A method for detecting the turning angle of a track-type construction machine

Abstract

This invention discloses a method for detecting the slewing angle of tracked construction machinery, relating to the field of construction machinery control technology. The detection method is applied to a device for detecting the slewing angle of tracked construction machinery. The device includes an image acquisition device, a video controller, and a vehicle controller. The method includes the following steps executed by the video controller: acquiring an image of the track plates using the image acquisition device; performing image processing on the track plate image to identify and fit a straight line at the outer edge of the track plate; generating an angle signal based on the angle between the straight line at the outer edge of the track plate and the horizontal direction of the image; and sending the angle signal to the vehicle controller, so that the vehicle controller can determine the slewing angle based on the angle signal and process and display the slewing angle.

Description

Technical Field

[0001] This invention relates to a method for detecting the rotation angle of tracked construction machinery, belonging to the field of construction machinery control technology. Background Technology

[0002] Upgrading construction machinery to intelligent levels using vision technology has become an industry trend. Cameras are easily deployed around construction machinery, whether for in-plant installation or subsequent modifications. Slewing angle detection is a crucial component of construction machinery attitude monitoring; for example, many overseas countries require slewing angle detection for cranes.

[0003] Currently, rotation angle detection typically requires the installation of an encoder that meshes with the rotation mechanism. However, the rotation meshing mechanism of some construction machinery is hidden inside the vehicle body. After the vehicle is assembled, it is difficult to install the relevant structure without disassembling the vehicle body. This makes it difficult to detect the rotation angle in external modification scenarios where there is no large-scale disassembly and assembly of the vehicle body.

[0004] Existing solutions mostly rely on adding encoders that mesh with the gears of the slewing mechanism to achieve angle detection. This not only involves significant modification and installation difficulties, but also results in high encoder costs and limited functionality. Currently, most construction machinery in China is not equipped with such sensors when it leaves the factory. Summary of the Invention

[0005] The purpose of this invention is to provide a method for detecting the rotation angle of tracked construction machinery. By installing an image acquisition device on the tracked construction machinery and calculating the rotation angle using an image recognition algorithm, the problem of externally installing a conventional rotation angle detection encoder can be solved, and the cost can be saved by eliminating the need to install a separate angle detection encoder.

[0006] To achieve the above objectives, the present invention is implemented using the following technical solution.

[0007] This invention provides a method for detecting the rotation angle of tracked construction machinery. The method is applied to a detection device for the rotation angle of tracked construction machinery, the device comprising an image acquisition device, a video controller, and a vehicle controller. The method includes the following steps executed by the video controller:

[0008] The image acquisition device acquires images of the track plates;

[0009] Image processing is performed on the track plate image to identify and fit the straight line of the outer edge of the track plate;

[0010] An angle signal is generated based on the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image; wherein, the horizontal direction of the image is the horizontal axis direction of the image acquired by the image acquisition device;

[0011] The included angle signal is sent to the vehicle controller, so that the vehicle controller can determine the turning angle based on the included angle signal, and process and display the turning angle.

[0012] Further, the step of performing image processing on the track plate image to identify and fit the straight line of the outer edge of the track plate includes:

[0013] The track plate image is subjected to image distortion correction processing to obtain a corrected image;

[0014] The corrected image is subjected to image projection transformation to obtain the projected image of the track plate on the horizontal plane;

[0015] The outer edge line of the track plate is fitted using an image recognition algorithm based on the projected image.

[0016] Further, the step of fitting the straight line of the outer edge of the track plate based on the projected image using an image recognition algorithm includes:

[0017] The track plate contour in the projected image is identified by a pre-trained deep learning object detection model, and the outer edge line of the track plate is obtained by fitting the contour points of the track plate contour.

[0018] Furthermore, the deep learning object detection model includes the YOLO series of object detection models; the pre-training dataset of the deep learning object detection model contains track plate images with different rotation angles.

[0019] Furthermore, the method also includes:

[0020] When the rotation angle of the tracked engineering machinery is zero, the image acquisition device acquires an initial image of the track plate, obtains the initial angle between the straight line of the outer edge of the track plate and the horizontal direction of the image, and stores the initial angle as a zero-point compensation value in the storage unit of the vehicle controller. The vehicle controller determines the rotation angle based on the angle signal and the zero-point compensation value.

[0021] Further, generating an angle signal based on the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image includes:

[0022] Obtain the two endpoints of the straight line at the outer edge of the track plate;

[0023] The slope of the straight line at the outer edge of the track plate is obtained based on the coordinates of the two endpoints.

[0024] Based on the slope, the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image is obtained by inverse trigonometric function calculation;

[0025] The included angle signal is generated based on the included angle.

[0026] Furthermore, the vehicle controller determines the turning angle based on the included angle signal and the zero-point compensation value, including:

[0027] The video controller sends the included angle signal to the vehicle controller;

[0028] The vehicle controller obtains the zero-point compensation value from the storage unit and performs a difference calculation between the included angle signal and the zero-point compensation value to determine the turning angle.

[0029] Furthermore, the method also includes:

[0030] The vehicle controller detects the slewing motion control signal.

[0031] The slewing motion control signal is used as the state control variable of the Kalman filter, and the slewing angle is used as the measurement variable of the Kalman filter. The slewing angle is then filtered and corrected using the Kalman filter.

[0032] Furthermore, the method also includes:

[0033] When the slewing motion control signal indicates that the slewing motion has stopped, the currently output slewing angle is locked to eliminate signal jitter in the stationary state, so that the value of the slewing angle remains unchanged until a new slewing motion control signal is detected.

[0034] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0035] This invention enables rapid and low-cost external modification of tracked construction machinery that lacks sensors by covering the entire track plate area with an image acquisition device. This avoids the need for construction work inside the tracked machinery, eliminating the need for hoisting and disassembling the vehicle body. Furthermore, this invention replaces the encoder with an image acquisition device, saving both the hardware cost of purchasing and installing encoders, as well as the labor and downtime costs associated with encoder installation. Simultaneously, this invention uses image recognition algorithms to calculate the rotation angle, simplifying the angle conversion logic and facilitating implementation and debugging in embedded systems. Attached Figure Description

[0036] Figure 1 The diagram shown is a flowchart illustrating a method for detecting the rotation angle of tracked engineering machinery provided by the present invention. Detailed Implementation

[0037] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0038] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0039] Example 1

[0040] This embodiment provides a detection device for the rotation angle of tracked engineering machinery. The detection device includes an image acquisition device, a video controller, and a vehicle controller.

[0041] The image acquisition device is mounted on the upper part of the tracked construction machinery. Its field of view covers the track plates area of ​​the machinery, and it is used to acquire images of the track plates. As a specific implementation, the image acquisition device can use a construction machinery camera, preferably mounted on the left / right side of the upper part of the tracked construction machinery. This ensures that the camera's mounting angle allows for clear and complete acquisition of the track plate images.

[0042] The video controller and the image acquisition device are electrically connected via analog or digital video signals. Specifically, the video controller can receive track images acquired by the image acquisition device through communication methods such as CAN bus, Ethernet, or dedicated video cables. The video controller processes the received track images and calculates the included angle signal corresponding to the rotation angle of the tracked construction machinery. The included angle signal reflects the quantized value of the rotation angle of the upper body of the tracked construction machinery relative to the lower track chassis in the image coordinate system.

[0043] The vehicle controller and video controller are electrically connected via a communication bus. The vehicle controller is standard equipment for tracked construction machinery and can collect and display vehicle information. The vehicle controller receives the angle signal sent by the video controller and converts the angle signal into the actual rotation angle through a preset conversion relationship within the vehicle controller. Finally, the vehicle controller processes the actual rotation angle and displays it through an interactive interface such as a display screen for operator reference.

[0044] Example 2

[0045] See Figure 1This embodiment provides a detection method for the detection device for the rotation angle of tracked engineering machinery provided in Embodiment 1. The specific steps are as follows:

[0046] Step S1: Acquire the image of the track plate captured by the image acquisition device;

[0047] The video controller in the tracked construction machinery rotation angle detection device acquires images of the track plates captured by the image acquisition device;

[0048] Step S2: Perform image processing on the track plate image to identify and fit the straight line of the outer edge of the track plate;

[0049] Because optical distortion exists through the camera lens, the video controller first performs image distortion correction processing on the received track plate image to eliminate the errors and effects caused by lens distortion. As a specific implementation, image distortion correction processing can employ camera calibration-based distortion correction, or methods such as radial distortion correction, tangential distortion correction, and perspective distortion correction. This embodiment preferentially uses a camera calibration-based distortion correction method to correct the track plate image. Specifically, the camera used in this embodiment is first calibrated to obtain its internal parameters and distortion parameters. Then, these parameters are used to perform distortion correction processing on the track plate image to be processed, resulting in a corrected image.

[0050] Next, the video controller performs image projection transformation on the corrected image to obtain the projected image of the track plate on the horizontal plane. The purpose of this step is to eliminate perspective distortion caused by the camera's installation angle, so that the geometric relationships in the image are consistent with the actual planar geometry.

[0051] Finally, the video controller fits the outer edge line of the track plate to the projected image using an image recognition algorithm. The image recognition algorithm is a deep learning object detection or segmentation algorithm pre-trained on a large number of training samples, including track plate images at different rotation angles, to ensure the model's accuracy and robustness under various working conditions. This embodiment preferentially uses the YOLO series segmentation model to recognize the contour of the track plate image in real time, and fits the identified contour points to obtain the outer edge line of the track plate.

[0052] Step S3: Generate an angle signal based on the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image;

[0053] In this embodiment, the horizontal direction of the image is preferably selected from the horizontal axis direction of the image acquired by the image acquisition device.

[0054] First, the video controller acquires the two endpoints of the straight line on the outer edge of the track plate; then, it calculates the slope of the straight line on the outer edge of the track plate based on the coordinates of the two endpoints; finally, it obtains the included angle by performing inverse trigonometric function calculation based on the slope, and generates the included angle signal based on the included angle.

[0055] In this embodiment, the default installation position of the image acquisition device is that the horizontal direction of the image is parallel to the outer edge of the track plate. When there is a deviation in the installation of the camera, the horizontal direction of the image can be corrected by zero-point calibration. For details of zero-point calibration, please refer to step S5.

[0056] Step S4: Send the included angle signal to the vehicle controller so that the vehicle controller can determine the turning angle based on the included angle signal, and process and display the turning angle.

[0057] The video controller sends the included angle signal generated in step S3 to the vehicle controller. The vehicle controller converts the included angle signal into the actual turning angle according to an internally preset conversion relationship. As a specific implementation, the conversion relationship can use a proportional coefficient or a calibration curve. The vehicle controller processes the turning angle and displays the current turning angle in real time on the display screen for the operator's reference.

[0058] Step S5: Zero point calibration;

[0059] When the rotation angle is 0, the video controller acquires the initial track plate image captured by the image acquisition device;

[0060] Image processing is performed on the initial track plate image to obtain the initial angle between the outer edge line of the track plate in the initial track plate image and the horizontal direction of the image;

[0061] The initial included angle is stored as a zero-point compensation value in the storage unit of the vehicle controller;

[0062] In the actual testing process, the actual rotation angle is determined based on the included angle and zero-point compensation value obtained in real time in step S3, that is, the actual rotation angle = real-time included angle value - zero-point compensation value.

[0063] Step S6: Filter correction;

[0064] To improve the stability of angle detection, the detection method provided in this embodiment also includes a filtering correction step.

[0065] Acquire the slewing control signal sent by the vehicle controller. The slewing control signal includes the slewing handle signal, slewing start / stop signal, etc.

[0066] The slewing motion control signal is used as the control variable, and the included angle calculated in real time in step S3 is used as the measurement variable.

[0067] The included angle is optimized by using the Kalman filter algorithm to filter out noise introduced by factors such as image recognition jitter, and a stable included angle signal is obtained.

[0068] Step S7: Angle locking;

[0069] To prevent the rotation angle value from drifting due to minor disturbances in the stationary state when the rotation motion stops, the detection method provided in the embodiment further includes an angle locking step:

[0070] When the rotation control signal obtained in step S6 indicates that the rotation action has stopped, the video controller locks the currently output angle signal to keep its value unchanged, eliminating signal jitter in the stationary state, thereby keeping the rotation angle value unchanged until a new rotation control signal is detected.

[0071] Through the above steps, this embodiment achieves non-contact, high-precision, and low-cost detection of the rotation angle of tracked engineering machinery, and has good stability and anti-interference ability.

[0072] Example 3

[0073] Based on the detection device for the rotation angle of tracked construction machinery provided in Embodiment 1 and the detection method for the rotation angle of tracked construction machinery provided in Embodiment 2, this embodiment provides an optimized scheme that integrates an inertial measurement unit (IMU) and / or a global positioning system (GPS) to further improve the accuracy and reliability of rotation angle detection.

[0074] An IMU and / or GPS are installed on the upper body of the tracked construction machinery to measure the angular velocity and acceleration of the upper body and to obtain its absolute position coordinates. The IMU and / or GPS are connected to the video controller and the vehicle controller, respectively.

[0075] During the detection process, multiple data fusions are added, which further improves the dynamic response speed, anti-interference ability and absolute accuracy of rotation angle detection while maintaining the core advantages of the vision method. It is suitable for complex working conditions with higher requirements for angle detection.

Claims

1. A method for detecting the slewing angle of tracked engineering machinery, characterized in that, The detection method is applied to a device for detecting the rotation angle of tracked engineering machinery. The device includes an image acquisition device, a video controller, and a vehicle controller. The method includes the following steps performed by the video controller: The image acquisition device acquires images of the track plates; Image processing is performed on the track plate image to identify and fit the straight line of the outer edge of the track plate; An angle signal is generated based on the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image; wherein, the horizontal direction of the image is the horizontal axis direction of the image acquired by the image acquisition device; The included angle signal is sent to the vehicle controller, so that the vehicle controller can determine the turning angle based on the included angle signal, and process and display the turning angle.

2. The method for detecting the rotation angle of tracked engineering machinery according to claim 1, characterized in that, The step of performing image processing on the track plate image to identify and fit the straight line of the outer edge of the track plate includes: The track plate image is subjected to image distortion correction processing to obtain a corrected image; The corrected image is subjected to image projection transformation to obtain the projected image of the track plate on the horizontal plane; The outer edge line of the track plate is fitted using an image recognition algorithm based on the projected image.

3. The method for detecting the rotation angle of tracked engineering machinery according to claim 2, characterized in that, The step of fitting the straight line of the outer edge of the track plate based on the projected image using an image recognition algorithm includes: The track plate contour in the projected image is identified by a pre-trained deep learning object detection model, and the outer edge line of the track plate is obtained by fitting the contour points of the track plate contour.

4. The method for detecting the rotation angle of tracked engineering machinery according to claim 3, characterized in that, The deep learning object detection model includes the YOLO series of object detection models; the pre-training dataset of the deep learning object detection model contains track plate images with different rotation angles.

5. The method for detecting the rotation angle of tracked engineering machinery according to claim 1, characterized in that, The method further includes: When the rotation angle of the tracked engineering machinery is zero, the image acquisition device acquires an initial image of the track plate, obtains the initial angle between the straight line of the outer edge of the track plate and the horizontal direction of the image, and stores the initial angle as a zero-point compensation value in the storage unit of the vehicle controller. The vehicle controller determines the rotation angle based on the angle signal and the zero-point compensation value.

6. The method for detecting the rotation angle of tracked engineering machinery according to claim 5, characterized in that, The angle signal is generated based on the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image, including: Obtain the two endpoints of the straight line at the outer edge of the track plate; The slope of the straight line at the outer edge of the track plate is obtained based on the coordinates of the two endpoints. Based on the slope, the angle between the straight line of the outer edge of the track plate and the horizontal direction of the image is obtained by inverse trigonometric function calculation; The included angle signal is generated based on the included angle.

7. The method for detecting the rotation angle of tracked engineering machinery according to claim 6, characterized in that, The vehicle controller determines the turning angle based on the included angle signal and the zero-point compensation value, including: The video controller sends the included angle signal to the vehicle controller; The vehicle controller obtains the zero-point compensation value from the storage unit and performs a difference calculation between the included angle signal and the zero-point compensation value to determine the turning angle.

8. The method for detecting the rotation angle of tracked engineering machinery according to claim 7, characterized in that, The method further includes: The vehicle controller detects the slewing motion control signal. The slewing motion control signal is used as the state control variable of the Kalman filter, and the slewing angle is used as the measurement variable of the Kalman filter. The slewing angle is then filtered and corrected using the Kalman filter.

9. The method for detecting the rotation angle of tracked engineering machinery according to claim 8, characterized in that, The method further includes: When the slewing motion control signal indicates that the slewing motion has stopped, the currently output slewing angle is locked to eliminate signal jitter in the stationary state, so that the value of the slewing angle remains unchanged until a new slewing motion control signal is detected.

Images