centers of the cameras themselves are aligned with the centers of the corresponding wheels through image processing.

[0047] 4. Perform correlation value operations on the steering video of the left front wheel through a correlation filtering method to obtain the coordinates of all moving trajectory points in the steering video of the left front wheel. Then, use the least squares method to fit the coordinates of all moving trajectory points of the tracking target in the steering video of the left front wheel to obtain a fitted circular curve, and solve the circular curve to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation center.

[0048] 5. Obtain the kingpin offset of the measured vehicle by calculating the first sum between the first extended distance and the second extended distance, calculating the second sum between the first distance and the second distance, calculating the difference between the first sum and the second sum, and using the difference as the kingpin offset of the measured vehicle.

[0049] 6. Obtain the current front wheel angle of the measured vehicle by calculating the central angle of the arc formed by all moving trajectory points in the steering video of the left front wheel.

[0050] 7. By fixing and connecting the two servo cylinders respectively to the vertical center line on the upper surface of the cylinder base plate and located on both sides of the horizontal center line on the upper surface of the cylinder base plate, the two servo cylinders are fixedly installed at the central position of the cylinder base plate, with balanced force, convenient installation, and good appearance.

[0051] 8. Determine the current front wheel angle of the measured vehicle based on a pre-established target deep learning neural network model and the current steering wheel angle.

[0052] 9. By setting up a motor, two front-back moving guide rails, and a slider, the control can control the slider to move back and forth, thereby driving two industrial motors to move back and forth. Also, by setting up an electric rotary table, the controller can control the electric rotary table to rotate, thereby driving two industrial cameras to rotate.

[0053] Of course, it is not necessary for any product or method implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0055] Figure 1Schematic diagram of a geometric parameter calibration system for a vehicle under test in a vehicle-in-the-loop test provided by an embodiment of the present invention;

[0056] Figure 2 Schematic diagram for calibrating geometric parameters of a vehicle under test;

[0057] Figure 3 Schematic diagram of a pit gantry provided by an embodiment of the present invention;

[0058] Figure 4 Schematic diagram of a parameter calibrator provided by an embodiment of the present invention;

[0059] Figure 5 Schematic diagram of wheel steering;

[0060] Figure 6(a) is a schematic diagram of the arc image of the trajectory points when the central angle of the arc is an acute angle;

[0061] Figure 6(b) is a schematic diagram of the arc image of the trajectory points when the central angle of the arc is an obtuse angle;

[0062] Figure 7 Schematic diagram of the non-linear function of the steering wheel angle and the front wheel angle of the vehicle.

[0063] Figures 1 - 7 In the figure, 1 is a pit, 2 is a pit gantry, 21 is a gantry bracket, 22 is a scissor lift, 3 is a parameter calibrator, 31 is a servo electric cylinder, 32 is an industrial camera, 33 is an electric cylinder bottom plate, 34 is a parameter calibrator bracket, 35 is a parameter calibrator bottom plate, 36 is a front and rear moving guide rail, 37 is a slider, 38 is an electric rotary turntable, 4 is a vehicle under test, r 1 The first distance, r 2 The second distance, d 1 The first extended distance, d 2 The second extended distance, U is the kingpin distance of the front wheels of the vehicle under test, β is the current front wheel angle of the vehicle, D is the starting point of the arc, E is the ending point of the arc, O is the center of the arc, M is the first included angle, N is the second included angle. Detailed implementation manners

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0065] It should be noted that the terms "including" and "having" and any variations thereof in the embodiments of the present invention and the accompanying drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0066] The embodiments of the present invention disclose a calibration system for geometric parameters of a vehicle under test in a vehicle-in-the-loop test, which can improve the accuracy of test results and avoid affecting the test progress. The embodiments of the present invention will be described in detail below.

[0067] Figure 1 It is a schematic structural diagram of a calibration system for geometric parameters of a vehicle under test in a vehicle-in-the-loop test provided by an embodiment of the present invention. Figure 2 It is a schematic diagram for calibrating the geometric parameters of the vehicle under test. Refer to Figure 1 and Figure 2 In the calibration system for geometric parameters of a vehicle under test in a vehicle-in-the-loop test provided by the embodiments of the present invention, it includes a controller and two pit gantries 2 and a parameter calibrator 3 arranged in the pit 1. Among them, the two pit gantries 2 and the parameter calibrator 3 can be fixed in the pit 1 through anchor bolts.

[0068] The vehicle under test 4 is placed directly above the pit 1, and the two pit gantries 2 are arranged in parallel. The controller is fixedly installed on the parameter calibrator 3. Among them, the fixed installation method can be any one of the existing technologies, and the embodiments of the present invention do not make any limitations in this regard.

[0069] The controller is communicatively connected to the ECU (Electronic Control Unit) of the vehicle under test 4. Among them, the controller can be a PLC (Programmable Logic Controller).

[0070] Figure 3 It is a schematic structural diagram of the pit gantry 2 provided by an embodiment of the present invention. Refer to Figure 3 Each pit gantry 2 includes a gantry bracket 21 and a scissor lift 22, and the scissor lift 22 is fixedly installed at the rear side of the gantry bracket 21. Among them, the fixed installation method can be bolt connection or any one of the existing technologies, and the embodiments of the present invention do not make any limitations in this regard.

[0071] Continue to refer to Figure 3, the truss bracket 21 can be a stepped structure. The front side of the truss bracket 21 is a high step, and the rear side of the truss bracket 21 is a low step. When the scissor lift 22 is not raised, the upper surface of the scissor lift 22 is flush with the upper surface of the front side of the truss bracket 21. And when the two scissor lifts 22 are not raised, their upper surfaces do not exceed the pit 1.

[0072] The front side of the truss bracket 21 can be a rectangular hollow structure, or it can be a structure with multiple through holes on the side, as long as it has a certain load-bearing capacity. Setting the front side of the truss bracket 21 as a hollow or perforated structure can reduce the weight. The rear side of the truss bracket 21 can be a groove structure.

[0073] Continue to refer to Figure 2 , the two truss brackets 21 are respectively arranged directly below the two front wheels of the vehicle under test 4, and the parameter calibrator 3 is arranged between the two truss brackets 21. Since the two truss brackets 21 are respectively arranged directly below the two front wheels of the vehicle under test 4, and the parameter calibrator 3 is arranged between the two truss brackets 21, therefore, the parameter calibrator 3 is very close to the two front wheels of the vehicle under test 4.

[0074] Figure 4 Schematic diagram of the structure of the parameter calibrator provided by the embodiment of the present invention, refer to Figure 4 , the parameter calibrator 3 at least includes two servo cylinders 31 and two industrial cameras 32. The two servo cylinders 31 are both signal-connected to the controller, and each servo cylinder 31 is drivingly connected to the corresponding industrial camera 32.

[0075] When parameter calibration is required, the tester controls the two scissor lifts 22 to rise through the controller of the scissor lift 22 to jack up the vehicle under test 4.

[0076] After the vehicle under test 4 is jacked up, parameter calibration can be carried out. The tester issues an alignment command to the controller. When the controller receives the alignment command, it controls the first servo cylinder to drive the corresponding industrial camera 32 to move until the camera center of itself is aligned with the center of the left front wheel, and controls the second servo cylinder to drive the corresponding industrial camera 32 to move until the camera center of itself is aligned with the center of the right front wheel. Among them, the first servo cylinder is the servo cylinder close to the left front wheel of the vehicle under test 4, and the second servo cylinder is the servo cylinder close to the right front wheel of the vehicle under test 4.

[0077] Among them, when the controller receives the alignment command, controlling the first servo cylinder to drive the corresponding industrial camera 32 to move until the camera center of itself is aligned with the center of the left front wheel, and controlling the second servo cylinder to drive the corresponding industrial camera 32 to move until the camera center of itself is aligned with the center of the right front wheel can be:

[0078] When the controller receives the alignment instruction, it takes the moving distance of the industrial camera 32 corresponding to the first servo cylinder included in the alignment instruction as the first current target distance, and takes the moving distance of the industrial camera 32 corresponding to the second servo cylinder included in the alignment instruction as the second current target distance;

[0079] After controlling the first servo cylinder to drive the corresponding industrial camera 32 to move the first current target distance, it controls the industrial camera 32 corresponding to the first servo cylinder to take a picture of the left front wheel to obtain a first image, performs image processing on the first image, obtains the second target distance between the camera center of the industrial camera 32 corresponding to the first servo cylinder and the center of the left front wheel, and determines whether the second target distance is 0. If not, it takes the second target distance as the first current target distance, and returns to execute the control of the first servo cylinder to drive the corresponding industrial camera 32 to move the first current target distance until the second target distance is 0. Among them, the second target distance is the alignment distance. When the alignment distance is 0, the camera center of the industrial camera 32 corresponding to the first servo cylinder is aligned with the center of the left front wheel;

[0080] After controlling the second servo cylinder to drive the corresponding industrial camera 32 to move the second current target distance, it controls the industrial camera 32 corresponding to the second servo cylinder to take a picture of the right front wheel to obtain a second image, performs image processing on the second image, obtains the third target distance between the camera center of the industrial camera 32 corresponding to the second servo cylinder and the center of the right front wheel, and determines whether the third target distance is 0. If not, it takes the third target distance as the second current target distance, and returns to execute the control of the second servo cylinder to drive the corresponding industrial camera 32 to move the second current target distance until the third target distance is 0. Among them, the third target distance is the alignment distance. When the alignment distance is 0, the camera center of the industrial camera 32 corresponding to the second servo cylinder is aligned with the center of the right front wheel.

[0081] The alignment instruction issued by the tester includes the moving distances of the two industrial cameras 32. After the controller receives the alignment instruction, it controls the two servo cylinders 31 to drive the corresponding industrial cameras 32 to move the corresponding moving distances respectively.

[0082] However, after the two industrial cameras 32 complete the movement, it is possible that the camera centers of the two industrial cameras 32 are not aligned with the centers of the corresponding front wheels. Therefore, it is necessary to take pictures of each corresponding front wheel through the two industrial cameras 32, and then perform image processing on the taken images to determine whether the camera centers of the two industrial cameras 32 are aligned with the centers of the corresponding front wheels. If not aligned, the controller needs to control the two servo cylinders 31 again to drive the corresponding industrial cameras 32 to move until the camera centers of the two industrial cameras 32 are both aligned with the centers of the corresponding front wheels. Among them, the image processing method can be any one of the existing technologies that can calculate the distance between the photographed camera and the photographed object. The embodiments of the present invention do not make any limitations on this. Among them, the distance between the photographed camera and the photographed object includes the alignment distance.

[0083] Thus, the controller controls the two servo cylinders 31 to drive the corresponding industrial cameras 32 to move until the camera centers of themselves are aligned with the centers of the corresponding wheels through image processing.

[0084] After the camera centers of the two industrial cameras 32 are aligned with the centers of the corresponding wheels, the controller controls the two industrial cameras 32 to respectively shoot the steering videos of the left front wheel and the right front wheel. Specifically, after the camera centers of the two industrial cameras 32 are aligned with the centers of the corresponding wheels, the tester controls the vehicle 4 to be tested to steer.

[0085] The two industrial cameras 32 respectively shoot the steering videos of the left front wheel and the right front wheel under the control of the controller and send them to the controller. Among them, the industrial camera 32 near the left front wheel shoots the steering video of the left front wheel, and the industrial camera 32 near the right front wheel shoots the steering video of the right front wheel. The steering videos of each wheel can be videos that first turn left and then turn right, or videos that first turn right and then turn left.

[0086] The controller receives the steering videos of the left front wheel and the right front wheel shot by the two industrial cameras 32, and respectively performs tracking processing on the steering videos of the left front wheel and the right front wheel based on the target tracking algorithm to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation and the second distance from the center of the right front wheel to the center of the right front wheel's rotation. Among them, the wheel rotation center is the center of the circular arc trajectory formed by a certain marked point on the wheel when tracking the marked point, and the marked point is the tracking target.

[0087] Among them, the determination method of the first distance can be:

[0088] For each video frame image in the steering video of the left front wheel, the controller performs a correlation value operation on the video frame image and a preset filtering template based on the correlation filtering method to obtain the coordinates of the movement trajectory points of the tracking target in the video frame image. Here, the preset filtering template is the filtering template that obtains the maximum response when acting on the tracking target in the video frame image, and the tracking target is any point on the wheel.

[0089] The coordinates of all the movement trajectory points of the tracking target in the steering video of the left front wheel are fitted by the least squares method to obtain a fitted circular curve.

[0090] The circular curve is solved to obtain the first distance from the center of the left front wheel to the center of rotation of the left front wheel.

[0091] The correlation filtering method is an object tracking algorithm. Compared with other object tracking algorithms, the greatest advantage of the correlation filtering method is its extremely fast operation speed and high accuracy. The correlation value is a measure of the similarity between two signals. If two signals are more similar, then their correlation value is higher. In the embodiments of the present invention, the two signals refer to each video frame image in the steering video and the preset filtering template h.

[0092] Here, the preset filtering template h is the filtering template that obtains the maximum response g when acting on the tracking target in the video frame image. The tracking target is any point on the wheel, and a point that is easy to be photographed by the industrial camera 32 can be selected. Here, the response g appears as a two-dimensional Gaussian peak centered on the tracking target in the video frame image f.

[0093] For each video frame image in the steering video of the left front wheel, the controller performs a correlation value operation on the video frame image and a preset filtering template based on the correlation filtering method to obtain the coordinates of the movement trajectory points of the tracking target in the video frame image, which can be:

[0094] For each video frame image in the steering video of the left front wheel, the controller respectively calculates the two-dimensional Fourier transform of the video frame image f: F = F(f), the two-dimensional Fourier transform of the preset filtering template h: H = F(h), and the two-dimensional Fourier transform of the response g: G = F(g).

[0095] According to the convolution theorem, the correlation value operation becomes the multiplication of elements in the Fourier domain. The symbol ⊙ represents the multiplication of elements, and the symbol * represents the complex conjugate. Then the form of the correlation value is:

[0096] G = F⊙H * (1)

[0097] Here, G is the two-dimensional Fourier transform of the response g, F is the two-dimensional Fourier transform of the video frame image f, and H *is the complex conjugate of the two-dimensional Fourier transform of the preset filter template h.

[0098] Adjusting the above formula (1), we have:

[0099]

[0100] In practical applications, due to factors such as the appearance transformation of the tracking target, it is necessary to consider m consecutive video frame images with the tracking target in the steering video of the left front wheel as references simultaneously to improve the robustness of the model. Then, there is the following objective function:

[0101]

[0102] where F i is the two-dimensional Fourier transform of the i-th video frame image in the steering video of the left front wheel, and G i is the two-dimensional Fourier transform of the response g of the i-th video frame image in the steering video of the left front wheel.

[0103] According to the convolution theorem, operations in the frequency domain are all element-wise. Therefore, each element in H * can be solved separately

[0104]

[0105] where is the complex conjugate of the two-dimensional Fourier transform of the element with abscissa w and ordinate v in the preset filter template h, F w，v，i is the two-dimensional Fourier transform of the element with abscissa w and ordinate v in the i-th video frame image in the steering video of the left front wheel, and G w，v，i is the two-dimensional Fourier transform of the response of the element with abscissa w and ordinate v in the i-th video frame image in the steering video of the left front wheel.

[0106] Taking the derivative of the above formula (4) and setting it to 0, we can solve to obtain:

[0107]

[0108] where is the complex conjugate of the two-dimensional Fourier transform of the i-th video frame image in the steering video of the left front wheel, is the complex conjugate of the two-dimensional Fourier transform of the response g of the i-th video frame image in the steering video of the left front wheel.

[0109] Obtaining H means obtaining H *, during the process of tracking a tracking target, it is only necessary to perform the correlation value operation shown in formula (1) on the preset filtering template h and each video frame image in the steering video of the left front wheel, and use the coordinates of the point with the largest value in the two-dimensional Fourier transform G of the obtained response g as the coordinates of the motion trajectory point of the tracking target in each video frame image.

[0110] Among them, the update method of the preset filtering template h can be carried out in the following way:

[0111] H t =(1 - η)H t-1 +ηH(t) (6)

[0112] Among them, H t is the two-dimensional Fourier transform of the preset filtering template h in the t-th video frame image, η is an empirical constant between 0 and 1, H t-1 is the two-dimensional Fourier transform of the preset filtering template h in the (t - 1)-th video frame image, and H(t) is the filtering template obtained in the t-th video frame image.

[0113] Thus, through the above-mentioned correlation filtering method, the coordinates of the motion trajectory points of the tracking target in each video frame image are obtained, and a series of motion trajectory point coordinates (x, y) of the tracking target on the left front wheel are obtained.

[0114] Then, the least squares method is used to fit all the motion trajectory point coordinates of the tracking target in the steering video of the left front wheel to obtain the fitted circular curve:

[0115] R 2 =(x - A) 2 +(y - B) 2 (7)

[0116] Among them, R is the radius of the circular curve and also the first distance from the center of the left front wheel to the center of rotation of the left front wheel, A is the abscissa of the center of the circular curve, B is the ordinate of the center of the circular curve, x is the abscissa of the motion trajectory point of the tracking target on the left front wheel, and y is the ordinate of the motion trajectory point of the tracking target on the left front wheel.

[0117] Let a = -2A, b = -2B, c = A 2 +B 2 -R 2 , and another form of the circular curve equation can be obtained:

[0118] x 2 +y 2 +ax + by + c = 0 (8)

[0119] Substitute the coordinates of all the motion trajectory points of the tracking target in the steering video of the left front wheel into the above formula (8) to solve the circular curve, and obtain the radius R of the circular curve, which is the first distance r from the center of the left front wheel to the center of rotation of the left front wheel. 1 :

[0120]

[0121] Thus, perform a correlation value operation on the steering video of the left front wheel through the correlation filtering method to obtain the coordinates of all the motion trajectory points in the steering video of the left front wheel. Then, fit the coordinates of all the motion trajectory points of the tracking target in the steering video of the left front wheel by the least squares method to obtain the fitted circular curve, and solve the circular curve to obtain the first distance from the center of the left front wheel to the center of rotation of the left front wheel.

[0122] Similarly, the determination method of the second distance can be:

[0123] For each video frame image in the steering video of the right front wheel, the controller performs a correlation value operation on the video frame image and a preset filtering template based on the correlation filtering method to obtain the coordinates of the motion trajectory points of the tracking target in the video frame image. Among them, the preset filtering template is the filtering template that obtains the maximum response when acting on the tracking target in the video frame image, and the tracking target is any point on the wheel;

[0124] Fit the coordinates of all the motion trajectory points of the tracking target in the steering video of the right front wheel by the least squares method to obtain the fitted circular curve;

[0125] Solve the circular curve to obtain the second distance from the center of the right front wheel to the center of rotation of the right front wheel.

[0126] The determination method of the second distance is the same as that of the first distance. For specific details, please refer to the above determination method of the first distance, which will not be elaborated here.

[0127] After obtaining the first distance and the second distance, the controller acquires the first extension distance of the industrial camera 32 corresponding to the first servo cylinder and the second extension distance of the industrial camera 32 corresponding to the second servo cylinder. Among them, the extension distance is the distance between the center of each industrial camera 32 and the installation position of the corresponding servo cylinder when the camera center of each industrial camera 32 is aligned with the center of the corresponding wheel.

[0128] Since the controller controls the camera centers of the two industrial cameras 32 to be aligned with the centers of the corresponding wheels, the controller can acquire the distance between the center of each industrial camera 32 and the installation position of the corresponding servo cylinder when the camera center of each industrial camera 32 is aligned with the center of the corresponding wheel, that is, the first extension distance and the second extension distance.

[0129] Figure 5 It is a schematic diagram of wheel steering, Figure 5 where r 1 is the first distance, r 2 is the second distance, d 1 is the first extension distance, d 2 is the second extension distance, U is the kingpin distance of the front wheels of the vehicle 4 to be measured, and β is the steering angle of the front wheels of the vehicle.

[0130] After obtaining the first extension distance and the second extension distance, the controller calculates the kingpin distance of the front wheels of the vehicle 4 to be measured based on the first distance, the second distance, the first extension distance, and the second extension distance.

[0131] Among them, the kingpin distance of the front wheels of the vehicle 4 to be measured calculated by the controller based on the first distance, the second distance, the first extension distance, and the second extension distance can be:

[0132] The controller calculates the first sum between the first extension distance and the second extension distance, calculates the second sum between the first distance and the second distance, calculates the difference between the first sum and the second sum, and uses the difference as the kingpin distance of the front wheels of the vehicle 4 to be measured.

[0133] The kingpin distance of the front wheels of the vehicle 4 to be measured is calculated according to the following formula:

[0134] U = (d 1 + d 2 ) - (r 1 + r 2 )

[0135] Among them, U is the kingpin distance of the front wheels of the vehicle 4 to be measured, r 1 is the first distance, r 2 is the second distance, d 1 is the first extension distance, d 2 is the second extension distance.

[0136] Thus, the kingpin distance of the front wheels of the vehicle 4 to be measured is obtained by calculating the first sum between the first extension distance and the second extension distance, calculating the second sum between the first distance and the second distance, calculating the difference between the first sum and the second sum, and using the difference as the kingpin distance of the front wheels of the vehicle 4 to be measured.

[0137] Since the controller is communicatively connected to the ECU, the controller can receive the current steering wheel angle of the vehicle 4 to be measured sent by the ECU, determine the current steering angle of the front wheels of the vehicle 4 to be measured, and calculate the ratio of the steering angle of the front wheels of the vehicle 4 to be measured to the steering wheel angle based on the current steering angle of the front wheels of the vehicle and the current steering wheel angle.

[0138] Among them, there are multiple ways to determine the current steering angle of the front wheels of the vehicle, including but not limited to the following two ways:

[0139] The first method:

[0140] When the controller obtains the coordinates of all the motion trajectory points in the steering video of the left front wheel, the controller generates a trajectory point arc image based on all the motion trajectory points of the tracking target in the steering video of the left front wheel;

[0141] Determine the coordinates of the starting point, the ending point, and the center of the circle of the arc formed by all the motion trajectory points in the steering video of the left front wheel in the trajectory point arc image;

[0142] Calculate the first angle between the straight line formed by the center of the circle and the starting point and the horizontal coordinate axis of the trajectory point arc image;

[0143] Calculate the second angle between the straight line formed by the center of the circle and the ending point and the horizontal coordinate axis of the trajectory point arc image;

[0144] Calculate the current vehicle front wheel angle of the vehicle under test 4 based on the first angle and the second angle.

[0145] Since the vehicle front wheel angles of the two front wheels are the same, therefore, the vehicle front wheel angle can be calculated based on the steering video of one side of the front wheel. In the embodiment of the present invention, the calculation is based on the steering video of the left front wheel. Of course, the calculation can also be based on the steering video of the right front wheel.

[0146] FIG. 6(a) is a schematic diagram of a trajectory point arc image when the central angle of the arc is an acute angle, and FIG. 6(b) is a schematic diagram of a trajectory point arc image when the central angle of the arc is an obtuse angle. Referring to FIGS. 6(a) and 6(b), when the controller obtains the coordinates of all the motion trajectory points in the steering video of the left front wheel, the controller generates a trajectory point arc image based on all the motion trajectory points of the tracking target in the steering video of the left front wheel. The method of generating an image through the motion trajectory points can be any one of the existing image generation methods, and the embodiment of the present invention does not make any limitation thereto.

[0147] As shown in FIGS. 6(a) and 6(b), all the motion trajectory points of the tracking target in the steering video of the left front wheel present as an arc. Determine the coordinates of the starting point, the ending point, and the center of the circle of the arc formed by all the motion trajectory points in the steering video of the left front wheel in the trajectory point arc image, that is, denote the starting point of the arc as D(x 1 , y 1 ), the ending point as E(x 2 , y 2 ), and the center of the circle as O(x 0 , y 0 ). Among them, the central angle of the arc formed by all the motion trajectory points in the steering video of the left front wheel is the current vehicle front wheel angle of the vehicle under test 4.

[0148] When the central angle of the arc is an acute angle, calculate the first angle M between the straight line OD formed by the center and the starting point and the horizontal coordinate axis of the arc image of the trajectory point according to the following formula (10), and calculate the second angle N between the straight line OE formed by the center and the ending point and the horizontal coordinate axis of the arc image of the trajectory point:

[0149]

[0150] When the central angle of the arc is an acute angle, the current front wheel steering angle of the vehicle under test 4 can be calculated based on the first angle and the second angle: calculate the difference between the first angle and the second angle as the current front wheel steering angle of the vehicle under test 4:

[0151] β = M - N (11)

[0152] When the central angle of the arc is an obtuse angle, calculate the first angle M between the straight line OD formed by the center and the starting point and the horizontal coordinate axis of the arc image of the trajectory point according to the following formula (12), and calculate the second angle N between the straight line OE formed by the center and the ending point and the horizontal coordinate axis of the arc image of the trajectory point:

[0153]

[0154] When the central angle of the arc is an obtuse angle, the current front wheel steering angle of the vehicle under test 4 can be calculated based on the first angle and the second angle: calculate the sum of the first angle and the second angle as the current front wheel steering angle of the vehicle under test 4:

[0155] β = M + N (13)

[0156] Thus, the current front wheel steering angle of the vehicle under test 4 is obtained by calculating the central angle of the arc formed by all the movement trajectory points in the steering video of the left front wheel.

[0157] The second method:

[0158] The controller determines the current front wheel steering angle of the vehicle under test 4 based on a pre-established target deep learning neural network model and the current steering wheel angle, where the target deep learning neural network model is used to correlate the steering wheel angle sample data with the corresponding front wheel steering angle sample data.

[0159] Among them, the training process of the target deep learning neural network model can be:

[0160] Obtain the steering wheel angle sample data and the corresponding front wheel steering angle sample data;

[0161] Use the steering wheel angle sample data and the corresponding vehicle front wheel angle sample data as model training data to train the initial deep learning neural network model to obtain the target deep learning neural network model.

[0162] When establishing the target deep learning neural network model, it is necessary to obtain the steering wheel angle sample data and the corresponding vehicle front wheel angle sample data.

[0163] Among them, the acquisition method of the steering wheel angle sample data and the corresponding vehicle front wheel angle sample data can be:

[0164] Turn the steering wheel multiple times to obtain multiple steering wheel angles and the corresponding vehicle front wheel angles, and then use the multiple steering wheel angles as the steering wheel angle sample data, and the vehicle front wheel angles corresponding to the multiple steering wheel angles as the vehicle front wheel angle sample data.

[0165] After obtaining the steering wheel angle sample data and the corresponding vehicle front wheel angle sample data, the steering wheel angle sample data and the corresponding vehicle front wheel angle sample data can be used as model training data to train the initial deep learning neural network model to obtain the target deep learning neural network model.

[0166] Among them, the condition for the end of model training can be that the required number of model iterations is reached or the error of the model is less than the required error threshold. The relationship between the steering wheel angle and the vehicle front wheel angle is a non-linear function relationship β = F(α), where α is the steering wheel angle, as Figure 7 shown Figure 7 is a schematic diagram of the non-linear function of the steering wheel angle and the vehicle front wheel angle.

[0167] Exemplarily, the target deep learning neural network model can be a BP (Back Propagation Neural Network) neural network model.

[0168] Thus, based on the pre-established target deep learning neural network model and the current steering wheel angle, the current vehicle front wheel angle of the vehicle under test 4 is determined.

[0169] Continue to refer to Figure 4 , the parameter calibrator 3 further includes an electric cylinder bottom plate 33. The two servo electric cylinders 31 are respectively fixedly connected to the vertical center line on the upper surface of the electric cylinder bottom plate 33 and are respectively located on both sides of the horizontal center line of the upper surface of the electric cylinder bottom plate 33. The two industrial cameras 32 are respectively fixedly installed at the ends of the two servo electric cylinders 31 away from the electric cylinder bottom plate 33. Among them, the fixed installation method can be bolt connection.

[0170] By fixedly connecting two servo cylinders 31 to the vertical center line on the upper surface of the cylinder base plate 33 respectively, and being located on both sides of the horizontal center line of the upper surface of the cylinder base plate 33 respectively, the two servo cylinders 31 are fixedly installed at the central position of the cylinder base plate 33, with balanced force, convenient installation and good appearance.

[0171] Continue to refer to Figure 4 When the parameter calibrator 3 further includes the cylinder base plate 33, the parameter calibrator 3 further includes a parameter calibrator bracket 34, a parameter calibrator base plate 35, a motor, two front and rear moving guide rails 36, a slider 37, a turntable base plate and an electric rotary turntable 38. The motor and the electric rotary turntable 38 are both connected to the controller in a signal manner. The controller is fixedly installed on the upper surface of the parameter calibrator base plate 35. Among them, the fixed installation method can be any one of the existing technical fixed installation methods, and the embodiments of the present invention do not make any limitations thereto.

[0172] The parameter calibrator bracket 34 is fixedly installed in the pit 1. The parameter calibrator base plate 35 is fixedly connected above the parameter calibrator bracket 34. The two front and rear moving guide rails 36 are fixedly connected to the upper surface of the parameter calibrator base plate 35 and are symmetrically arranged about the vertical center line of the upper surface of the parameter calibrator base plate 35. The slider 37 is slidably connected above the two front and rear moving guide rails 36. The motor is fixedly connected to the upper surface of the parameter calibrator base plate 35 and the output end of the motor is drivingly connected to the slider 37. Among them, the parameter calibrator bracket 34 can be fixedly installed in the pit 1 through anchor bolts. The fixed connection methods of the parameter calibrator base plate 35, the two front and rear moving guide rails 36 and the motor can be any one of the existing technical fixed connection methods, and the embodiments of the present invention do not make any limitations thereto.

[0173] By fixedly connecting the two front and rear moving guide rails 36 to the upper surface of the parameter calibrator base plate 35 and being symmetrically arranged about the vertical center line of the upper surface of the parameter calibrator base plate 35, the two front and rear moving guide rails 36 are fixedly installed at the central position of the parameter calibrator base plate 35, with balanced force, convenient installation and good appearance.

[0174] In one implementation, the parameter calibrator bracket 34 and the parameter calibrator base plate 35 can be integrally formed. In the embodiments of the present invention, splitting the two into two components can reduce the weight and processing cost.

[0175] The turntable bottom plate is fixedly connected above the slider 37, and the electric rotary turntable 38 is fixedly connected to the center position of the upper surface of the turntable bottom plate, with balanced force, convenient installation and good appearance. The electric cylinder bottom plate 33 is fixedly installed at the center position of the upper surface of the electric rotary turntable 38, with balanced force, convenient installation and good appearance. Among them, the fixed connection methods of the turntable bottom plate, the electric rotary turntable 38 and the electric cylinder bottom plate 33 can be any one of the fixed connection methods in the prior art, and the embodiments of the present invention do not make any limitations on this.

[0176] Sometimes, the position where the measured vehicle 4 stops is deviated. Therefore, the tester needs to send a moving instruction or a rotation instruction to the controller.

[0177] When the controller receives the moving instruction, it controls the motor to drive the slider 37 to move a first target distance along the two front and rear moving guide rails 36 in the first target direction. Among them, the moving instruction includes the first target direction and the first target distance.

[0178] Exemplarily, the first target direction is forward or backward.

[0179] When the controller receives the rotation instruction, it controls the electric rotary turntable 39 to rotate a target angle in the second target direction. Among them, the rotation instruction includes the second target direction and the target angle.

[0180] Exemplarily, the second target direction is the clockwise direction or the counterclockwise direction.

[0181] Thus, by setting the motor, the two front and rear moving guide rails 36 and the slider 37, the controller can control the slider 37 to move back and forth, thereby driving the two industrial motors 32 to move back and forth. And, by setting the electric rotary turntable 39, the controller can control the electric rotary turntable 39 to rotate, thereby driving the two industrial cameras 32 to rotate.

[0182] As can be seen from the above, a geometric parameter calibration system for a vehicle under test in a vehicle-in-the-loop test provided by an embodiment of the present invention includes a controller, two pit gantries and a parameter calibrator disposed in a pit. The vehicle under test is placed directly above the pit. The two pit gantries are arranged in parallel. The controller is fixedly installed on the parameter calibrator and is communicatively connected to the electronic control unit (ECU) of the vehicle under test. Each pit gantry includes a gantry support and a scissor lift. The scissor lift is fixedly installed at the rear side of the gantry support. The two gantry supports are respectively disposed directly below the two front wheels of the vehicle under test. The parameter calibrator is disposed between the two gantry supports. The parameter calibrator includes at least two servo cylinders and two industrial cameras. The two servo cylinders are both signal-connected to the controller. Each servo cylinder is drivingly connected to the corresponding industrial camera. The two scissor lifts lift the vehicle under test. When the controller receives an alignment instruction, it controls the first servo cylinder to drive the corresponding industrial camera to move until the camera center of itself is aligned with the center of the left front wheel, and controls the second servo cylinder to drive the corresponding industrial camera to move until the camera center of itself is aligned with the center of the right front wheel. Herein, the first servo cylinder is the servo cylinder close to the left front wheel of the vehicle under test, and the second servo cylinder is the servo cylinder close to the right front wheel of the vehicle under test. After the camera centers of the two industrial cameras are aligned with the centers of the corresponding wheels, it controls the two industrial cameras to respectively capture the steering videos of the left front wheel and the right front wheel. It receives the steering videos of the left front wheel and the right front wheel captured by the two industrial cameras, and respectively performs tracking processing on the steering videos of the left front wheel and the right front wheel based on a target tracking algorithm to obtain a first distance from the center of the left front wheel to the center of the left front wheel's rotation center and a second distance from the center of the right front wheel to the center of the right front wheel's rotation center. It obtains the first extended distance of the industrial camera corresponding to the first servo cylinder and the second extended distance of the industrial camera corresponding to the second servo cylinder, and calculates the kingpin offset of the front wheels of the vehicle under test according to the first distance, the second distance, the first extended distance and the second extended distance. Herein, the extended distance is the distance between the center of each industrial camera and the installation position of the corresponding servo cylinder when the camera center of each industrial camera is aligned with the center of the corresponding wheel. It receives the current steering wheel angle of the vehicle under test sent by the ECU, determines the current front wheel angle of the vehicle under test, and calculates the ratio of the front wheel angle to the steering wheel angle of the vehicle under test according to the current front wheel angle and the current steering wheel angle.Thus, by setting up a controller, two pit gantries, and a parameter calibrator, the controller controls two industrial cameras of the parameter calibrator to capture the steering videos of the left front wheel and the right front wheel. Then, based on the target tracking algorithm, the steering videos of the left front wheel and the right front wheel are respectively tracked to obtain the first distance from the center of the left front wheel to the center of the left front wheel's rotation and the second distance from the center of the right front wheel to the center of the right front wheel's rotation. Then, the first extended distance of the industrial camera corresponding to the first servo cylinder and the second extended distance of the industrial camera corresponding to the second servo cylinder are obtained. The kingpin offset of the measured vehicle is calculated based on the first distance, the second distance, the first extended distance, and the second extended distance. And the current steering wheel angle of the measured vehicle sent by the ECU is received to determine the current front wheel angle of the measured vehicle. The ratio of the front wheel angle to the steering wheel angle of the measured vehicle is calculated based on the current front wheel angle and the current steering wheel angle, achieving the purpose of automatically obtaining the kingpin offset and the ratio of the front wheel angle to the steering wheel angle in real time without on-site measurement, without manual measurement errors, improving the accuracy of the obtained parameters, further improving the accuracy of the test results, and avoiding affecting the test progress.

[0183] Moreover, since the embodiments of the present invention can obtain the kingpin offset and the ratio of the front wheel angle to the steering wheel angle in real time, it greatly reduces the preparation time for the vehicle-in-the-loop test, improves the test efficiency, and also ensures the reliability of the test results.

[0184] Those of ordinary skill in the art can understand that the drawings are only schematic diagrams of one embodiment, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0185] Those of ordinary skill in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment as described in the embodiment, or can be correspondingly changed to be located in one or more devices different from this embodiment. The modules in the above embodiments can be combined into one module, or further split into multiple sub-modules.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle-in-the-loop test vehicle geometric parameter calibration system, characterized in that: It includes a controller, two pit racks and a parameter calibrator arranged in a pit, the vehicle under test is placed directly above the pit, the two pit racks are arranged in parallel, the controller is fixedly mounted on the parameter calibrator, and the controller is communicatively connected with an electronic control unit ECU of the vehicle under test; Each pit frame includes a frame support and a scissor lift, the scissor lift is fixedly installed on the rear side of the frame support, the two frame supports are respectively arranged directly below the two front wheels of the vehicle under test, and the parameter calibrator is arranged between the two frame supports; The parameter calibration instrument comprises at least two servo electric cylinders and two industrial cameras, the two servo electric cylinders are both connected to the controller signal, each servo electric cylinder is connected to the corresponding industrial camera drive, and the two scissor lifts support the vehicle under test; When receiving the alignment instruction, the controller controls the first servo electric cylinder to drive the corresponding industrial camera to move until the center of its camera is aligned with the center of the left front wheel, and controls the second servo electric cylinder to drive the corresponding industrial camera to move until the center of its camera is aligned with the center of the right front wheel, wherein the first servo electric cylinder is a servo electric cylinder close to the left front wheel of the vehicle under test, and the second servo electric cylinder is a servo electric cylinder close to the right front wheel of the vehicle under test; After the camera centers of the two industrial cameras have been aligned with the centers of the corresponding wheels, controlling the two industrial cameras to respectively shoot the steering video of the left front wheel and the steering video of the right front wheel; Receive the steering video of the left front wheel and the steering video of the right front wheel taken by the two industrial cameras, track and process the steering video of the left front wheel and the steering video of the right front wheel respectively based on the target tracking algorithm, obtain a first distance from the center of the left front wheel to the rotation center of the left front wheel and a second distance from the center of the right front wheel to the rotation center of the right front wheel, obtain a first extension distance of the industrial camera corresponding to the first servo electric cylinder and a second extension distance of the industrial camera corresponding to the second servo electric cylinder, and calculate the front wheel kingpin spacing of the vehicle under test according to the first distance, the second distance, the first extension distance and the second extension distance, wherein the extension distance is the distance between the center of each industrial camera and the installation position of the corresponding servo electric cylinder when the camera center of each industrial camera is aligned with the center of the corresponding wheel; Receive the current steering wheel angle of the tested vehicle sent by the ECU, determine the current front wheel angle of the tested vehicle, and calculate the ratio of the front wheel angle of the tested vehicle to the steering wheel angle based on the current front wheel angle and the current steering wheel angle.

2. The vehicle-in-the-loop test vehicle geometric parameter calibration system according to claim 1, characterized in that: The parameter calibration instrument also includes an electric cylinder bottom plate; The two servo electric cylinders are respectively fixedly connected to the vertical center line of the upper surface of the electric cylinder base plate, and are respectively located on both sides of the horizontal center line of the upper surface of the electric cylinder base plate. The two industrial cameras are respectively fixedly installed on the ends of the two servo electric cylinders away from the electric cylinder base plate.

3. The vehicle-in-the-loop test vehicle geometric parameter calibration system as claimed in claim 2, characterized in that: The parameter calibration instrument further includes a parameter calibration instrument bracket, a parameter calibration instrument bottom plate, a motor, two forward and backward movable guide rails, a slider, a turntable bottom plate and an electric rotating turntable, the motor and the electric rotating turntable are both connected to the controller signal, and the controller is fixedly mounted on the upper surface of the parameter calibration instrument bottom plate; The parameter calibration instrument bracket is fixedly installed in the pit, the parameter calibration instrument bottom plate is fixedly connected to the top of the parameter calibration instrument bracket, the two forward and backward movable guide rails are fixedly connected to the upper surface of the parameter calibration instrument bottom plate and are symmetrically arranged about the vertical center line of the upper surface of the parameter calibration instrument bottom plate, the slider is slidably connected to the top of the two forward and backward movable guide rails, the motor is fixedly connected to the upper surface of the parameter calibration instrument bottom plate and the output end of the motor is drivingly connected to the slider; The turntable bottom plate is fixedly connected to the top of the slider, the electric rotary turntable is fixedly connected to the center position of the upper surface of the turntable bottom plate, and the electric cylinder bottom plate is fixedly installed at the center position of the upper surface of the electric rotary turntable; When receiving a movement instruction, the controller controls the motor to drive the slider to move a first target distance along the two front and rear moving guide rails in a first target direction, wherein the movement instruction includes the first target direction and the first target distance; When receiving a rotation instruction, the controller controls the electric rotary turntable to rotate toward a second target direction by a target angle, wherein the rotation instruction includes the second target direction and the target angle.

4. The vehicle-in-the-loop test vehicle geometric parameter calibration system according to claim 1, characterized in that: When receiving the alignment instruction, the controller uses the moving distance of the industrial camera corresponding to the first servo electric cylinder contained in the alignment instruction as the first current target distance, and uses the moving distance of the industrial camera corresponding to the second servo electric cylinder contained in the alignment instruction as the second current target distance; After controlling the first servo electric cylinder to drive the corresponding industrial camera to move the first current target distance, controlling the industrial camera corresponding to the first servo electric cylinder to take a picture of the left front wheel to obtain a first image, performing image processing on the first image to obtain a second target distance between the camera center of the industrial camera corresponding to the first servo electric cylinder and the center of the left front wheel, judging whether the second target distance is 0, if not, taking the second target distance as the first current target distance, returning to execute the control of the first servo electric cylinder to drive the corresponding industrial camera to move the first current target distance until the second target distance is 0, wherein the second target distance is an alignment distance, and when the alignment distance is 0, the camera center of the industrial camera corresponding to the first servo electric cylinder is aligned with the center of the left front wheel; After controlling the second servo electric cylinder to drive the corresponding industrial camera to move the second current target distance, the industrial camera corresponding to the second servo electric cylinder is controlled to take a picture of the right front wheel to obtain a second image, and the second image is processed to obtain a third target distance between the camera center of the industrial camera corresponding to the second servo electric cylinder and the center of the right front wheel, and it is determined whether the third target distance is 0. If not, the third target distance is used as the second current target distance, and the control of the second servo electric cylinder to drive the corresponding industrial camera to move the second current target distance until the third target distance is 0 is returned, wherein the third target distance is the alignment distance. When the alignment distance is 0, the camera center of the industrial camera corresponding to the second servo electric cylinder is aligned with the center of the right front wheel.

5. The vehicle-in-the-loop test vehicle geometric parameter calibration system according to claim 1, characterized in that: For each video frame image in the steering video of the left front wheel, the controller performs a correlation value operation on the video frame image and a preset filter template based on a correlation filtering method to obtain the coordinates of the motion trajectory point of the tracking target in the video frame image, wherein the preset filter template is a filter template with the maximum response obtained when the preset filter template acts on the tracking target in the video frame image, and the tracking target is any point on the wheel; Fitting all the motion trajectory point coordinates of the tracking target in the steering video of the left front wheel by the least square method to obtain a fitted circular curve; The circular curve is solved to obtain a first distance from the center of the left front wheel to the center of rotation of the left front wheel.

6. The vehicle-in-the-loop test vehicle geometric parameter calibration system according to claim 1, characterized in that: The controller calculates a first sum between the first extension distance and the second extension distance, calculates a second sum between the first distance and the second distance, calculates a difference between the first sum and the second sum, and uses the difference as the front wheel kingpin distance of the tested vehicle.

7. The vehicle-in-the-loop test vehicle geometric parameter calibration system as claimed in claim 5, characterized in that: The controller generates a trajectory point arc image based on all motion trajectory points of the tracking target in the steering video of the left front wheel; Determine the coordinates of the starting point, the ending point and the center of the arc formed by all the motion trajectory points in the steering video of the left front wheel in the trajectory point arc image; Calculating a first angle between a straight line formed by the center of the circle and the starting point and a horizontal coordinate axis of the trajectory point arc image; Calculating a second angle between a straight line formed by the center of the circle and the end point and a horizontal coordinate axis of the trajectory point arc image; The current front wheel turning angle of the tested vehicle is calculated based on the first angle and the second angle.

8. The vehicle-in-the-loop test vehicle geometric parameter calibration system according to claim 1, characterized in that: The controller determines the current front wheel angle of the vehicle under test based on a pre-established target deep learning neural network model and the current steering wheel angle, wherein the target deep learning neural network model is used to correlate the steering wheel angle sample data with the corresponding vehicle front wheel angle sample data.

9. The vehicle-in-the-loop test vehicle geometric parameter calibration system as claimed in claim 8, characterized in that: The training process of the target deep learning neural network model is: Obtaining steering wheel angle sample data and corresponding vehicle front wheel angle sample data; The steering wheel angle sample data and the corresponding vehicle front wheel angle sample data are used as model training data to train the initial deep learning neural network model to obtain the target deep learning neural network model.

10. The vehicle geometric parameter calibration system for the vehicle-in-the-loop test as claimed in claim 8 or 9, characterized in that: The target deep learning neural network model is a back propagation BP neural network model.

Images