IMU-based unmanned vehicle body attitude detection method and system, and medium

Abstract

The invention relates to an IMU-based unmanned vehicle body attitude detection method and system and a medium, and the method comprises the steps: Q1, in the operation process of a port unmanned container truck, obtaining multiple groups of motion attitude data information of a vehicle body in real time based on multiple groups of vehicle-mounted IMU inertial sensors, an improved weighted recursive average filtering algorithm is adopted to preprocess the multiple sets of motion attitude data of the vehicle body, and processed multiple sets of motion attitude data information of the vehicle body is obtained; and Q2, based on the processed multiple groups of motion attitude data information of the vehicle body, performing curve fitting on each group of motion attitude data of the vehicle body by adopting a Fourier series fitting algorithm based on gradient descent. The six-degree-of-freedom posture change of each part of the vehicle body can be effectively detected, the input hardware cost is low, the transformation difficulty is small, the data acquisition precision is high, reference information is provided for operation supervision personnel to adjust the running speed of the vehicle end, the purpose of stable running is achieved, and the method has high practicability and advancement.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned container trucks, and in particular to an IMU-based unmanned vehicle body posture detection method, system and medium. Background Art

[0002] As one of the world's largest trading countries, my country's ports play the role of connecting domestic and foreign trade hubs. In recent years, the container throughput of my country's ports has continued to show an upward trend. This growth has also promoted the continuous upgrading of China's port facilities and services to meet the growing demand for freight. Intelligent transportation of unmanned container trucks in ports is an important innovation in today's port logistics field. The application of autonomous driving technology and intelligent scheduling system can realize the automation and intelligence of the container truck transportation process, reduce human operation links, reduce transportation time and cost, and greatly improve the efficiency of port logistics transportation. In the process of loading containers, unmanned container trucks will undergo elastic deformation due to excessive load pressure, causing the body posture to change. When obstacles are sensed during operation, they often brake to ensure safety. Due to inertia, the container will have a huge impact with the vehicle, bringing the risk of damage to the box or body. The uneven road surface causes shaking and brings the risk of overturning and the risk of bumper touching the ground. Intelligent body status and environmental monitoring can improve the safety of unmanned container truck operations, avoid collision accidents, effectively reduce safety hazards, and ensure the safety of port operators and equipment. When the vehicle posture is inconsistent due to the heavy container, the supervisor should issue a deceleration command in time to avoid the inability to brake in time due to excessive load and ensure safe operation. The traditional vehicle posture monitoring scheme assumes that the vehicle is a rigid body, but the unmanned container truck body is too long, the container is too heavy, and the weight distribution in the container is not absolutely uniform. The existing body posture detection method cannot truly reflect the posture changes of various parts of the body. Therefore, there is an urgent need for an intelligent body posture detection solution that is more suitable for unmanned container trucks.

[0003] In the prior art, a Chinese patent (application number: 201810813720.8, publication number: CN110823181A) discloses a sanitation vehicle and a body posture detection method, a body stabilization method and a corresponding system thereof. First, a three-dimensional coordinate system of the body is established, wherein the direction of the X-axis is the direction of the vehicle head, the direction of the Y-axis is the lateral direction of the body perpendicular to the direction of the vehicle head, and the direction of the Z-axis is perpendicular to the plane of the body chassis; then, the acceleration of the body in the X-axis direction, the acceleration in the Y-axis direction and the Z-axis direction, as well as the vehicle body rotation axis with the X-axis as the rotation axis are detected. The body roll angle, the body pitch angle with the Y axis as the rotation axis, and the body yaw angle with the Z axis as the rotation axis; finally, the body posture is obtained according to the obtained data, and the body can be stabilized according to the body posture. This type of posture detection method using IMU generally chooses to install an IMU sensor near the center of the rear wheel of the vehicle, while for unmanned container trucks, posture changes are generally flexible changes of the suspension system and the container fixed platform. The center of the rear wheel is relatively stable. The data detected by the IMU installed here cannot accurately reflect the changes in the posture of various parts of the body. Therefore, this type of method is not suitable for unmanned container trucks.

[0004] In the prior art, a Chinese patent (application number: 202011492933.9, publication number: CN112598705A) discloses a vehicle posture detection method based on binocular vision, which uses a vehicle-mounted binocular camera to collect dynamic video during vehicle operation, preprocess the dynamic video, and detect the feature information of the reference object; then the three-dimensional coordinates of the center of mass of the reference object are calculated through stereo matching of the binocular images, thereby inferring the vehicle posture information. This method involves complex image processing and computer vision technology, which requires a lot of computing resources and algorithm support, resulting in high computational complexity and reducing the real-time performance of the algorithm. Unmanned container trucks need to operate 24 hours a day, and the camera is sensitive to environmental changes. Factors such as lighting, shadows, reflections, and occlusions may affect the accuracy of the algorithm, making it difficult for the binocular camera algorithm to accurately calculate the depth and position of the object, thereby affecting the calculation results of its own posture. In addition, the performance and accuracy of the algorithm are usually affected by the calibration parameters. In order to ensure the accuracy of the binocular camera algorithm, precise camera calibration and calibration are required. However, unmanned container trucks will undergo elastic deformation due to the pressure of containers during operation. Therefore, this method is not suitable for the unmanned container truck’s own posture monitoring and estimation work. Summary of the invention

[0005] In view of the above deficiencies in the prior art, the present invention provides an IMU-based unmanned vehicle body posture detection method, system and medium, which can not only effectively detect the posture changes of the six freedoms of various parts of the vehicle body, but also have low hardware cost, low modification difficulty, and high data collection accuracy, but also provide reference information for operation and supervision personnel to adjust the vehicle-side operating speed to achieve the purpose of smooth operation. It has high practicality and advancement.

[0006] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows: a method for detecting the body posture of an unmanned vehicle based on an IMU, the method comprising:

[0007] Q1. During the unmanned container truck operation at the port, multiple sets of motion posture data information of the vehicle body are obtained in real time based on multiple sets of on-board IMU inertial sensors, and the multiple sets of motion posture data of the vehicle body are pre-processed using an improved weighted recursive average filtering algorithm to obtain the processed multiple sets of motion posture data information of the vehicle body;

[0008] Q2. Based on the processed multiple sets of motion posture data information of the vehicle body, a Fourier series fitting algorithm based on gradient descent is used to perform curve fitting on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the vehicle body after fitting;

[0009] Q3. Based on the multiple sets of motion posture fitting curve data information of the fitted vehicle body, a motion posture variation model of the multi-point vehicle body optimized by the crown porcupine is constructed, the motion posture variation of the vehicle body is characterized, and the motion posture variation data information of the vehicle body is obtained;

[0010] Q4. Based on the vehicle body motion posture change data information, construct the vehicle body posture warning function F, calculate the vehicle body posture warning value, and obtain the vehicle body posture warning value data information.

[0011] Furthermore, the method further comprises:

[0012] Q5. Based on the vehicle's body posture warning value data information, a preset threshold is set. If the vehicle's body posture warning value is less than the preset threshold, the vehicle's body posture is normal and the task continues to be completed. If the vehicle's body posture warning value is greater than the preset threshold, the vehicle's body posture is abnormal, a warning is issued, and the vehicle's body posture is adjusted.

[0013] Furthermore, the vehicle body posture warning function F is:

[0014]

[0015] Among them, r is the data information of the change in the motion posture of the vehicle body, and α1, α2 and α3 are the vehicle body posture warning constant factors.

[0016] Furthermore, the constraint function f of the vehicle's body posture warning constant factors α1, α2 and α3 is,

[0017]

[0018] The value range of the constraint function f is (0,1).

[0019] Furthermore, in step Q1, the preprocessing of multiple groups of motion posture data of the vehicle body by using the improved weighted recursive average filtering algorithm includes:

[0020] Q11. Based on the multiple groups of motion posture data information of the vehicle body, for each group of motion posture data information of the vehicle body, a queue matrix of the vehicle body motion posture is constructed to obtain the queue matrix data information of the vehicle body motion posture;

[0021] Q12. Based on the queue matrix data information of the vehicle body motion posture, construct a weight function G of the vehicle body motion posture queue,

[0022]

[0023] Among them, y i is the queue matrix data information of the vehicle body motion posture, β1 and β2 are the weight determining factors of the queue of the vehicle body motion posture, and the weight of the vehicle body motion posture queue is calculated to obtain the weight data information of the vehicle body motion posture queue;

[0024] Q13. Based on the weight data information of the vehicle body motion posture queue, a weighted average filter function H of the vehicle body motion posture is constructed.

[0025]

[0026] Among them, y i is the queue matrix data information of the vehicle body motion posture, n is the sample capacity, δ i The weight data information of the vehicle body motion posture queue is preprocessed on multiple groups of vehicle body motion posture data to obtain processed multiple groups of vehicle body motion posture data information.

[0027] Furthermore, the weight determining factors β1 and β2 of the vehicle body motion posture queue are:

[0028]

[0029] Where n is the sample size, y i It is the queue matrix data information of the vehicle body motion posture.

[0030] Furthermore, in step Q2, the use of a Fourier series fitting algorithm based on gradient descent to perform curve fitting on each set of motion posture data of the vehicle body includes:

[0031] Q21. Based on the processed multiple groups of motion posture data information of the vehicle body, a gradient descent function W for each group of motion postures of the vehicle body is constructed.

[0032]

[0033] Wherein, z is the processed multiple groups of motion posture data information of the vehicle body, γ1, γ2 and γ3 are the harmonic constant parameters of the multiple groups of motion postures of the vehicle body, and each group of motion postures of the vehicle body is optimized to obtain the optimized multiple groups of motion posture data information of the vehicle body;

[0034] Q22. Based on the multiple sets of motion posture data information of the optimized vehicle body, a Fourier series fitting function R of the vehicle body motion posture is established.

[0035]

[0036] Among them, r is the multiple sets of motion posture data information of the optimized vehicle body, J is a positive integer, a j and b j is the Fourier technology fitting factor of the vehicle body motion state;

[0037] Q23. Based on the Fourier series fitting function R of the vehicle body motion posture, curve fitting is performed on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the fitted vehicle body.

[0038] Further, in step Q3, the construction of the crown porcupine optimized multi-point vehicle body motion posture variation model and characterization of the vehicle body motion posture variation include:

[0039] Q31. Based on the fitted data information of multiple groups of motion posture fitting curves of the vehicle body, a motion posture variation model of a multi-point vehicle body is constructed, and the model parameters are initialized to obtain the initialized model parameter matrix data information;

[0040] Q32. Based on the initialized model parameter matrix data information, establish the crown porcupine optimization function P of the vehicle body's motion posture,

[0041]

[0042] Among them, g is the initialized model parameter matrix data information, η1, η2 and η3 are optimization factors, and the model parameters are optimized to obtain the optimized multi-point vehicle body motion posture change model;

[0043] Q33. Based on the optimized multi-point vehicle body motion posture change model, multiple sets of motion posture fitting curve data information of the fitted vehicle body are input to characterize the vehicle body motion posture change, and obtain the vehicle body motion posture change data information.

[0044] In order to achieve the above-mentioned objectives and other related objectives, the present invention also provides an IMU-based unmanned vehicle body posture detection system, including a computer device, which is programmed or configured to execute the steps of any one of the IMU-based unmanned vehicle body posture detection methods.

[0045] In order to achieve the above-mentioned purpose and other related purposes, the present invention also provides a computer-readable storage medium, which stores a computer program programmed or configured to execute any one of the IMU-based unmanned vehicle body posture detection methods.

[0046] The present invention has the following positive effects:

[0047] 1. The present invention pre-processes multiple groups of motion posture data of the vehicle body by adopting an improved weighted recursive average filtering algorithm, and performs curve fitting on each group of motion posture data of the vehicle body in combination with a Fourier series fitting algorithm based on gradient descent. It can not only effectively detect the posture changes of the six freedoms of various parts of the vehicle body, with low hardware cost, low modification difficulty, and high data collection accuracy, but also improves the accuracy of the vehicle body motion posture, provides reference information for operation supervisors to adjust the vehicle-side running speed, and achieves the purpose of smooth operation, and has high practicality and advancement.

[0048] 2. The present invention characterizes the variation of the motion posture of the vehicle body by constructing a crown porcupine optimized multi-point vehicle body model, and calculates the vehicle body posture warning value in combination with the construction of the vehicle body posture warning function F. It not only monitors the phenomenon that the driving state of the unmanned container truck may change, but also detects the vehicle body state change information by installing multiple six-axis IMUs around the vehicle body. The controller further processes the data to determine whether the vehicle body posture when stopped or in motion meets the warning conditions. If so, a prompt is issued according to the corresponding vehicle body posture. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0050] Figure 2 It is a flow chart of the improved weighted recursive average filtering algorithm of the present invention;

[0051] Figure 3 It is a schematic diagram of the process of the Fourier series fitting algorithm based on gradient descent of the present invention;

[0052] Figure 4 A schematic diagram of a process for constructing a motion posture variation model of a multi-point vehicle body optimized for a crested porcupine according to the present invention;

[0053] Figure 5 The first is a schematic diagram of the vehicle body motion posture of the present invention;

[0054] Figure 6 The second is a schematic diagram of the vehicle body motion posture of the present invention;

[0055] Figure 7 The third is a schematic diagram of the vehicle body motion posture of the present invention;

[0056] Figure 8 Schematic diagram of the vehicle body motion posture of the present invention (IV);

[0057] Fig. 9 The fifth is a schematic diagram of the vehicle body motion posture of the present invention;

[0058] Fig.10 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0059] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0060] Example 1: Figure 1 As shown, a method for detecting the body posture of an unmanned vehicle based on an IMU, the method comprising:

[0061] Q1. During the unmanned container truck operation at the port, multiple sets of motion posture data information of the vehicle body are obtained in real time based on multiple sets of on-board IMU inertial sensors, and the multiple sets of motion posture data of the vehicle body are pre-processed using an improved weighted recursive average filtering algorithm to obtain the processed multiple sets of motion posture data information of the vehicle body;

[0062] Q2. Based on the processed multiple sets of motion posture data information of the vehicle body, a Fourier series fitting algorithm based on gradient descent is used to perform curve fitting on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the vehicle body after fitting;

[0063] Q3. Based on the multiple sets of motion posture fitting curve data information of the fitted vehicle body, a motion posture variation model of the multi-point vehicle body optimized by the crown porcupine is constructed, the motion posture variation of the vehicle body is characterized, and the motion posture variation data information of the vehicle body is obtained;

[0064] Q4. Based on the vehicle body motion posture change data information, construct the vehicle body posture warning function F, calculate the vehicle body posture warning value, and obtain the vehicle body posture warning value data information.

[0065] In this embodiment, the method further includes:

[0066] Q5. Based on the vehicle's body posture warning value data information, a preset threshold is set. If the vehicle's body posture warning value is less than the preset threshold, the vehicle's body posture is normal and the task continues to be completed. If the vehicle's body posture warning value is greater than the preset threshold, the vehicle's body posture is abnormal, a warning is issued, and the vehicle's body posture is adjusted.

[0067] In this embodiment, the vehicle body posture warning function F is:

[0068]

[0069] Among them, r is the data information of the change in the motion posture of the vehicle body, and α1, α2 and α3 are the vehicle body posture warning constant factors.

[0070] In this embodiment, the constraint function f of the vehicle body posture warning constant factors α1, α2 and α3 is,

[0071]

[0072] The value range of the constraint function f is (0,1).

[0073] In this embodiment, if Figure 2 As shown, in step Q1, the preprocessing of multiple groups of motion posture data of the vehicle body by using the improved weighted recursive average filtering algorithm includes:

[0074] Q11. Based on the multiple groups of motion posture data information of the vehicle body, for each group of motion posture data information of the vehicle body, a queue matrix of the vehicle body motion posture is constructed to obtain the queue matrix data information of the vehicle body motion posture;

[0075] Q12. Based on the queue matrix data information of the vehicle body motion posture, construct a weight function G of the vehicle body motion posture queue,

[0076]

[0077] Among them, yi is the queue matrix data information of the vehicle body motion posture, β1 and β2 are the weight determining factors of the queue of the vehicle body motion posture, and the weight of the vehicle body motion posture queue is calculated to obtain the weight data information of the vehicle body motion posture queue;

[0078] Q13. Based on the weight data information of the vehicle body motion posture queue, a weighted average filter function H of the vehicle body motion posture is constructed.

[0079]

[0080] Among them, y i is the queue matrix data information of the vehicle body motion posture, n is the sample capacity, δ i The weight data information of the vehicle body motion posture queue is preprocessed on multiple groups of vehicle body motion posture data to obtain processed multiple groups of vehicle body motion posture data information.

[0081] In this embodiment, the weight determining factors β1 and β2 of the vehicle body motion posture queue are:

[0082]

[0083] Where n is the sample size, y i It is the queue matrix data information of the vehicle body motion posture.

[0084] In this embodiment, if Figure 3 As shown, in step Q2, the curve fitting of each group of motion posture data of the vehicle body using the Fourier series fitting algorithm based on gradient descent includes:

[0085] Q21. Based on the processed multiple groups of motion posture data information of the vehicle body, a gradient descent function W for each group of motion postures of the vehicle body is constructed.

[0086]

[0087] Wherein, z is the processed multiple groups of motion posture data information of the vehicle body, γ1, γ2 and γ3 are the harmonic constant parameters of the multiple groups of motion postures of the vehicle body, and each group of motion postures of the vehicle body is optimized to obtain the optimized multiple groups of motion posture data information of the vehicle body;

[0088] Q22. Based on the multiple sets of motion posture data information of the optimized vehicle body, a Fourier series fitting function R of the vehicle body motion posture is established.

[0089]

[0090] Among them, r is the multiple sets of motion posture data information of the optimized vehicle body, J is a positive integer, a j and bj is the Fourier technology fitting factor of the vehicle body motion state;

[0091] Q23. Based on the Fourier series fitting function R of the vehicle body motion posture, curve fitting is performed on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the fitted vehicle body.

[0092] Example 2: Based on the IMU-based unmanned vehicle body posture detection method in Example 1, the present invention is further illustrated and described below.

[0093] like Figure 1 As shown, a method for detecting the body posture of an unmanned vehicle based on an IMU, the method comprising:

[0094] Q1. During the unmanned container truck operation at the port, multiple sets of motion posture data information of the vehicle body are obtained in real time based on multiple sets of on-board IMU inertial sensors, and the multiple sets of motion posture data of the vehicle body are pre-processed using an improved weighted recursive average filtering algorithm to obtain the processed multiple sets of motion posture data information of the vehicle body;

[0095] Q2. Based on the processed multiple sets of motion posture data information of the vehicle body, a Fourier series fitting algorithm based on gradient descent is used to perform curve fitting on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the vehicle body after fitting;

[0096] Q3. Based on the multiple sets of motion posture fitting curve data information of the fitted vehicle body, a motion posture variation model of the multi-point vehicle body optimized by the crown porcupine is constructed, the motion posture variation of the vehicle body is characterized, and the motion posture variation data information of the vehicle body is obtained;

[0097] Q4. Based on the vehicle body motion posture change data information, construct the vehicle body posture warning function F, calculate the vehicle body posture warning value, and obtain the vehicle body posture warning value data information.

[0098] In this embodiment, if Figure 4 As shown, in step Q3, the construction of the crown porcupine optimized multi-point vehicle body motion posture variation model and characterization of the vehicle body motion posture variation include:

[0099] Q31. Based on the fitted data information of multiple groups of motion posture fitting curves of the vehicle body, a motion posture variation model of a multi-point vehicle body is constructed, and the model parameters are initialized to obtain the initialized model parameter matrix data information;

[0100] Q32. Based on the initialized model parameter matrix data information, establish the crown porcupine optimization function P of the vehicle body's motion posture,

[0101]

[0102] Among them, g is the initialized model parameter matrix data information, η1, η2 and η3 are optimization factors, and the model parameters are optimized to obtain the optimized multi-point vehicle body motion posture change model;

[0103] Q33. Based on the optimized multi-point vehicle body motion posture change model, multiple sets of motion posture fitting curve data information of the fitted vehicle body are input to characterize the vehicle body motion posture change, and obtain the vehicle body motion posture change data information.

[0104] In this embodiment, the present invention provides an IMU-based unmanned vehicle body posture detection system, including a computer device, which is programmed or configured to execute the steps of any one of the IMU-based unmanned vehicle body posture detection methods.

[0105] In this embodiment, the present invention provides a computer-readable storage medium, on which is stored a computer program programmed or configured to execute any one of the IMU-based unmanned vehicle body posture detection methods.

[0106] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM (RDRAM).

[0107] In this embodiment, the traditional vehicle posture monitoring method focuses on the heading angle and position, ignoring the changes in pitch angle and roll angle. The container spreader places the container at a random position on the unmanned container truck. Containers of different positions and weights, insufficient tire pressure, and whether the road surface is flat will cause inconsistent posture changes in different areas of the vehicle. According to the operation of unmanned container trucks in ports, the vehicle body deflection state is divided into the following five situations:

[0108] (a) If Figure 5 As shown, in case 1, the left side of the single / double box body is tilted downward: the deviation of the two IMU devices on the left side exceeds 1° (lasting for more than 30 seconds), which is judged as the left side body tilting;

[0109] (b) If Figure 6 As shown, in case 2, the right side of the single / double box body is tilted downward: the deviation of the two front and rear IMU devices on the right side exceeds 1° (lasting for more than 30 seconds), which is judged as the right side body tilting;

[0110] (c) Figure 7 As shown, in case three, the right side of the single / double box body is tilted downward: the two imu devices on the front beam deviate more than 1° from the horizontal state (lasting for more than 30s), which is judged as the front body tilting;

[0111] (d) If Figure 8 As shown, in case 4, the right side of the single / double box body is tilted downward: the two IMU devices on the rear beam deviate more than 1° from the horizontal state (lasting for more than 30s), which is judged as the rear body tilting;

[0112] (e) If Fig. 9 As shown, in situation five, the middle of the double-box body is downward and the four corners are tilted upward: if three or more IMU devices deviate from the horizontal state by more than 1° (lasting for more than 30 seconds), it is judged that the middle of the body is tilted.

[0113] In this embodiment, if Fig.10 As shown, the system consists of: vehicle motion posture acquisition module, posture solution and judgment module, information transmission and communication module, and platform receiving end display module.

[0114] Among them, the IMU sensor terminals installed at the four corners of the vehicle are used to obtain the posture of the vehicle body in real time, and then the data is transmitted to the controller, and the posture solution and judgment module determines the vehicle posture change and operating status at the current moment. The posture change judgment result is sent to the remote monitoring platform through the 5G data transmission module, providing a reference for the supervisory operation personnel to adjust the vehicle operation status in time, so as to take corresponding actions to improve the safety of the operation process.

[0115] The vehicle motion posture acquisition module selects the front lower right coordinate system commonly used in mobile carrier research as the local coordinate system of the vehicle body. Specifically, the forward direction of the vehicle body is the X-axis direction, and the rotation around the X-axis is the roll angle. According to the right-hand rule, counterclockwise rotation is positive, and its change represents the degree of left and right tilt of the vehicle body when overloaded; the axis perpendicular to the forward direction of the vehicle body and pointing to the right side of the vehicle body is the Y-axis, and the rotation around the Y-axis is the pitch angle. According to the right-hand rule, counterclockwise rotation is positive, and its change represents the degree of tilting or pressing down of the vehicle body when overloaded; the Z-axis is perpendicular to the chassis plane downward, and the rotation around the Z-axis is the yaw angle. According to the right-hand rule, counterclockwise rotation is positive; the attitude angle of the vehicle body in the flat and unloaded state is recorded as the initial value.

[0116] The posture calculation and judgment module collects the body posture change information obtained by the six-axis IMU installed around the body. First, the data processing algorithm is used to filter out the abnormal interference signals such as clutter and ripple in the data to obtain continuous, smooth and stable real-time change information of the body posture. Then it is compared with the initial value stored in the static state on the flat ground to determine which situation the body posture change currently belongs to. If the body posture change reaches the warning condition, an alarm message will be issued.

[0117] The information transmission and communication module sends the alarm information issued by the posture calculation and judgment module to the remote operation supervisor through 5G communication.

[0118] In summary, the present invention can not only effectively detect the posture changes of the six freedoms of various parts of the vehicle body, with low hardware cost, low modification difficulty, and high data collection accuracy, but also provide reference information for operation supervisors to adjust the vehicle-side operating speed to achieve the purpose of smooth operation. It has high practicality and advancement.

[0119] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for detecting the body posture of an unmanned vehicle based on IMU, characterized in that: The method comprises: Q1. During the unmanned container truck operation at the port, multiple sets of motion posture data information of the vehicle body are obtained in real time based on multiple sets of on-board IMU inertial sensors, and the multiple sets of motion posture data of the vehicle body are pre-processed using an improved weighted recursive average filtering algorithm to obtain the processed multiple sets of motion posture data information of the vehicle body; Q2. Based on the processed multiple sets of motion posture data information of the vehicle body, a Fourier series fitting algorithm based on gradient descent is used to perform curve fitting on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the vehicle body after fitting; Q3. Based on the multiple sets of motion posture fitting curve data information of the fitted vehicle body, a motion posture variation model of the multi-point vehicle body optimized by the crown porcupine is constructed, the motion posture variation of the vehicle body is characterized, and the motion posture variation data information of the vehicle body is obtained; Q4. Based on the vehicle body motion posture change data information, construct the vehicle body posture warning function F, calculate the vehicle body posture warning value, and obtain the vehicle body posture warning value data information.

2. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 1, characterized in that: The method further comprises: Q5. Based on the vehicle's body posture warning value data information, a preset threshold is set. If the vehicle's body posture warning value is less than the preset threshold, the vehicle's body posture is normal and the task continues to be completed. If the vehicle's body posture warning value is greater than the preset threshold, the vehicle's body posture is abnormal, a warning is issued, and the vehicle's body posture is adjusted.

3. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 1, characterized in that: The vehicle body posture warning function F is: Among them, r is the data information of the change in the motion posture of the vehicle body, and α1, α2 and α3 are the vehicle body posture warning constant factors.

4. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 3 is characterized in that: The constraint function f of the vehicle's body posture warning constant factors α1, α2 and α3 is, The value range of the constraint function f is (0,1).

5. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 1, characterized in that: In step Q1, the preprocessing of multiple groups of motion posture data of the vehicle body by using the improved weighted recursive average filtering algorithm includes: Q11. Based on the multiple groups of motion posture data information of the vehicle body, for each group of motion posture data information of the vehicle body, a queue matrix of the vehicle body motion posture is constructed to obtain the queue matrix data information of the vehicle body motion posture; Q12. Based on the queue matrix data information of the vehicle body motion posture, construct a weight function G of the vehicle body motion posture queue, Among them, y i is the queue matrix data information of the vehicle body motion posture, β1 and β2 are the weight determining factors of the queue of the vehicle body motion posture, and the weight of the vehicle body motion posture queue is calculated to obtain the weight data information of the vehicle body motion posture queue; Q13. Based on the weight data information of the vehicle body motion posture queue, a weighted average filter function H of the vehicle body motion posture is constructed. Among them, y i is the queue matrix data information of the vehicle body motion posture, n is the sample capacity, δ i The weight data information of the vehicle body motion posture queue is preprocessed on multiple groups of vehicle body motion posture data to obtain processed multiple groups of vehicle body motion posture data information.

6. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 5, characterized in that: The weight determining factors β1 and β2 of the vehicle body motion posture queue are: Where n is the sample size, y i It is the queue matrix data information of the vehicle body motion posture.

7. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 1, characterized in that: In step Q2, the curve fitting of each group of motion posture data of the vehicle body using the Fourier series fitting algorithm based on gradient descent includes: Q21. Based on the processed multiple groups of motion posture data information of the vehicle body, a gradient descent function W for each group of motion postures of the vehicle body is constructed. Wherein, z is the processed multiple groups of motion posture data information of the vehicle body, γ1, γ2 and γ3 are the harmonic constant parameters of the multiple groups of motion postures of the vehicle body, and each group of motion postures of the vehicle body is optimized to obtain the optimized multiple groups of motion posture data information of the vehicle body; Q22. Based on the multiple sets of motion posture data information of the optimized vehicle body, a Fourier series fitting function R of the vehicle body motion posture is established. Among them, r is the multiple sets of motion posture data information of the optimized vehicle body, J is a positive integer, a j and b j is the Fourier technology fitting factor of the vehicle body motion state; Q23. Based on the Fourier series fitting function R of the vehicle body motion posture, curve fitting is performed on each set of motion posture data of the vehicle body to obtain multiple sets of motion posture fitting curve data information of the fitted vehicle body.

8. The method for detecting the body posture of an unmanned vehicle based on IMU according to claim 1, characterized in that: In step Q3, the construction of the crown porcupine optimized multi-point vehicle body motion posture variation model and characterization of the vehicle body motion posture variation include: Q31. Based on the fitted data information of multiple groups of motion posture fitting curves of the vehicle body, a motion posture variation model of a multi-point vehicle body is constructed, and the model parameters are initialized to obtain the initialized model parameter matrix data information; Q32. Based on the initialized model parameter matrix data information, establish the crown porcupine optimization function P of the vehicle body's motion posture, Among them, g is the initialized model parameter matrix data information, η1, η2 and η3 are optimization factors, and the model parameters are optimized to obtain the optimized multi-point vehicle body motion posture change model; Q33. Based on the optimized multi-point vehicle body motion posture change model, multiple sets of motion posture fitting curve data information of the fitted vehicle body are input to characterize the vehicle body motion posture change, and obtain the vehicle body motion posture change data information.

9. An unmanned vehicle body posture detection system based on IMU, including computer equipment, characterized in that: The computer device is programmed or configured to execute the steps of the unmanned vehicle body posture detection method based on IMU as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that is programmed or configured to execute the IMU-based unmanned vehicle body posture detection method as described in any one of claims 1 to 8.

Images