Control method, device and system, unmanned loader, medium and program product

Abstract

The invention provides a control method, device and system, an unmanned loader, a medium and a program product. The control method comprises the steps that data information related to a material pile and cavity pressure information of an execution mechanism are obtained; obtaining a looseness score and a viscosity score based on the temperature information and the image information of the material pile; calculating to obtain a proportion parameter and an integral parameter based on the height information of the bucket cutting into the material pile and the volume information, the looseness score and the viscosity score of the material pile; calculating to obtain a pressure difference value between the cavity pressure threshold value and the cavity pressure information; the proportional parameter and the integral parameter serve as parameters of a proportional-integral control algorithm, the pressure difference value serves as input data of the proportional-integral control algorithm, and a compensation angle is obtained through the proportional-integral control algorithm; calculating an angle sum value between the compensation angle and a main control angle obtained through a main control algorithm; and adjusting the angle of the actuator based on the angle sum value such that the angle of the actuator conforms to the angle sum value.

Description

Technical Field

[0001] The present disclosure relates to the technical field of unmanned control of engineering machinery, and in particular to a control method, device, system, unmanned loader, medium and program product. Background Art

[0002] With the advent of the era of big data and artificial intelligence, the demand for unmanned construction machinery is becoming increasingly urgent. Unmanned loaders, as key equipment in scenarios such as mines and ports, have a direct impact on operational efficiency due to their automation level. They can not only achieve high work efficiency and quality, but also greatly reduce operating costs and improve operational safety and reliability. However, the related technical research on unmanned construction machinery still uses a rigid shoveling method. The shoveling and unloading methods are too simple and the movements are rigid, which cannot achieve the delicate and gentle operation of manual operators. However, manual operation has inherent physiological limitations. As the working hours increase, the efficiency and quality of the operation will also decline. Summary of the Invention

[0003] A technical problem solved by the present disclosure is that in the related art, the efficiency and quality of unmanned loaders when shoveling materials are low.

[0004] According to one aspect of the present disclosure, a control method for an unmanned loader is provided, comprising: obtaining data information related to a material pile and cavity pressure information of an actuator, wherein the data information includes: height information of a bucket cutting into the material pile and volume information, temperature information and image information of the material pile; the cavity pressure information of the actuator includes at least one of cavity pressure information of a boom and cavity pressure information of a bucket; obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile; obtaining a looseness score and a viscosity score of the material pile based on the height information of the bucket cutting into the material pile and the volume information, temperature information and image information of the material pile; obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile; obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile; obtaining a looseness score and a viscosity score of the material pile based on the The method comprises the following steps: calculating the proportional parameter and the integral parameter based on the cavity pressure information, the looseness score and the viscosity score; calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator; using the proportional parameter and the integral parameter as parameters of a proportional-integral control algorithm, using the pressure difference as input data of the proportional-integral control algorithm, and obtaining a compensation angle through the proportional-integral control algorithm; calculating the angle sum between the compensation angle and the main control angle obtained by the main control algorithm; and adjusting the angle of the actuator based on the angle sum so that the angle of the actuator conforms to the angle sum.

[0005] In some embodiments, a proportional parameter and an integral parameter are calculated based on the height information of the bucket cutting into the material pile and the volume information, looseness score and viscosity score of the material pile, including: normalizing the height information of the bucket cutting into the material pile and the volume information, looseness score and viscosity score of the material pile respectively to obtain normalized height information, normalized volume information, normalized looseness score and normalized viscosity score; calculating a first parameter based on the normalized height information and the normalized volume information, wherein the first parameter is a comprehensive parameter of the height and volume of the material pile; calculating a second parameter based on the normalized looseness score and the normalized viscosity score, wherein the second parameter is a comprehensive parameter of the looseness and viscosity of the material pile; and calculating the proportional parameter and the integral parameter based on the first parameter and the second parameter.

[0006] In some embodiments, the first parameter for:

[0007] ,

[0008] in, is the normalized height information, is the normalized volume information, a is the weight corresponding to the normalized height information, and b is the weight corresponding to the normalized volume information, wherein a is a value between 0 and 1, b is a value between 0 and 1, and a+b=1.

[0009] In some embodiments, the second parameter for:

[0010] ,

[0011] in, is the normalized looseness score, is the normalized stickiness score, c is the weight corresponding to the normalized looseness score, and d is the weight corresponding to the normalized stickiness score, wherein c is a value between 0 and 1, d is a value between 0 and 1, and c+d=1.

[0012] In some embodiments, the ratio parameter for:

[0013] ,

[0014] in, is the first parameter, is the second parameter, e is the weight corresponding to the first parameter, and f is the weight corresponding to the second parameter, wherein e is a value between 0 and 1, f is a value between 0 and 1, and e+f=1.

[0015] In some embodiments, the integral parameter for:

[0016] ,

[0017] in, is the first parameter, is the second parameter, g is the weight corresponding to the first parameter, and h is the weight corresponding to the second parameter, wherein g is a value between 0 and 1, h is a value between 0 and 1, and g+h=1.

[0018] In some embodiments, obtaining the looseness score and the viscosity score of the material pile based on the temperature information and the image information of the material pile includes: inputting the temperature information and the image information of the material pile into a pre-trained transfer learning model, and outputting the looseness score and the viscosity score of the material pile through analysis by the transfer learning model.

[0019] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the boom, and the cavity pressure information of the boom includes the large cavity pressure information of the boom; calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator includes: calculating the pressure difference between the set large cavity pressure threshold of the boom and the large cavity pressure information of the boom; adjusting the angle of the actuator based on the angle and value includes: adjusting the angle of the boom based on the angle and value.

[0020] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the boom, and the cavity pressure information of the boom includes the small cavity pressure information of the boom; calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator includes: calculating the pressure difference between the set small cavity pressure threshold of the boom and the small cavity pressure information of the boom; adjusting the angle of the actuator based on the angle and value includes: adjusting the angle of the boom based on the angle and value.

[0021] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the bucket, and the cavity pressure information of the bucket includes the large cavity pressure information of the bucket; calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator includes: calculating the pressure difference between the set large cavity pressure threshold of the bucket and the large cavity pressure information of the bucket; adjusting the angle of the actuator based on the angle and value includes: adjusting the angle of the bucket based on the angle and value.

[0022] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the bucket, and the cavity pressure information of the bucket includes the small cavity pressure information of the bucket; calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator includes: calculating the pressure difference between the set small cavity pressure threshold of the bucket and the small cavity pressure information of the bucket; adjusting the angle of the actuator based on the angle and value includes: adjusting the angle of the bucket based on the angle and value.

[0023] According to another aspect of the present disclosure, a control device for an unmanned loader is provided, comprising: a data acquisition module for receiving three-dimensional point cloud data of a material pile and its surrounding environment detected by a radar and temperature information and image information of the material pile detected by a camera; a data processing module for obtaining, based on the three-dimensional point cloud data, information on the height at which a bucket cuts into the material pile and volume information of the material pile, obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile, and calculating a proportional parameter and an integral parameter based on the information on the height at which the bucket cuts into the material pile and the volume information, looseness score and viscosity score of the material pile; an auxiliary control module for calculating, based on the obtained cavity pressure of the actuator, information, calculates the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator, uses the proportional parameter and the integral parameter as parameters of the proportional-integral control algorithm, uses the pressure difference as input data of the proportional-integral control algorithm, and obtains the compensation angle through the proportional-integral control algorithm, wherein the cavity pressure information of the actuator includes at least one of the cavity pressure information of the boom and the cavity pressure information of the bucket; and a main control module, for calculating the angle and value between the compensation angle and the main control angle obtained by the main control algorithm, and outputting the angle and value to the vehicle controller, so that the vehicle controller adjusts the angle of the actuator so that the angle of the actuator conforms to the angle and value.

[0024] According to another aspect of the present disclosure, a control device for an unmanned loader is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the aforementioned control method based on instructions stored in the memory.

[0025] According to another aspect of the present disclosure, a control system for an unmanned loader is provided, comprising: a control device as described above; a radar for obtaining three-dimensional point cloud data of a material pile and its surrounding environment, and transmitting the three-dimensional point cloud data to the control device, wherein the control device obtains height information of a bucket cutting into the material pile and volume information of the material pile based on the three-dimensional point cloud data; a camera for obtaining temperature information and image information of the material pile, and transmitting the temperature information and the image information to the control device; and a pressure sensor for obtaining cavity pressure information of an actuator, and transmitting the cavity pressure information to the control device.

[0026] In some embodiments, the control system further includes: a vehicle controller for receiving the angle and value from the control device and adjusting the angle of the actuator according to the angle and value; wherein the control device is used to output the angle and value to the vehicle controller.

[0027] According to another aspect of the present disclosure, an unmanned loader is provided, comprising: the control system as described above.

[0028] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the control method as described above is implemented.

[0029] According to another aspect of the present disclosure, a computer program product is provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the control method as described above is implemented.

[0030] In the above control method, since the proportional parameters and integral parameters of the proportional-integral control algorithm are obtained by using the data information related to the material pile during the control process of the unmanned loader, and the pressure difference between the cavity pressure threshold and the cavity pressure information of the actuator is used as the input data of the proportional-integral control algorithm, a compensation angle is output, and the main control angle is compensated by using the compensation angle, so that the actuator is adjusted to a suitable angle so that the bucket can softly cut into the material pile. This enables the bucket to cut into the material pile more comprehensively considering the current situation of the material pile and the cavity pressure, thereby improving the efficiency and quality of the unmanned loader when shoveling materials.

[0031] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0033] The present disclosure can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0034] Figure 1 is a flow chart illustrating a control method for an unmanned loader according to some embodiments of the present disclosure;

[0035] Figure 2 is a flowchart illustrating a control method for an unmanned loader according to other embodiments of the present disclosure;

[0036] Figure 3 is a block diagram schematically illustrating a structure of a control device for an unmanned loader according to some embodiments of the present disclosure;

[0037] Figure 4 is a block diagram schematically illustrating a structure of a control device for an unmanned loader according to other embodiments of the present disclosure;

[0038] Figure 5 is a block diagram schematically illustrating a structure of a control device for an unmanned loader according to other embodiments of the present disclosure;

[0039] Figure 6 is a block diagram schematically illustrating a structure of a control system for an unmanned loader according to some embodiments of the present disclosure;

[0040] Figure 7 is a structural block diagram schematically illustrating a control system for an unmanned loader according to other embodiments of the present disclosure;

[0041] Figure 8 is a control module connection diagram illustrating a control method for an unmanned loader according to some embodiments of the present disclosure;

[0042] Figure 9 is a data flow diagram illustrating a control method for an unmanned loader according to some embodiments of the present disclosure;

[0043] Figure 10 3 is a schematic diagram showing the motion trajectory effect of the bucket of an unmanned loader in a control method for an unmanned loader according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0044] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure.

[0045] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0046] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0047] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0048] In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0049] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0050] Technical solutions in related technologies include: control methods based on preset path planning, such as using offline planning algorithms (for example, A* (A-Star Algorithm), DWA (Dynamic Window Approach, dynamic window method), etc.) to generate the bucket motion trajectory and achieve path tracking through encoders and gyroscopes; control methods based on a single sensor, such as visual guidance solutions (using cameras to identify material edges (for example, YOLOv5 (You Only Look Once, Version 5) algorithm)), solutions for controlling bucket alignment with lidar (using point cloud data to build a three-dimensional model, plan the cutting angle, force control solutions, and simple control based on torque feedback); control strategies based on traditional shaking shovels, such as fixed-frequency vibration (2-4 Hz (Hertz)) with a fixed flip angle, segmented control based on time parameters (for example, high-speed flipping at the beginning of unloading combined with low-speed compensation at the end), etc.

[0051] However, the known technology still has the following problems:

[0052] (1) Physiological limitations of manual operation. Traditional loader operators need to maintain constant attention to operate the multi-way valve group, and single-shift operation time generally exceeds 8 hours. Ergonomic studies have shown that after 2 hours of continuous operation, hand muscle fatigue increases by 37%, reaction delay reaches 0.4-0.7 seconds, resulting in bucket cutting error exceeding ±50mm.

[0053] (2) Difficulty adapting to unstructured environments. Characteristics of sticky materials: Wet clay has an internal friction angle of 35°-45° and a cohesion of 80-120 kPa (kilopascals), exceeding the internal friction angle of ordinary sand and gravel (internal friction angle 25°-35°, cohesion <10 kPa). When encountering sticky materials, the control strategy based on a preset entry angle (usually 30°-45°) causes the bucket resistance to surge by over 300%, resulting in a 40% increase in hydraulic system overflow losses.

[0054] (3) Bottlenecks in unmanned control technology. ① Lack of flexible cutting: Unmanned systems in related technologies mostly use a fixed cutting speed. When encountering a sudden change in material hardness, the system cannot dynamically adjust the cutting angle, resulting in a 57% increase in the risk of bucket stress concentration exceeding the material yield strength (e.g., 460 MPa). ② Rough shaking shovel strategy: Unloading in related technologies relies on fixed-frequency vibration (e.g., 2-3 Hz), without considering the material's adhesion characteristics.

[0055] In summary, in the related art, the efficiency and quality of unmanned loaders when shoveling materials are low.

[0056] In view of this, an embodiment of the present disclosure provides a control method for an unmanned loader to improve the efficiency and quality of the unmanned loader when shoveling materials.

[0057] Figure 1 1 is a flow chart showing a control method for an unmanned loader according to some embodiments of the present disclosure. The control method may be executed by a control device. Figure 1 As shown, the control method includes steps S102 to S114.

[0058] In step S102, data information related to the material pile and cavity pressure information of the actuator are obtained, wherein the data information includes: height information of the bucket cutting into the material pile and volume information, temperature information and image information of the material pile; the cavity pressure information of the actuator includes at least one of the cavity pressure information of the boom and the cavity pressure information of the bucket.

[0059] The required data information can be collected through sensors installed on the unmanned loader.

[0060] For example, radar (e.g., lidar and millimeter-wave radar) can be used to obtain information about the material pile, its surroundings, and its shape. The lidar can generate three-dimensional point cloud data of the material pile and its surroundings, and transmit this data to a control device. Based on this data, the control device can determine information such as the height at which the bucket penetrated the material pile and its volume. The lidar can obtain three-dimensional point cloud information of the material pile and its surroundings, thereby determining its shape. Furthermore, the millimeter-wave radar can obtain information such as the distance and angle between the radar and the material pile.

[0061] For example, a camera can be used to obtain information such as material surface temperature, material type, volume, and image information. The camera can obtain temperature and image information of the material pile and transmit these information to a control device. The camera can include a thermal imaging camera and a binocular camera. The thermal imaging camera can obtain surface temperature information of the material pile, while the binocular camera can obtain image information of the material pile. Furthermore, after obtaining the image information, a deep learning algorithm can be used to identify the material type and volume of the material pile.

[0062] For example, a pressure sensor can be used to obtain cavity pressure information of an actuator and transmit the cavity pressure information to a control device. For example, a pressure sensor can obtain pressure information of the large cavity and / or small cavity of a loader's boom, as well as the large cavity and / or small cavity of a bucket.

[0063] In step S104 , a looseness score and a viscosity score of the material pile are obtained based on the temperature information and image information of the material pile.

[0064] In some embodiments, step S104 includes: inputting the temperature information and image information of the material pile into a pre-trained transfer learning model, and outputting the looseness score and viscosity score of the material pile through analysis by the transfer learning model.

[0065] Here, the transfer learning model is a pre-trained and annotated model. For example, this transfer learning model is based on a Residual Network (ResNet). For example, sample data of temperature and image information of a material pile can be used as input data, and the material type (e.g., bulkiness and viscosity) and volume information can be used as annotations to train the transfer learning model. This pre-trained ResNet-based transfer learning model can then be used to analyze material type (e.g., bulkiness and viscosity) and estimate the volume of the material pile.

[0066] Therefore, based on the pre-trained transfer learning model, after the temperature information and image information of the material pile are input into the transfer learning model, the transfer learning model outputs the looseness score and viscosity score of the material pile through analysis.

[0067] ResNet is a deep convolutional neural network with excellent performance. It can effectively process image data and, through residual connections, address the vanishing gradient and degradation issues that arise with increasing network depth. Before training, the collected data is labeled and classified, for example, by assigning different degrees of looseness to a scale of 1-10. Then, the data is scored based on the real-time image data to obtain a looseness score. The process for obtaining the stickiness score is similar.

[0068] In step S106 , a proportional parameter and an integral parameter are calculated based on the height information of the bucket penetrating the material pile, the volume information of the material pile, the looseness score, and the viscosity score.

[0069] In some embodiments, step S106 includes normalizing the height information of the bucket entering the material pile, the volume information of the material pile, the looseness score, and the viscosity score to obtain normalized height information, normalized volume information, normalized looseness score, and normalized viscosity score. The following describes the normalization process for these data.

[0070] For example, find the highest point of a material pile in a 3D environment map built by dual radars and the lowest point , the maximum height of the material pile That is, the maximum value of the height information of the bucket cutting into the material pile, and the minimum value of the height of the material pile This is the minimum value of the height information of the bucket cutting into the material pile, for example, It can be 0. Normalize the height data to:

[0071] , (1)

[0072] in, is the normalized height information, H is the height information of the bucket cutting into the material pile, The maximum value of the height information of the bucket cutting into the material pile, The minimum height information for the bucket to cut into the material pile.

[0073] During the loading process, the height of the material pile may change. Each time the control method of the present disclosure is executed, that is, before each subsequent compensation angle is calculated, the maximum value of the height information of the bucket cutting into the material pile may be remeasured and calculated and brought into the above-mentioned relational formula (1).

[0074] The volume of the material pile is estimated by combining lidar point cloud data with camera images , normalize the volume data to:

[0075] , (2)

[0076] in, is the normalized volume information, V is the current volume of the material pile, is the maximum volume of the material pile, is the minimum volume of the material pile, for example, is 0.

[0077] During the shoveling process, the volume of the material pile may change. Each time the control method of the present disclosure is executed, that is, before each subsequent compensation angle is calculated, the maximum volume of the material pile may be remeasured and calculated and brought into the above-mentioned relational expression (2).

[0078] The material type is analyzed through the ResNet-based transfer learning model to obtain the looseness score of the material , normalize the looseness score data to:

[0079] , (3)

[0080] in, is the normalized looseness score, L is the looseness score output by the transfer learning model, is the maximum value of the looseness score corresponding to the type and volume of the current material pile, It is the minimum value of the looseness score corresponding to the type and volume of the current material pile.

[0081] Here, through the transfer learning model, the type information and volume information of the material pile can be analyzed. Based on the type information and volume information of the material pile, the maximum and minimum values of the looseness score of the material pile can be obtained according to the pre-known empirical data, and used as the maximum and minimum values of the looseness score L output by the model.

[0082] Obtain the viscosity score of the material through the ResNet-based transfer learning model , normalize the stickiness score data to:

[0083] , (4)

[0084] in, is the normalized stickiness score, S is the stickiness score output by the transfer learning model, is the maximum value of the viscosity score corresponding to the type and volume of the current material pile, It is the minimum value of the viscosity score corresponding to the type and volume of the current material pile.

[0085] Here, through the transfer learning model, the type information and volume information of the material pile can be analyzed. Based on the type information and volume information of the material pile, the maximum and minimum values of the viscosity score of the material pile can be obtained according to the pre-known empirical data, which can be used as the maximum and minimum values of the model output viscosity score S.

[0086] In some embodiments, the above step S106 further includes: calculating a first parameter based on the normalized height information and the normalized volume information, wherein the first parameter is a comprehensive parameter of the height and volume of the material pile.

[0087] For example, the first parameter for:

[0088] , (5)

[0089] in, is the normalized height information, is the normalized volume information, a is the weight corresponding to the normalized height information (which may be called the first weight), and b is the weight corresponding to the normalized volume information (which may be called the second weight). Here, a is a value between 0 and 1, b is a value between 0 and 1, and a + b = 1. Here, a and b are weight parameters that can be adjusted according to actual application conditions.

[0090] In some embodiments, the above step S106 further includes: calculating a second parameter based on the normalized bulkiness score and the normalized viscosity score, wherein the second parameter is a comprehensive parameter of the bulkiness and viscosity of the material pile.

[0091] For example, the second parameter for:

[0092] , (6)

[0093] in, is the normalized looseness score, is the normalized stickiness score, c is the weight corresponding to the normalized looseness score (which can be called the third weight), and d is the weight corresponding to the normalized stickiness score (which can be called the fourth weight). c is a value between 0 and 1, d is a value between 0 and 1, and c + d = 1. Here, c and d are weight parameters that can be adjusted according to actual application conditions.

[0094] In some embodiments, the above step S106 further includes: calculating a proportional parameter and an integral parameter based on the first parameter and the second parameter.

[0095] For example, the scale parameter for:

[0096] , (7)

[0097] in, is the first parameter, is the second parameter, e is the weight corresponding to the first parameter (which can be called the fifth weight), and f is the weight corresponding to the second parameter (which can be called the sixth weight), where e is a value between 0 and 1, f is a value between 0 and 1, and e + f = 1. Here, e and f are weight parameters that can be adjusted according to actual application conditions.

[0098] For example, the integral parameter for:

[0099] , (8)

[0100] in, is the first parameter, is the second parameter, g is the weight corresponding to the first parameter (which can be called the seventh weight), and h is the weight corresponding to the second parameter (which can be called the eighth weight), where g is a value between 0 and 1, h is a value between 0 and 1, and g + h = 1. Here, g and h are weight parameters that can be adjusted according to actual application conditions.

[0101] In this way, the coefficients corresponding to different material forms and types are calculated based on the information obtained by the radar and camera, and summarized into a parameter reference table (including proportional parameters and integral parameters) as the basis for parameter selection of the proportional integral control algorithm (i.e., PI control algorithm).

[0102] In step S108, the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator is calculated. In this step, the pressure difference = cavity pressure threshold - cavity pressure information of the actuator.

[0103] Here, the pre-set cavity pressure threshold (also called the pressure overlimit threshold) is based on the actual engineering structure range. For example, the specific measurement information for a loader is that the range of the boom's large cavity pressure is generally 0-40 MPa, and the range of the small cavity pressure is generally 0-30 MPa. Under full load conditions, the large cavity pressure is approximately 25-32 MPa when the boom is raised, and the small cavity pressure is approximately 18-25 MPa during unloading. The range of the loader bucket's large cavity pressure is generally 0-22 MPa, and the range of the small cavity pressure is generally 0-18 MPa. Under full load conditions, the peak large cavity pressure is approximately 25 MPa when the bucket is raised, and the small cavity pressure is approximately 15 MPa during unloading. According to the above-mentioned measuring range and the collected material information, the cavity pressure threshold is set. For example, the threshold set in the algorithm is that the cavity pressure of the boom (such as the upper arm) is 30 MPa. When the pressure of the boom exceeds the current threshold, the angle is compensated to control the lowering of the boom, realize the boom pressure relief control, and shovel the material at a reasonable angle.

[0104] It should be noted that the cavity pressure threshold can be set according to actual conditions, and the scope of this disclosure is not limited to the specific value of the cavity pressure threshold.

[0105] In some embodiments, the actuator's cavity pressure information includes the boom's cavity pressure information, which in turn includes the boom's large cavity pressure information. In such cases, the cavity pressure threshold includes the boom's large cavity pressure threshold. Step S108 includes calculating the pressure difference between the boom's large cavity pressure threshold and the boom's large cavity pressure information. This achieves the purpose of calculating the pressure difference corresponding to the boom's large cavity, thereby facilitating subsequent adjustment of the boom's angle.

[0106] In other embodiments, the actuator's cavity pressure information includes the boom's cavity pressure information, which in turn includes the boom's small cavity pressure information. In such cases, the cavity pressure threshold includes the boom's small cavity pressure threshold. Step S108 includes calculating the pressure difference between the set boom's small cavity pressure threshold and the boom's small cavity pressure information. This achieves the purpose of calculating the pressure difference corresponding to the boom's small cavity, thereby facilitating subsequent adjustment of the boom's angle.

[0107] In other embodiments, the actuator's cavity pressure information includes the bucket's cavity pressure information, which in turn includes the bucket's large cavity pressure information. In such cases, the cavity pressure threshold includes the bucket's large cavity pressure threshold. Step S108 includes calculating the pressure difference between the bucket's large cavity pressure threshold and the bucket's large cavity pressure information. This achieves the purpose of calculating the pressure difference corresponding to the bucket's large cavity, thereby facilitating subsequent adjustment of the bucket's angle.

[0108] In other embodiments, the actuator's cavity pressure information includes the bucket's cavity pressure information, which in turn includes the bucket's small cavity pressure information. In such cases, the cavity pressure threshold includes the bucket's small cavity pressure threshold. Step S108 includes calculating the pressure difference between the set bucket's small cavity pressure threshold and the bucket's small cavity pressure information. This achieves the purpose of calculating the pressure difference corresponding to the bucket's small cavity, thereby facilitating subsequent adjustment of the bucket's angle.

[0109] In step S110, the proportional parameter and the integral parameter are used as parameters of the proportional-integral control algorithm, the pressure difference is used as input data of the proportional-integral control algorithm, and the compensation angle is obtained through the proportional-integral control algorithm.

[0110] Here, the proportional-integral control algorithm, also known as the PI-assisted control algorithm, utilizes the proportional and integral components of the PID control logic while ignoring the differential component. This algorithm uses the proportional parameter (i.e., the P parameter reference table) and the integral parameter (i.e., the I parameter reference table) calculated in step S106 as the proportional and integral parameters of the control algorithm. The pressure difference calculated in step S108 serves as the input data for the proportional-integral control algorithm, and the output data of the algorithm serves as the compensation angle.

[0111] In other words, the PI-assisted control algorithm uses the obtained P parameter reference table and I parameter reference table as parameter inputs for the PI-assisted control algorithm, uses the cavity pressure threshold as the set threshold standard value, and uses the difference between the cavity pressure threshold and the pressure information collected by the pressure sensor as the input data for the PI-assisted control algorithm. This implements a PID (Proportional-Integral-Derivative Control) control method and generates a compensation angle. In other words, the proportional-integral control algorithm used in the embodiments of the present disclosure is a special PID control algorithm.

[0112] In step S112, the angle sum between the compensation angle and the main control angle obtained by the main control algorithm is calculated. That is, the angle sum = compensation angle + main control angle.

[0113] For example, the main control algorithm may adopt an MPC (Model Predictive Control) algorithm. For example, angle sensors may be provided at the boom and / or bucket to detect the angles of the boom and / or bucket. For example, the angle sensor at the boom transmits the detected boom angle to the control device, and the angle sensor at the bucket transmits the detected bucket angle to the control device. The control device may plan the next angle of the boom based on the boom angle and through the main control algorithm. The control device may also plan the next angle of the bucket based on the bucket angle and through the main control algorithm. Here, the angle of the actuator (boom and / or bucket) planned by the main control algorithm is the above-mentioned main control angle. The main control algorithm is a known algorithm used by unmanned loaders, and therefore, it will not be described in detail here.

[0114] For the angle control of the boom and bucket, the main control algorithm MPC adopts a decoupled control method to perform separate angle control.

[0115] In step S114 , the angle of the actuator is adjusted based on the angle and value so that the angle of the actuator conforms to the angle and value.

[0116] In some embodiments, the actuator cavity pressure information includes the boom cavity pressure information, which includes the boom's large cavity pressure information. In such cases, step S114 includes adjusting the boom angle based on the angle and value. In other words, the boom angle is adjusted based on the collected boom large cavity pressure information.

[0117] In other embodiments, the actuator cavity pressure information includes the boom cavity pressure information, and the boom cavity pressure information includes the boom's small cavity pressure information. In such cases, step S114 includes adjusting the boom angle based on the angle and value. In other words, the boom angle is adjusted based on the collected boom small cavity pressure information.

[0118] In other embodiments, the actuator cavity pressure information includes bucket cavity pressure information, and the bucket cavity pressure information includes the bucket's bulk cavity pressure information. In such cases, step S114 includes adjusting the bucket angle based on the angle and value. In other words, the bucket angle is adjusted based on the collected bucket bulk cavity pressure information.

[0119] In other embodiments, the actuator cavity pressure information includes bucket cavity pressure information, which includes the bucket's small cavity pressure information. In such cases, step S114 includes adjusting the bucket angle based on the angle and value. In other words, the bucket angle is adjusted based on the collected bucket small cavity pressure information.

[0120] For the angle control of the boom and bucket, the main control algorithm MPC adopts a decoupled control method to perform separate angle control. In addition, the auxiliary angle control implemented also uses two separate sets of PI auxiliary control angle compensation algorithms to achieve reasonable soft cutting for different situations.

[0121] In some embodiments, the above step S114 includes: outputting the angle and value to the vehicle controller, so that the vehicle controller adjusts the angle of the actuator so that the angle of the actuator conforms to the angle and value.

[0122] Thus, a control method for an unmanned loader according to some embodiments of the present disclosure is provided. The control method includes: obtaining data information related to a material pile and cavity pressure information of an actuator, wherein the data information includes: information on the height at which a bucket cuts into the material pile, as well as volume information, temperature information, and image information of the material pile; the cavity pressure information of the actuator includes at least one of cavity pressure information of the boom and cavity pressure information of the bucket; obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile; calculating a proportional parameter and an integral parameter based on the information on the height at which the bucket cuts into the material pile, as well as the volume information, looseness score, and viscosity score of the material pile; calculating a pressure difference between a set cavity pressure threshold and the cavity pressure information of the actuator; using the proportional parameter and the integral parameter as parameters of a proportional-integral control algorithm and the pressure difference as input data of the proportional-integral control algorithm to obtain a compensation angle through the proportional-integral control algorithm; calculating an angle sum between the compensation angle and a main control angle obtained by a main control algorithm; and adjusting the angle of the actuator based on the angle sum so that the actuator angle conforms to the angle sum. In this control method, during the control process of the unmanned loader, the proportional parameters and integral parameters of the proportional-integral control algorithm are obtained by using data information related to the material pile, and the pressure difference between the cavity pressure threshold and the cavity pressure information of the actuator is used as the input data of the proportional-integral control algorithm, thereby outputting a compensation angle, and using the compensation angle to compensate for the main control angle, so that the actuator is adjusted to a suitable angle so that the bucket softly cuts into the material pile. This enables the bucket to cut into the material pile more comprehensively considering the current situation of the material pile and the cavity pressure, thereby improving the efficiency and quality of the unmanned loader when shoveling materials.

[0123] It should be noted that the auxiliary angle compensation is performed during the shoveling process, and in the process of transporting materials while maintaining the bucket angle, in order to avoid abnormal unloading, the compensation process may not be involved in the control.

[0124] Figure 2 FIG. 1 is a flow chart showing a control method for an unmanned loader according to other embodiments of the present disclosure. Figure 2As shown, the control method includes steps S201 to S207.

[0125] In step 201, the sensor is initialized and calibrated, and preset operating parameters are loaded.

[0126] That is, the required sensors (eg, radar, camera, and / or IMU (Inertial Measurement Unit)) are initialized and calibrated, and preset operating parameters are loaded.

[0127] In step 202, the material pile is scanned by radar perception, and the material attributes are identified by camera. That is, the material pile is scanned and identified by radar and camera to obtain specific information and other attributes of the material pile.

[0128] In step 203, a reference parameter matching table is prepared based on the prior information of different materials. That is, based on the acquired information, a calculation formula is used to generate a corresponding parameter reference matching table.

[0129] In step 204, the pressure feedback of the loader arm and the pressure feedback of the bucket are collected in real time. For example, the pressure feedback of the loader arm and the pressure feedback of the bucket are collected in real time.

[0130] In step 205, the angle compensation is achieved by using the pressure feedback-assisted control program.

[0131] For example, pressure feedback and set pressure thresholds are used as input to an auxiliary control program to generate angle compensation for the unmanned loader's boom and bucket.

[0132] In step 206, the main control program plus the auxiliary control program realizes the control of the angle.

[0133] That is, the control of the boom and bucket of the unmanned loader is achieved by the angle control output by the main control program plus the angle compensation of the auxiliary control program.

[0134] In step 207, the unmanned loader implements adaptive soft cutting for different materials.

[0135] Thus, a control method for an unmanned loader according to other embodiments of the present disclosure is provided, which can improve the efficiency, stability, and robustness of the unmanned loader in automatic shoveling.

[0136] Embodiments of the present disclosure provide an intelligent, assisted shoveling control method for an unmanned loader based on pressure feedback and multimodal sensing, relating to intelligent, assisted shoveling control technology for unmanned loaders. This method primarily involves: utilizing multimodal sensor fusion information (real-time reading of information from the unmanned loader's radar, camera, and pressure sensors) to collect relevant information about the loading point; analyzing this integrated information to generate a corresponding parameter reference table for different materials and loading situations, thereby adaptively adjusting control parameters; using the resulting parameter reference table, compensating the unmanned loader's angle control through PI control and parameter adjustment, providing real-time incremental changes to improve the unmanned loader's adaptability to different situations during the shoveling process, reduce unnecessary capacity loss, and maximize shoveling utilization; finally, combining it with advanced control algorithms (e.g., MPC, PID, etc.) to achieve appropriate soft cut-in, making the shoveling cut-in more flexible and precise. This method can improve the efficiency, stability, and robustness of unmanned loaders during automated shoveling.

[0137] Figure 3 FIG. 1 is a block diagram schematically illustrating a control device for an unmanned loader according to some embodiments of the present disclosure. Figure 3 As shown, the control device includes a data acquisition module 301 , a data processing module 302 , an auxiliary control module 303 and a main control module 304 .

[0138] The data acquisition module 301 is used to receive the three-dimensional point cloud data of the material pile and the surrounding environment detected by the radar and the temperature information and image information of the material pile detected by the camera.

[0139] The data processing module 302 is used to obtain the height information of the bucket cutting into the material pile and the volume information of the material pile based on the three-dimensional point cloud data, obtain the looseness score and viscosity score of the material pile based on the temperature information and image information of the material pile, and calculate the proportional parameter and the integral parameter based on the height information of the bucket cutting into the material pile and the volume information, looseness score and viscosity score of the material pile.

[0140] In some embodiments, the data processing module 302 is used to normalize the height information of the bucket cutting into the material pile, as well as the volume information, looseness score and viscosity score of the material pile, to obtain normalized height information, normalized volume information, normalized looseness score and normalized viscosity score; based on the normalized height information and normalized volume information, a first parameter is calculated, wherein the first parameter is a comprehensive parameter of the height and volume of the material pile; based on the normalized looseness score and the normalized viscosity score, a second parameter is calculated, wherein the second parameter is a comprehensive parameter of the looseness and viscosity of the material pile; and based on the first parameter and the second parameter, a proportional parameter and an integral parameter are calculated.

[0141] In some embodiments, the data processing module 302 is used to input the temperature information and image information of the material pile into a pre-trained transfer learning model, and output the looseness score and viscosity score of the material pile through analysis by the transfer learning model.

[0142] The auxiliary control module 303 is used to calculate the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator based on the obtained cavity pressure information of the actuator, use the proportional parameter and the integral parameter as parameters of the proportional-integral control algorithm, use the pressure difference as input data of the proportional-integral control algorithm, and obtain the compensation angle through the proportional-integral control algorithm, wherein the cavity pressure information of the actuator includes at least one of the cavity pressure information of the boom and the cavity pressure information of the bucket.

[0143] In some embodiments, the actuator cavity pressure information includes the boom cavity pressure information, and the boom cavity pressure information includes the boom large cavity pressure information. The auxiliary control module 303 is configured to calculate a pressure difference between a set boom large cavity pressure threshold and the boom large cavity pressure information.

[0144] In some embodiments, the actuator cavity pressure information includes the boom cavity pressure information, which includes the boom small cavity pressure information. The auxiliary control module 303 is configured to calculate a pressure difference between a set boom small cavity pressure threshold and the boom small cavity pressure information.

[0145] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the bucket, and the cavity pressure information of the bucket includes the large cavity pressure information of the bucket. The auxiliary control module 303 is used to calculate the pressure difference between the set large cavity pressure threshold of the bucket and the large cavity pressure information of the bucket.

[0146] In some embodiments, the cavity pressure information of the actuator includes the cavity pressure information of the bucket, and the cavity pressure information of the bucket includes the small cavity pressure information of the bucket. The auxiliary control module 303 is used to calculate the pressure difference between the set small cavity pressure threshold of the bucket and the small cavity pressure information of the bucket.

[0147] The main control module 304 is used to calculate the angle sum between the compensation angle and the main control angle obtained by the main control algorithm, and output the angle sum to the vehicle controller so that the vehicle controller adjusts the angle of the actuator so that the angle of the actuator conforms to the angle sum.

[0148] In some embodiments, the actuator cavity pressure information includes the boom cavity pressure information, and the boom cavity pressure information includes the boom large cavity pressure information or the boom small cavity pressure information. The main control module 304 is configured to output the angle sum value to the vehicle controller, so that the vehicle controller adjusts the boom angle so that the boom angle conforms to the angle sum value.

[0149] In some embodiments, the actuator cavity pressure information includes bucket cavity pressure information, and the bucket cavity pressure information includes large cavity pressure information or small cavity pressure information of the bucket. The main control module 304 is configured to output the angle and value to the vehicle controller, so that the vehicle controller adjusts the bucket angle so that the bucket angle conforms to the angle and value.

[0150] So far, a control device for an unmanned loader according to some embodiments of the present disclosure is provided. The control device includes: a data acquisition module, a data processing module, an auxiliary control module and a main control module. In the control device, since in the control process of the unmanned loader, the proportional parameters and integral parameters of the proportional integral control algorithm are obtained by using the data information related to the material pile, and the pressure difference between the cavity pressure threshold and the cavity pressure information of the actuator is used as the input data of the proportional integral control algorithm, a compensation angle is output, and the main control angle is compensated by using the compensation angle, so that the actuator is adjusted to a suitable angle so that the bucket softly cuts into the material pile, which enables the bucket to cut into the material pile more comprehensively considering the current situation of the material pile and the cavity pressure, thereby improving the efficiency and quality of the unmanned loader when shoveling materials.

[0151] Figure 4 4 is a block diagram schematically illustrating a control device for an unmanned loader according to some other embodiments of the present disclosure. The control device includes a memory 410 and a processor 420.

[0152] The memory 410 can be a disk, a flash memory or any other non-volatile storage medium. Figure 1 and / or Figure 2 The instructions in the corresponding embodiment.

[0153] The processor 420 is coupled to the memory 410 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 420 is used to execute instructions stored in the memory, thereby improving the efficiency and quality of the unmanned loader when shoveling materials.

[0154] In one embodiment, it is also possible to Figure 5As shown, the control device 500 includes a memory 510 and a processor 520. The processor 520 is coupled to the memory 510 via a BUS 530. The control device 500 can also be connected to an external storage device 550 via a storage interface 540 to access external data, and can also be connected to a network or another computer system (not shown) via a network interface 560, which will not be described in detail here.

[0155] In this embodiment, the memory stores data instructions, and the processor processes the instructions, thereby improving the efficiency and quality of the unmanned loader when shoveling materials.

[0156] In some embodiments, an electronic device for intelligent assisted shoveling control of an unmanned loader based on pressure feedback and multimodal perception is also provided, comprising: one or more processors, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to implement any step of the above-mentioned intelligent assisted control shoveling method of the unmanned loader: obtaining three-dimensional point cloud, image, pressure and other information of the material; forming a parameter reference table of the intelligent assisted PI control algorithm based on the calculated material height, volume, viscosity and looseness; deploying and implementing the control of the main control algorithm and the auxiliary PI control algorithm; collecting sensor data information transmission and issuing commands to the control program; and receiving and executing the control instructions by the vehicle controller.

[0157] Figure 6 is a block diagram schematically illustrating a structure of a control system for an unmanned loader according to some embodiments of the present disclosure.

[0158] like Figure 6 As shown, the control system includes a control device 610. For example, the control device 610 is as follows Figure 3 、 Figure 4 or Figure 5 Controls shown.

[0159] like Figure 6 As shown, the control system also includes a radar 601. Radar 601 is used to obtain three-dimensional point cloud data of the material pile and its surrounding environment and transmit the 3D point cloud data to a control device 610. Based on the 3D point cloud data, control device 610 can obtain information about the height at which the bucket penetrates the material pile and the volume of the material pile. For example, the radar may include a laser radar.

[0160] like Figure 6 As shown, the control system further includes a camera 602. The camera 602 is used to obtain temperature information and image information of the material pile and transmit the temperature information and image information to the control device 610. For example, the camera can include a thermal imaging camera and a binocular camera.

[0161] like Figure 6 As shown, the control system further includes a pressure sensor 603. The pressure sensor 603 is used to obtain cavity pressure information of the actuator and transmit the cavity pressure information to the control device 610.

[0162] Thus, a control system for an unmanned loader according to some embodiments of the present disclosure is provided. The control system includes a control device, a radar, a camera, and a pressure sensor. This control system can improve the efficiency and quality of the unmanned loader when shoveling materials.

[0163] In some embodiments, as Figure 6 As shown, the control system may further include a vehicle controller 606. The vehicle controller 606 is configured to receive the angle and value from the control device 610 and adjust the angle of the actuator according to the angle and value. Here, the control device 610 is configured to output the angle and value to the vehicle controller.

[0164] Figure 7 is a structural block diagram schematically illustrating a control system for an unmanned loader according to some other embodiments of the present disclosure.

[0165] In some embodiments, as Figure 7 As shown, a combination of a mechanical lidar and a solid-state lidar mounted on the roof of an unmanned loader can be used to construct a three-dimensional environmental map using the point cloud data collected by the radar, and to identify information such as the shape of the material pile; a thermal imaging camera mounted on the roof of the unmanned loader and near-infrared cameras and global shutter cameras on both sides of the vehicle body are used, and a pre-trained ResNet-based transfer learning model is deployed to analyze the material type (such as looseness and viscosity) and estimate the volume information of the material; pressure sensors mounted on the arm and bucket of the unmanned loader are used to measure the pressure information of the large cavity of the arm, the small cavity of the arm, the large cavity of the bucket, and the small cavity of the bucket.

[0166] like Figure 7As shown, the data acquisition module 301 in the control device 610 collects the required information from the radar 601 and camera 602 and then transmits it to the data processing module 302 for data processing. The radar point cloud information is denoised and stitched together to generate a three-dimensional point cloud map. The camera image information is trained using a deep learning-based object detection model to generate information such as the material volume. The first communication module 305 of the control device 610 converts the processed information into corresponding information and transmits it to the auxiliary control module 303 and the main control module 304. This information includes collected sensor information and a generated parameter reference table. The main control module 304 performs primary control of the unmanned loader's boom and bucket, while the auxiliary control module 303 uses feedback pressure information to perform corresponding angle compensation. Finally, the second communication module 306 of the control device 610 transmits control command information to the vehicle controller 606, which controls the actuators (e.g., boom and bucket) 620 of the unmanned loader.

[0167] In the above embodiment, the control device 610 of the control system includes, in addition to the data acquisition module 301, the data processing module 302, the auxiliary control module 303, and the main control module 304, a first communication module 305 and a second communication module 306. The first communication module 305 is used to transmit data, and the second communication module 306 is used to transmit data or commands.

[0168] In the above embodiment, the data acquisition module is used to obtain relevant data information from sensors such as radar, camera and / or IMU; the data processing module is used to perform processing operations such as filtering and normalization on the data based on the collected data information; the communication module is responsible for transmitting the processed information and control commands to different modules; the main control module and the auxiliary control module (i.e., the PI auxiliary control module) are used to control the angles of the boom and bucket of the unmanned loader; the vehicle controller is used to receive control commands and execute control actions such as lowering the arm, shoveling, raising the arm, and unloading.

[0169] Furthermore, in the data acquisition module, the information acquired by each sensor, including the three-dimensional environment map information, the material pile shape information, and the vehicle body tilt information, is transmitted to the data processing module.

[0170] Furthermore, in the data processing module, the collected information is preliminarily analyzed to determine the volume, location, type of the material, and the distance between the loader and the material. All the collected data information is integrated into a parameter table (including proportional parameters and integral parameters) through a normalized proportion formula, which serves as a parameter reference for the auxiliary control module.

[0171] Furthermore, in the communication module, the first communication module is responsible for transmitting the processed data to the main control module and the auxiliary control module of the vehicle, and the second communication module is responsible for sending the control command of the control program to the vehicle controller.

[0172] Furthermore, the auxiliary control module takes the large chamber pressure of the bucket and boom as input, and uses the proportional parameter and integral parameter obtained by the data processing module as the parameter reference of PI. The optimal compensation angle is determined by actual debugging of the P and I parameters, and the calculated optimal compensation angle is added to the control angle output by the main control algorithm model predictive control (MPC) module, so that the boom and bucket can be raised or lowered according to different situations to alleviate the problem of mechanical damage or rigid cutting caused by excessive pressure in the loader's pressure valve exceeding the threshold when the material is too hard, too large or too wet.

[0173] The intelligent assisted shoveling control method and system for an unmanned loader based on pressure feedback and multimodal perception provided by the embodiments of the present disclosure take into account the problems of rigid cutting-in during autonomous shoveling operations by the unmanned loader, which may lead to mechanical damage and reduced operating efficiency. By utilizing the fusion information of sensors such as radar, camera, and pressure sensor equipped on the unmanned loader, a corresponding parameter reference table is adaptively generated; and based on the pressure characteristics of the unmanned loader shoveling, a PI assisted control angle compensation strategy is constructed, and the real-time changing angle increment and the angle control value output by the main control algorithm MPC are added to achieve reasonable soft cutting-in, making the shoveling cutting more flexible and delicate, and improving the adaptability of the unmanned loader to different situations and the shoveling utilization rate during the shoveling process.

[0174] Figure 8 is a control module connection diagram illustrating a control method for an unmanned loader according to some embodiments of the present disclosure.

[0175] like Figure 8 As shown, module 801 is the control demand information acquisition module, which is responsible for collecting the angle feedback of the unmanned loader's boom and bucket, as well as radar information and camera information, and then transmits the data to modules 804 and 805. Module 804 also needs to receive the planning angle information 802 from the planning algorithm to achieve smooth and reasonable shoveling movements of the unmanned loader. Module 805 needs to receive the pressure information 803 transmitted by the pressure sensors on the loader's boom and bucket to implement compensation control of the loader's boom and bucket angles when encountering irregular or hard materials. Finally, the control angle output by module 804 and the compensation angle output by module 805 are added and transmitted to module 806, which then implements the actual control of the loader.

[0176] Figure 9 is a data flow diagram illustrating a control method for an unmanned loader according to some embodiments of the present disclosure.

[0177] like Figure 9 As shown, module 901 is the information acquisition module required by the auxiliary control module, responsible for collecting radar and camera information. By analyzing and calculating the information in module 901, module 902 generates a P parameter reference table (i.e., proportional parameters), and module 903 generates an I parameter reference table (i.e., integral parameters). Module 904 collects pressure sensor information from the unmanned loader, combines the information from the parameter reference tables in modules 902 and 903, and transmits it to module 905. Module 906, similarly analyzing the information in module 901, transmits the generated pressure overlimit threshold information to module 905. In module 905, all the acquired information is subjected to a PID control algorithm consisting only of P and I parameters. The final output is the control compensation angle, which is transmitted to module 907, completing the auxiliary PI control process.

[0178] Figure 10 3 is a schematic diagram showing the motion trajectory effect of the bucket of an unmanned loader in a control method for an unmanned loader according to some embodiments of the present disclosure.

[0179] like Figure 10 As shown, step 1001 shows the initial state of the unmanned loader, with the bucket flush against the ground and shoveling material. The process then proceeds to step 1002, where the material is shoveled. However, if the shoveling encounters excessive resistance, the intelligent auxiliary shoveling control module takes effect, applying a negative compensation angle to the bucket (increasing the bucket angle upward), causing the bucket to proceed to step 1003, releasing pressure. Once the pressure is released below the overload threshold, the main control program continues to control the bucket's upward movement, proceeding to step 1004. If significant resistance is still encountered during operation, steps 1002 to 1003 are repeated to achieve a reasonable soft entry into the material.

[0180] In some embodiments of the present disclosure, an unmanned loader is further provided, which includes the control system as described above.

[0181] In some embodiments of the present disclosure, a computer program product is further provided. The computer program product includes a computer program or instructions. When the computer program or instructions are executed by a processor, the control method as described above is implemented.

[0182] In some embodiments, the present disclosure further provides a computer-readable storage medium (eg, a non-transitory computer-readable storage medium) having computer program instructions stored thereon. When the instructions are executed by a processor, the computer program instructions are executed. Figure 1 and / or Figure 2The steps of the method in the corresponding embodiment. Those skilled in the art will understand that the embodiments of the present disclosure can be provided as methods, apparatuses, or computer program products. Therefore, the present disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure can take the form of a computer program product implemented on one or more computer-usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0183] An embodiment of the present disclosure also provides a non-transitory computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement any step of the intelligent auxiliary control shoveling method of the unmanned loader described above: obtaining three-dimensional point cloud, image, pressure and other information of the material; forming a parameter reference table of the intelligent auxiliary PI control algorithm based on the calculated material height, volume, viscosity and looseness; deploying and implementing the control of the main control algorithm and the auxiliary PI control algorithm; collecting sensor data information transmission and issuing commands to the control program; and receiving and executing the control instructions by the vehicle controller.

[0184] The technical solution disclosed in the present invention is an improvement to the flexible shoveling of unmanned loaders. The compensation angle obtained by the intelligent auxiliary control shoveling method is added to the angle control value output by the main control algorithm MPC to achieve reasonable soft cutting, making the shoveling cutting more flexible and delicate, and improving the adaptability of the unmanned loader to different situations and the shoveling utilization rate during the shoveling process.

[0185] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0186] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0187] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0188] The present disclosure has been described in detail so far. To avoid obscuring the concept of the present disclosure, some details known in the art have not been described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

[0189] Although some specific embodiments of the present disclosure have been described in detail by way of examples, those skilled in the art will appreciate that the above examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Those skilled in the art will appreciate that modifications may be made to the above embodiments without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A control method for an unmanned loader, comprising: Obtaining data information related to the material pile and cavity pressure information of the actuator, wherein the data information includes: height information of the bucket penetrating the material pile, and volume information, temperature information, and image information of the material pile; and the cavity pressure information of the actuator includes at least one of cavity pressure information of the boom and cavity pressure information of the bucket; obtaining a looseness score and a viscosity score of the material pile based on the temperature information and image information of the material pile; Calculating a proportional parameter and an integral parameter based on information about the height at which the bucket cuts into the material pile, as well as information about the volume, looseness score, and viscosity score of the material pile; Calculating a pressure difference between a set cavity pressure threshold and the cavity pressure information of the actuator; Using the proportional parameter and the integral parameter as parameters of a proportional-integral control algorithm, using the pressure difference as input data of the proportional-integral control algorithm, and obtaining a compensation angle through the proportional-integral control algorithm; calculating an angle sum value between the compensation angle and a main control angle obtained by a main control algorithm; and The angle of the actuator is adjusted based on the angle and value so that the angle of the actuator conforms to the angle and value.

2. The control method according to claim 1, wherein: Based on the height information of the bucket penetrating the material pile, the volume information of the material pile, the looseness score, and the viscosity score, a proportional parameter and an integral parameter are calculated, including: Normalizing the height information of the bucket entering the material pile, the volume information, the looseness score, and the viscosity score of the material pile, respectively, to obtain normalized height information, normalized volume information, normalized looseness score, and normalized viscosity score; Calculating a first parameter based on the normalized height information and the normalized volume information, wherein the first parameter is a comprehensive parameter of the height and volume of the material pile; Calculating a second parameter based on the normalized bulkiness score and the normalized viscosity score, wherein the second parameter is a comprehensive parameter of the bulkiness and viscosity of the material pile; and The proportional parameter and the integral parameter are calculated based on the first parameter and the second parameter.

3. The control method according to claim 2, wherein: The first parameter for: ， in, is the normalized height information, is the normalized volume information, a is the weight corresponding to the normalized height information, and b is the weight corresponding to the normalized volume information, wherein a is a value between 0 and 1, b is a value between 0 and 1, and a+b=1.

4. The control method according to claim 2, wherein: The second parameter for: ， in, is the normalized looseness score, is the normalized stickiness score, c is the weight corresponding to the normalized looseness score, and d is the weight corresponding to the normalized stickiness score, wherein c is a value between 0 and 1, d is a value between 0 and 1, and c+d=1.

5. The control method according to claim 2, wherein: The scale parameter for: ， in, is the first parameter, is the second parameter, e is the weight corresponding to the first parameter, and f is the weight corresponding to the second parameter, wherein e is a value between 0 and 1, f is a value between 0 and 1, and e+f=1.

6. The control method according to claim 2, wherein: The integration parameter for: ， in, is the first parameter, is the second parameter, g is the weight corresponding to the first parameter, and h is the weight corresponding to the second parameter, wherein g is a value between 0 and 1, h is a value between 0 and 1, and g+h=1.

7. The control method according to claim 1, wherein: Obtaining a looseness score and a viscosity score of the material pile based on the temperature information and the image information of the material pile includes: The temperature information and image information of the material pile are input into a pre-trained transfer learning model, and the looseness score and viscosity score of the material pile are output through analysis by the transfer learning model.

8. The control method according to claim 1, wherein: The cavity pressure information of the actuator includes the cavity pressure information of the movable arm, and the cavity pressure information of the movable arm includes the large cavity pressure information of the movable arm; Calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator, including: calculating the pressure difference between the set large cavity pressure threshold of the movable arm and the large cavity pressure information of the movable arm; Adjusting the angle of the actuator based on the angle and the value includes adjusting the angle of the boom based on the angle and the value.

9. The control method according to claim 1, wherein: The cavity pressure information of the actuator includes the cavity pressure information of the movable arm, and the cavity pressure information of the movable arm includes the small cavity pressure information of the movable arm; Calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator, including: calculating the pressure difference between the set small cavity pressure threshold of the movable arm and the small cavity pressure information of the movable arm; Adjusting the angle of the actuator based on the angle and the value includes adjusting the angle of the boom based on the angle and the value.

10. The control method according to claim 1, wherein: The cavity pressure information of the actuator includes cavity pressure information of the bucket, and the cavity pressure information of the bucket includes large cavity pressure information of the bucket; Calculating a pressure difference between a set cavity pressure threshold and the cavity pressure information of the actuator, including: calculating a pressure difference between a set large cavity pressure threshold of the bucket and the large cavity pressure information of the bucket; Adjusting the angle of the actuator based on the angle and the value includes adjusting the angle of the bucket based on the angle and the value.

11. The control method according to claim 1, wherein: The cavity pressure information of the actuator includes cavity pressure information of the bucket, and the cavity pressure information of the bucket includes small cavity pressure information of the bucket; Calculating the pressure difference between the set cavity pressure threshold and the cavity pressure information of the actuator, including: calculating the pressure difference between the set small cavity pressure threshold of the bucket and the small cavity pressure information of the bucket; Adjusting the angle of the actuator based on the angle and the value includes adjusting the angle of the bucket based on the angle and the value.

12. A control device for an unmanned loader, comprising: a data acquisition module, configured to receive three-dimensional point cloud data of the material pile and its surrounding environment detected by the radar, and temperature information and image information of the material pile detected by the camera; a data processing module, configured to obtain information about the height at which the bucket penetrates the material pile and volume information of the material pile based on the three-dimensional point cloud data, obtain a looseness score and a viscosity score of the material pile based on temperature information and image information of the material pile, and calculate a proportional parameter and an integral parameter based on the information about the height at which the bucket penetrates the material pile, the volume information of the material pile, the looseness score, and the viscosity score; an auxiliary control module, configured to calculate, based on the obtained cavity pressure information of the actuator, a pressure difference between a set cavity pressure threshold and the cavity pressure information of the actuator, use the proportional parameter and the integral parameter as parameters of a proportional-integral control algorithm, use the pressure difference as input data of the proportional-integral control algorithm, and obtain a compensation angle through the proportional-integral control algorithm, wherein the cavity pressure information of the actuator includes at least one of cavity pressure information of the boom and cavity pressure information of the bucket; and A main control module is used to calculate the angle sum between the compensation angle and the main control angle obtained by the main control algorithm, and output the angle sum to the vehicle controller so that the vehicle controller adjusts the angle of the actuator so that the angle of the actuator conforms to the angle sum.

13. A control device for an unmanned loader, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the control method according to any one of claims 1 to 11 based on instructions stored in the memory.

14. A control system for an unmanned loader, comprising: A control device as claimed in claim 12 or 13; a radar for obtaining three-dimensional point cloud data of the material pile and its surroundings, and transmitting the three-dimensional point cloud data to the control device, wherein the control device obtains information about the height at which the bucket penetrates the material pile and information about the volume of the material pile based on the three-dimensional point cloud data; a camera, configured to obtain temperature information and image information of the material pile, and transmit the temperature information and the image information to the control device; and The pressure sensor is used to obtain the cavity pressure information of the actuator and transmit the cavity pressure information to the control device.

15. The control system according to claim 14, further comprising: a vehicle controller, configured to receive the angle and value from the control device and adjust the angle of the actuator according to the angle and value; Wherein, the control device is used to output the angle and value to the vehicle controller.

16. An unmanned loader, comprising: A control system as claimed in claim 14 or 15.

17. A computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are executed by a processor to implement the control method according to any one of claims 1 to 11.

18. A computer program product, comprising a computer program or instructions, wherein when the computer program or instructions are executed by a processor, the control method according to any one of claims 1 to 11 is implemented.

Images