Excavator operation action control method and device based on bucket trajectory planning

Abstract

The invention discloses an excavator operation action control method and device based on bucket trajectory planning, and belongs to the technical field of engineering machinery automation. The method comprises the steps that excavator attitude sensing data and excavator environment sensing data are obtained; the excavator attitude sensing data comprises all joint angles of the excavator; according to the excavator attitude sensing data, obtaining the position and attitude of the bucket and constructing a dynamic model; the dynamic model represents a mapping relation between the bucket stress information and the bucket force output constraint information; acquiring a three-dimensional construction model according to the excavator environment sensing data; acquiring a three-dimensional deviation field according to the three-dimensional construction model and a preset building information model; planning a target construction track of the bucket according to the position and posture of the bucket, the three-dimensional deviation field and the dynamic model; and the target pose sequence of the joint space of the excavator is obtained based on the target construction track of the bucket, action control is conducted on all joints of the excavator according to the target pose sequence, and the high-precision excavator control method capable of achieving real-time adjustment according to the operation quality is achieved.

Description

Technical Field

[0001] This invention belongs to the field of engineering machinery automation technology, and in particular relates to a method and device for controlling the operation of excavators based on bucket trajectory planning. Background Technology

[0002] The development of intelligent and automated construction machinery technology has roughly gone through the following stages: The manual operation stage relied entirely on the operator's visual observation and manual experience. Construction quality and efficiency were directly linked to the operator's skill level and fatigue. This stage suffered from drawbacks such as difficulty in quantifying quality, high costs, and poor safety. The two-dimensional mechanical guidance stage saw the emergence of 2D mechanical guidance systems with the widespread adoption of sensors and GNSS (Global Navigation Satellite System) technology. By installing GNSS receivers and tilt sensors on excavators, the real-time two-dimensional position of the bucket cutting edge and its deviation from the design drawings were displayed on a screen in the cab. This elevated construction guidance from pure experience to a data-driven level, reducing the number of measurements and improving the accuracy of horizontal and elevation control. However, it also had limitations such as insufficient information dimensions, lack of real-time quality feedback, and over-reliance on GNSS.

[0003] In recent years, with the development of 3D scanning, BIM (Building Information Modeling), and automatic control technologies, the industry has begun to explore more advanced solutions. 3D guidance based on BIM models: By importing the design BIM model into the onboard computer and combining it with GNSS and machine attitude sensors, the real-time position of the bucket relative to the 3D design model can be displayed on the screen, achieving guidance accuracy higher than 2D systems. Semi-automated excavators capable of performing simple repetitive actions have emerged, incorporating 3D design models and laying the foundation for automated control.

[0004] However, existing technologies also have certain limitations, such as: lack of environmental perception: most systems still only focus on the machine's own state and lack the ability to perceive changes in the surrounding environment (such as the shape of the excavated trench) in real time; insufficient intelligence in decision-making: automated decision-making functions are relatively simple, mostly using preset programs or simple control logic, which are difficult to cope with complex and dynamic construction environments; lack of correction ability: in automated construction control, it is often impossible to accurately estimate the next action based on the target construction situation, nor can it adjust the control strategy according to the real-time situation during the excavator's operation. Summary of the Invention

[0005] The purpose of this invention is to provide an excavator operation motion control method and device based on bucket trajectory planning. By sensing the vehicle posture and working environment, the excavator construction site is reconstructed and compared with the target construction site to complete construction quality monitoring and generate the bucket target trajectory. The bucket target trajectory is then used to control the movement of each joint, realizing an excavator motion control method that can be adjusted in real time according to the operation quality.

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

[0007] In a first aspect, the present invention provides a method for controlling the operational actions of an excavator based on bucket trajectory planning, comprising: Acquire excavator posture perception data and excavator environment perception data; among which, excavator posture perception data includes the angles of each joint of the excavator; The bucket pose is obtained based on the excavator's attitude perception data, and a dynamic model is constructed. The dynamic model represents the mapping relationship between the bucket's force information and the bucket's output force constraint information. A three-dimensional model of the construction environment is obtained based on the excavator's environmental perception data; The three-dimensional deviation field is obtained based on the three-dimensional model of the construction environment and the pre-designed building information model; The bucket target construction trajectory is planned based on the bucket posture, three-dimensional deviation field, and dynamic model. The target pose sequence of the excavator joint space is obtained based on the target construction trajectory of the bucket, and the motion control of each joint of the excavator is performed according to the target pose sequence.

[0008] Optionally, methods for acquiring excavator attitude perception data include: The vehicle's rotation angle and the angle between the vehicle body and the horizontal plane are obtained by an inertial measurement unit and tilt sensor installed in the core of the fuselage. The amplitude of each joint is obtained by angle sensors installed at each joint of the excavator; the joints of the excavator include the boom, stick, bucket, and hydraulic actuators. The Kalman filter algorithm is used to fuse the vehicle body rotation angle, horizontal plane angle, and joint amplitude to obtain excavator posture perception data.

[0009] Optionally, methods for acquiring excavator environmental perception data include: The three-dimensional point cloud information of the construction environment is obtained by using a lidar sensor installed on the excavator; Color information of the construction environment is obtained by using a binocular camera installed on the excavator; By fusing 3D point cloud information and color information, environmental perception data of the excavator is obtained.

[0010] Optionally, methods for obtaining the bucket pose based on excavator attitude perception data include: Geometric models of each joint of the excavator were established using the DH method; Based on the angles of each joint of the excavator, the current bucket position is obtained through a forward kinematics algorithm.

[0011] Optionally, methods for constructing a dynamic model based on excavator attitude perception data include: Construct a multibody dynamics model that includes the boom, stick, bucket, and hydraulic actuators; The parameters of the multibody dynamics model are set according to the physical parameters of the excavator; among them, the physical parameters of the excavator include mass, moment of inertia and hydraulic system characteristics; Use simulation software or custom algorithms to validate models and adjust parameters.

[0012] Optionally, methods for obtaining a 3D model of the construction environment based on excavator environmental perception data include: The environmental perception data of the excavator is sequentially processed by denoising, registration and segmentation to obtain effective point cloud data; A 3D model of the current construction environment is constructed based on effective point cloud data using a finite element mesh generation algorithm.

[0013] Optionally, methods for obtaining the three-dimensional deviation field based on the three-dimensional model of the construction environment and the pre-designed building information model include: Unify and register the coordinates of the current 3D model of the construction environment with the Building Information Model (BIM). The difference between the current construction environment and the target construction environment is calculated based on the registered model, and a three-dimensional deviation field is generated.

[0014] Optionally, methods for planning the target construction trajectory of the bucket based on the bucket pose, three-dimensional deviation field, and dynamic model include: The current construction area to be adjusted is determined based on the three-dimensional deviation field; Based on the current construction area to be adjusted and the current bucket position, an optimization algorithm is used to plan the target construction trajectory of the bucket that satisfies the constraints of the dynamic model.

[0015] Optionally, methods for obtaining the target pose sequence of the excavator joint space based on the bucket target construction trajectory include: The target construction trajectory of the bucket is decomposed into bucket poses at multiple work points to obtain a bucket pose sequence. Based on the bucket pose sequence, the target angles of each joint at each working point are calculated using an inverse kinematics algorithm. By combining the operational objectives and the mechanical constraints of the excavator, the target angles of each joint corresponding to each operational point are optimized to obtain the target pose sequence of the joint space of each operational point; among them, the operational objectives include the bucket full rate and obstacle avoidance; the mechanical constraints of the excavator include the angle limits of each joint and the speed and acceleration boundaries.

[0016] Optionally, the method for motion control of each joint of the excavator based on the target pose sequence includes: Based on the target pose sequence, the Model Predictive Control (MPC) algorithm is used to generate control commands. Control commands are sent to the excavator's hydraulic system to adjust the current movements of each joint; It receives the current excavator body posture data to obtain the current action execution status of each joint, and adjusts the control commands according to the current action execution status.

[0017] Secondly, the present invention provides an excavator operation action control device based on bucket trajectory planning, comprising: Working condition perception module: used to acquire excavator posture perception data and excavator environment perception data; among which, the excavator posture perception data includes the angles of each joint of the excavator; Bucket posture acquisition module: used to acquire bucket posture based on excavator posture perception data and build a dynamic model; wherein, the dynamic model represents the mapping relationship between bucket force information and bucket output constraint information; Construction Environment Model Acquisition Module: Used to acquire a 3D model of the construction environment based on the excavator's environmental perception data; Target Deviation Field Acquisition Module: Used to acquire the three-dimensional deviation field based on the three-dimensional model of the construction environment and the pre-designed building information model; Target construction trajectory acquisition module: used to plan the target construction trajectory of the bucket based on the bucket posture, three-dimensional deviation field and dynamic model; Joint motion control module: used to obtain the target pose sequence of the excavator joint space based on the target construction trajectory of the bucket, and to perform motion control on each joint of the excavator according to the target pose sequence.

[0018] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: Through the tight coupling of 3D point cloud reconstruction and machine posture perception, it not only provides the excavator's precise pose in the global map, but also reconstructs the dynamic 3D model of the surrounding environment in real time. Furthermore, it can accurately capture the motion state of each joint of the excavator, enabling the system to perceive its own posture and the external environment simultaneously within a unified coordinate system. This provides a rich, accurate, and real-time data foundation for bucket trajectory decision-making. Moreover, by placing quality control at the forefront of the construction process, after each excavation action, the real-time acquired construction scene point cloud is precisely registered with the preset model and subjected to 3D difference analysis to generate a 3D deviation field in real time. Based on the real-time quality deviation, subsequent excavation trajectories and control strategies are dynamically adjusted, forming a complete closed loop of perception-evaluation-control. This solves the problem of the inefficiency and unavoidable consequences of traditional construction quality control relying heavily on post-construction manual measurement and acceptance. The rework issue ensured that the construction process always proceeded in line with the design goals, fundamentally guaranteeing the final construction quality and improving construction efficiency. By using model predictive control algorithms to predict the future state of the excavator system using its dynamic model, and generating a control sequence for the excavator joint space by solving a constrained optimization problem, precise and efficient control of this complex nonlinear system was achieved. It can explicitly handle physical constraints such as speed and force limits of the hydraulic system, effectively cope with system delays and external disturbances, and achieve more accurate and stable control. More importantly, the model predictive control algorithm can receive real-time construction quality assessment data, upgrading its control objective from simple trajectory tracking to quality control, and obtaining guidance for deviation correction in real time. This not only significantly improves operational accuracy and efficiency, but also enables intelligent response to uncertainties such as changes in soil hardness and mechanical wear, achieving adaptive adjustment of control strategies. Attached Figure Description

[0019] Figure 1 The diagram shown is a schematic diagram of an excavator operation action control method based on bucket trajectory planning in one embodiment of the present invention.

[0020] Figure 2 The diagram shown is a flowchart of a sensing and positioning method in one embodiment of the present invention;

[0021] Figure 3 The diagram shown is a schematic diagram of the excavator sensor layout in one embodiment of the present invention;

[0022] Figure 4 The diagram shown is a schematic diagram of the excavator forward model in one embodiment of the present invention;

[0023] Figure 5 The diagram shown is a flowchart of a decision-making and control method in one embodiment of the present invention.

[0024] Figure 6 The diagram shown is a flowchart of the interaction and display method in one embodiment of the present invention.

[0025] Figure 3 The components are: 1. First position sensor; 2. Second position sensor; 3. Third position sensor; 4. Fourth position sensor; 5. LiDAR; 6. Pressure sensor; 7. Binocular / TOF camera. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0027] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0028] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0029] Example 1

[0030] like Figure 1As shown, this embodiment provides an excavator operation motion control method based on bucket trajectory planning. It reconstructs the excavator construction site using 3D SLAM technology and sensor position sensing. A point cloud registration algorithm is used to compare the construction site point cloud with a preset building model point cloud to complete the construction quality assessment. Based on the construction quality assessment results, the construction target is obtained, and the bucket's construction trajectory is planned based on the construction target. Simultaneously, an attitude controller installed on the excavator body controls the bucket's motion trajectory. An MPC algorithm is used to improve control accuracy. Finally, a 3D AR interface is used to complete the interaction between the system and the operator. The specific steps are as follows:

[0031] I. For example Figure 2 As shown, the specific steps of perception and localization are as follows:

[0032] Step S11: Environmental Perception and Data Fusion

[0033] like Figure 3 As shown, a first position sensor 1 is installed at the bottom of the cab. The mounting surface of the first position sensor 1 is as parallel to the ground as possible, and the sensor's pitch detection axis is as parallel to the centerline of the excavator body as possible. It is used to detect the swing angle and the angle between the excavator body and the horizontal plane, which affect the relative position and attitude between the bucket and the target working surface. A second position sensor 2 is installed on the boom structure. The detection axis of the second position sensor 2 is as parallel as possible to the line connecting the hinge points at both ends of the boom. It is used to monitor the amplitude angle between the excavator boom and the horizontal plane in real time, and can calculate the current position of the boom to achieve precise control of the boom's movement. A third position sensor 3 is installed on the stick structure. The detection axis of the third position sensor 3 is as parallel as possible to the line connecting the hinge points at both ends of the stick. It is responsible for detecting the amplitude angle between the stick and the horizontal plane, and can calculate the precise position of the stick. A fourth position sensor 4 is installed on the bucket linkage structure. The detection axis of the fourth position sensor 4 is as parallel as possible to the bucket linkage, and is used to detect the amplitude angle between the bucket and the horizontal plane. LiDAR sensors 5 are mounted on both sides of the boom to detect high-density 3D point cloud data of the construction conditions. They obtain 3D information about the construction surface through hemispherical scanning areas on both sides, enabling analysis of state changes and construction conditions during the process. Pressure sensors 6 are installed on the hydraulic valves and pilot valves to detect the pressure in the working and pilot oil circuits, ensuring the hydraulic system operates at optimal pressure, providing stable power output, and preventing system overload. A binocular / TOF camera 7 is mounted on the top of the cab, with an optional GNSS receiver to provide the corresponding absolute position value, used to obtain a 3D information description of the construction surface. Unlike high-density pure point cloud information, the binocular / TOF camera 7 can acquire low-density spatial data that can be combined with color information, which is then fused into the data perception module to supplement the color information lacking in the LiDAR sensors 5.

[0034] A real-time 3D SLAM algorithm based on laser point cloud or vision-inertial fusion is adopted. First, real-time reconstruction is performed. During the construction process, the pits, soil piles, etc. around the excavator are scanned and 3D modeled in real time to generate a dense point cloud map. Then, self-localization is completed to estimate the precise position and attitude of the excavator in the global map of the construction scene in real time.

[0035] Step S12, Attitude-aware data fusion

[0036] Data is collected from high-precision sensors installed on key joints of the excavator, such as the boom, stick, and bucket, as well as from IMU and tilt sensors installed in the core of the machine. A Kalman filter algorithm is used to fuse the data from different sensors, accurately measuring and feeding back the three-dimensional coordinates and attitude of the excavator bucket cutting edge in the global coordinate system.

[0037] Step S13, Forward and Inverse Kinematics Solutions of the Excavator

[0038] like Figure 4 As shown, a geometric relationship model (such as the DH method) is established between the boom, stick, and bucket to achieve a two-way coordinate transformation from joint space to Cartesian space. Using forward kinematics, based on the displacement sensor readings of the boom, stick, and bucket cylinders, the precise three-dimensional coordinates of the bucket teeth tip and the bucket's attitude angle in the global coordinate system are calculated in real time. Using inverse kinematics, based on the planned target motion trajectory of the bucket teeth tip and the desired attitude angle, the precise displacement required by each cylinder is calculated inversely, serving as the reference input for the control system.

[0039] Step S14: Solving the excavator dynamics problem

[0040] A multibody dynamics model incorporating the boom, stick, bucket, and hydraulic actuators is established to comprehensively analyze the various loads acting on the bucket during operation, including digging resistance, material gravity, inertial forces, and the dynamic characteristics of the hydraulic cylinders. The model can solve for the dynamic digging force at the bucket teeth, the force at each hinge point, and the pressure fluctuations in the hydraulic system in real time. It can also simulate the dynamic response of the bucket interacting with the soil under different soil conditions (considering parameters such as soil cohesion coefficient and internal friction angle). This provides accurate dynamic data support for optimizing the operating efficiency of excavator working devices and implementing advanced control strategies (such as load-sensitive control and energy consumption optimization).

[0041] II. Figure 5 As shown, the specific steps of the decision-making and control method are as follows:

[0042] Step S21: Point cloud analysis and evaluation of construction quality

[0043] This embodiment utilizes an intelligent quality monitoring system based on 3D point cloud data to perform point cloud analysis and evaluation of excavator construction quality. It collects high-density 3D point cloud data of the construction work surface (such as the bottom of a trench or the surface of a slope) using a LiDAR 5 and a binocular / TOF camera 7 installed on the excavator. The point cloud data undergoes preprocessing such as denoising, registration, and segmentation, and a high-precision 3D model of the work surface is reconstructed based on triangulation or quadrilateral mesh generation algorithms. The reconstructed 3D model is automatically compared with the design model (such as a BIM model), enabling precise calculation and quantitative evaluation of deviations in key construction indicators, such as excavation elevation, slope, flatness, cross-sectional dimensions, and earthwork volume. Simultaneously, a visual evaluation report containing color deviation maps and quantitative statistical reports is automatically generated, thus achieving real-time, accurate, and digital evaluation and traceability of construction quality, effectively improving construction accuracy and operational efficiency.

[0044] Step S22, Intelligent Bucket Trajectory Planning

[0045] Using LiDAR, RTK positioning system, and position sensors, real-time construction environment data is collected, including the surface morphology of the excavation area, obstacle information, and machine posture, and a precise 3D model of the area to be excavated is constructed. Subsequently, based on the excavation task requirements such as trench cross-section and excavation depth, and the optimization goals of full bucket and obstacle avoidance, the area to be excavated is discretized. By establishing a functional relationship between the bucket tooth tip path and the excavation volume, a preliminary bucket tooth tip path is planned within the task space that meets operational habits, achieves full bucket capacity, and complies with mechanical constraints (such as joint angle, speed, and acceleration limits).

[0046] In this embodiment, to further improve trajectory quality, time-optimal trajectory algorithms such as cubic spline curves and NURBS curve interpolation are used to replan and smooth the initial path, generating a smooth, continuous, and time-optimal tooth tip motion trajectory in joint space, as well as the angle and angular velocity sequences of each joint. This planning result will serve as the input to the motion control system, and ultimately, through a multi-joint collaborative control algorithm, drive the excavator's working device to accurately and efficiently complete autonomous excavation operations.

[0047] Step S23: Nonlinear controller predictive control

[0048] A nonlinear model, including Hammerstein and other nonlinear models, is established to incorporate the nonlinear dynamic characteristics of the hydraulic system (such as pressure-flow relationship and cylinder friction) and the multibody dynamics of the working device. Combined with a model predictive control algorithm, the motion trajectory of the bucket teeth tip is optimized in real-time over a future period. At each step, based on the current system state (such as joint angles and hydraulic pressure) and the predictive model, the optimal control sequence that satisfies mechanical constraints and operational objectives is calculated. A seventh-order polynomial curve is used to smooth the joint angle trajectory, ensuring the smoothness of the excavator's joint movements. This directly generates precise control commands for the hydraulic cylinders, achieving high-precision, adaptive tracking of the planned trajectory. Even under complex working conditions such as varying soil resistance, the control remains robust and stable.

[0049] III. Figure 6 As shown, the specific steps of the interaction and display methods are as follows:

[0050] Step S31, 3D AR Augmented Reality Interaction

[0051] A 3D AR augmented reality interactive interface for excavator bucket control, based on AR glasses or mobile devices, is introduced. This interface utilizes gesture recognition or virtual controls for interaction, allowing operators to directly define the digging area with gestures in the air or adjust the bucket posture by touching virtual sliders. Based on the design model BIM and environmental perception data, the deviation between the current bucket position and the target elevation is displayed intuitively in a virtual-real fusion manner (e.g., using color coding: green indicates acceptable, red indicates over-digging / under-digging), and the optimal digging angle is dynamically suggested. This significantly reduces the difficulty of operation, achieving precise and efficient control where what you see is what you dig, effectively avoiding the visual blind spot problem in traditional operations, and improving construction quality and safety.

[0052] Step S32: Construction scene and 3D to 2D planar marking of the bucket

[0053] By integrating GNSS receivers and tilt sensors onto the excavator boom, stick, and bucket, the three-dimensional spatial coordinates of the bucket tip in the absolute coordinate system are acquired in real time. Subsequently, using orthophoto projection algorithms (specifically implemented by constructing a projection matrix) or perspective projection technology, the complex three-dimensional motion trajectory of the bucket, the design cross-sectional model, and the real-time working face elevation information are efficiently converted into easily identifiable two-dimensional planar graphics. Finally, on the visualization terminal in the cab, the current position of the bucket, the design boundary, the planned trajectory, and key construction indicators (such as the deviation between the current elevation and the design elevation) are dynamically overlaid and displayed in the form of a two-dimensional planar map.

[0054] Step S33: Generation of 3D construction trajectory for bucket working point

[0055] Based on the imported BIM or CAD design model (such as design cross-sections and elevation information), the construction task is transformed into a specific sequence of excavation work points. Through 2D inverse projection and 3D calculation, the 2D coordinates of the planar work points specified on the operator's screen are converted into 3D coordinates under actual working conditions, and kinematic calculations are used to determine the target angles of each joint of the excavator. Furthermore, considering factors such as bucket capacity, obstacle avoidance, mechanical motion constraints, and soil characteristics, interpolation algorithms (such as cubic spline curves) or optimization algorithms are employed to generate a smooth, efficient, and continuous trajectory sequence in 3D space that directly guides the bucket from the entry point to the exit point, providing precise path instructions for the excavator's automatic control or operator guidance.

[0056] Example 2

[0057] This embodiment provides an excavator operation motion control device based on bucket trajectory planning, including: Working condition perception module: used to acquire excavator posture perception data and excavator environment perception data; among which, the excavator posture perception data includes the angles of each joint of the excavator; Bucket posture acquisition module: used to acquire bucket posture based on excavator posture perception data and build a dynamic model; wherein, the dynamic model represents the mapping relationship between bucket force information and bucket output constraint information; Construction Environment Model Acquisition Module: Used to acquire a 3D model of the construction environment based on the excavator's environmental perception data; Target Deviation Field Acquisition Module: Used to acquire the three-dimensional deviation field based on the three-dimensional model of the construction environment and the pre-designed building information model; Target construction trajectory acquisition module: used to plan the target construction trajectory of the bucket based on the bucket posture, three-dimensional deviation field and dynamic model; Joint motion control module: used to obtain the target pose sequence of the excavator joint space based on the target construction trajectory of the bucket, and to perform motion control on each joint of the excavator according to the target pose sequence.

[0058] The device provided in this embodiment can execute the excavator operation action control method based on bucket trajectory planning provided in any step of Embodiment 1, and has the corresponding functional modules and beneficial effects of the execution method.

[0059] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0060] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0061] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0062] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0063] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for controlling the operational actions of an excavator based on bucket trajectory planning, characterized in that, include: Acquire excavator posture perception data and excavator environment perception data; among which, excavator posture perception data includes the angles of each joint of the excavator; The bucket pose is obtained based on the excavator's attitude perception data, and a dynamic model is constructed. The dynamic model represents the mapping relationship between the bucket's force information and the bucket's output force constraint information. A three-dimensional model of the construction environment is obtained based on the excavator's environmental perception data; The three-dimensional deviation field is obtained based on the three-dimensional model of the construction environment and the pre-designed building information model; The bucket target construction trajectory is planned based on the bucket posture, three-dimensional deviation field, and dynamic model. The target pose sequence of the excavator joint space is obtained based on the target construction trajectory of the bucket, and the motion control of each joint of the excavator is performed according to the target pose sequence.

2. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for acquiring excavator attitude perception data include: The vehicle's rotation angle and the angle between the vehicle body and the horizontal plane are obtained by an inertial measurement unit and tilt sensor installed in the core of the fuselage. The amplitude of each joint is obtained by angle sensors installed at each joint of the excavator; the joints of the excavator include the boom, stick, bucket, and hydraulic actuators. The Kalman filter algorithm is used to fuse the vehicle body rotation angle, horizontal plane angle, and joint amplitude to obtain excavator posture perception data.

3. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for acquiring environmental perception data from excavators include: The three-dimensional point cloud information of the construction environment is obtained by using a lidar sensor installed on the excavator; Color information of the construction environment is obtained by using a binocular camera installed on the excavator; By fusing 3D point cloud information and color information, environmental perception data of the excavator is obtained.

4. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for obtaining bucket pose based on excavator posture perception data include: Geometric models of each joint of the excavator were established using the DH method; Based on the angles of each joint of the excavator, the current bucket position is obtained through a forward kinematics algorithm.

5. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for constructing dynamic models based on excavator attitude perception data include: Construct a multibody dynamics model that includes the boom, stick, bucket, and hydraulic actuators; The parameters of the multibody dynamics model are set according to the physical parameters of the excavator; among them, the physical parameters of the excavator include mass, moment of inertia and hydraulic system characteristics; Use simulation software or custom algorithms to validate models and adjust parameters.

6. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for obtaining a 3D model of the construction environment based on excavator environmental perception data include: The environmental perception data of the excavator is sequentially processed by denoising, registration and segmentation to obtain effective point cloud data; A 3D model of the current construction environment is constructed based on effective point cloud data using a finite element mesh generation algorithm.

7. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for obtaining the three-dimensional deviation field based on the three-dimensional model of the construction environment and the pre-designed building information model include: Unify and register the coordinates of the current 3D model of the construction environment with the Building Information Model (BIM). The difference between the current construction environment and the target construction environment is calculated based on the registered model, and a three-dimensional deviation field is generated.

8. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for planning the target construction trajectory of the bucket based on the bucket posture, three-dimensional deviation field, and dynamic model include: The current construction area to be adjusted is determined based on the three-dimensional deviation field; Based on the current construction area to be adjusted and the current bucket position, an optimization algorithm is used to plan the target construction trajectory of the bucket that satisfies the constraints of the dynamic model.

9. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for obtaining the target pose sequence of the excavator joint space based on the bucket target construction trajectory include: The target construction trajectory of the bucket is decomposed into bucket poses at multiple work points to obtain a bucket pose sequence. Based on the bucket pose sequence, the target angles of each joint at each working point are calculated using an inverse kinematics algorithm. By combining the operational objectives and the mechanical constraints of the excavator, the target angles of each joint corresponding to each operational point are optimized to obtain the target pose sequence of the joint space of each operational point; among them, the operational objectives include the bucket full rate and obstacle avoidance; the mechanical constraints of the excavator include the angle limits of each joint and the speed and acceleration boundaries.

10. The excavator operation motion control method based on bucket trajectory planning according to claim 1, characterized in that, Methods for motion control of excavator joints based on target pose sequences include: Based on the target pose sequence, the Model Predictive Control (MPC) algorithm is used to generate control commands. Control commands are sent to the excavator's hydraulic system to adjust the current movements of each joint; It receives the current excavator body posture data to obtain the current action execution status of each joint, and adjusts the control commands according to the current action execution status.

11. A control device for excavator operation actions based on bucket trajectory planning, characterized in that, include: Working condition perception module: used to acquire excavator posture perception data and excavator environment perception data; among which, the excavator posture perception data includes the angles of each joint of the excavator; Bucket posture acquisition module: used to acquire bucket posture based on excavator posture perception data and build a dynamic model; wherein, the dynamic model represents the mapping relationship between bucket force information and bucket output constraint information; Construction Environment Model Acquisition Module: Used to acquire a 3D model of the construction environment based on the excavator's environmental perception data; Target Deviation Field Acquisition Module: Used to acquire the three-dimensional deviation field based on the three-dimensional model of the construction environment and the pre-designed building information model; Target construction trajectory acquisition module: used to plan the target construction trajectory of the bucket based on the bucket posture, three-dimensional deviation field and dynamic model; Joint motion control module: used to obtain the target pose sequence of the excavator joint space based on the target construction trajectory of the bucket, and to perform motion control on each joint of the excavator according to the target pose sequence.

Images