Riemannian manifold excavator behavior cloning method and device based on improved double quaternion

Abstract

The invention relates to a Riemannian manifold excavator behavior cloning method and device based on improved double quaternions. The method comprises the steps that a forward dynamical model of the excavator is constructed through a DH parameter method, a transfer matrix under a Cartesian coordinate system of the tail end of the excavator and translation variables and rotation variables under the Cartesian coordinate system of the excavator are obtained according to the forward dynamical model, quaternion representation of the rotation part of the tail end of the excavator is obtained, and the rotation part of the tail end of the excavator is calculated according to the quaternion representation. The method comprises the following steps: obtaining double quaternion representation of the position and attitude of the whole excavator, representing the position and attitude of the tail end of the excavator in a Cartesian coordinate system as Riemannian manifold double quaternion representation of a Riemannian manifold tangent space through Riemannian manifold mapping, constructing a non-parametric Bayesian Gaussian process model, giving a query timestamp, and constructing a non-parametric Bayesian Gaussian process model; and calculating a mean value and a variance of output behavior data components to obtain a final behavior cloning result. By adopting the method, the expression ability of the excavator operation behavior can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of excavator behavior cloning, and in particular to a method and device for cloning an excavator behavior based on an improved dual quaternion Riemannian manifold. Background Art

[0002] In the development process of intelligent construction machinery, although intelligent unmanned aerial vehicle excavators have broad prospects, they are currently facing many problems that need to be solved urgently. The efficiency of their autonomous excavation is extremely low. In actual operation, it is difficult to complete the excavation task quickly and accurately. Compared with manual excavation, the efficiency gap is obvious. It cannot truly replace the operator for efficient excavation, resulting in a significant increase in construction costs. At the same time, autonomous excavation embodied intelligence technology is not yet mature, especially in terms of imitating the excavation behavior of cloned operators, there is a lack of effective methods. This makes it impossible for intelligent unmanned excavators to accurately learn and reproduce human operating skills, and it is difficult to adapt to complex and changeable excavation conditions, which seriously restricts the development and practical application of this technology. Therefore, seeking an effective method to improve the imitation learning ability of intelligent unmanned excavators has become the key to promoting the advancement of autonomous excavation embodied intelligence technology. Summary of the invention

[0003] Based on this, it is necessary to provide a Riemann manifold excavator behavior cloning method and device based on improved dual quaternions to address the above technical problems.

[0004] A Riemannian manifold excavator behavior cloning method based on improved dual quaternion, the method comprising: A forward dynamics model of the excavator is constructed by the DH parameter method, and a transfer matrix in a Cartesian coordinate system at the end of the excavator, a translation variable and a rotation variable in a Cartesian coordinate system at the end of the excavator are obtained according to the forward dynamics model; According to the transfer matrix and the rotation variable, a quaternion representation of the rotating part of the excavator end is obtained, and according to the translation variable and the quaternion representation, a dual quaternion representation of the position and attitude of the entire excavator is obtained; According to the dual quaternion representation, the position and posture of the end of the excavator in the Cartesian coordinate system are represented as a dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping; According to the projection of the behavior data of the excavator operator onto the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed. According to the non-parametric Bayesian Gaussian process model, given the query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

[0005] In one of the embodiments, it also includes: performing electronic control modification on the excavator; the modification includes: setting IMU sensors on the excavator's boom, dipper arm, bucket and body rotation mechanism, and installing a proportional valve on the excavator's manual mechanical drive, and connecting a data acquisition card.

[0006] In one embodiment, the method further includes: constructing a forward dynamics model of the excavator by using a DH parameter method; the DH parameters include: the distance between two adjacent axes along the x-axis , the angle of rotation of two adjacent axes along the x-axis , the distance between two adjacent axes translated along the z-axis , the angle of rotation of two adjacent axes along the z axis ; According to the forward dynamics model, the transfer matrix in the Cartesian coordinate system of the excavator end is obtained. for: ; in, , , , . , , , , , , , These are the posture data collected by the IMU sensors placed on the excavator's boom, arm, bucket, and body slewing mechanism; The translation variables of the excavator in the Cartesian coordinate system are: .

[0007] The rotation variable is expressed in Euler angles as: .

[0008] In one embodiment, the method further includes: obtaining the quaternion representation of the rotating part of the excavator end according to the transfer matrix, the translation variable and the rotation variable as follows: .

[0009] in, , , and It is expressed in Cartesian coordinate system.

[0010] In one embodiment, the method further comprises: obtaining a dual quaternion representation of the position and attitude of the entire excavator according to the quaternion representation, including: .

[0011] In one embodiment, the method further includes: obtaining the end position and posture of the excavator according to the dual quaternion representation as: .

[0012] in, is the position of the excavator end in the Cartesian coordinate system, It is the corner mark of the excavator's position and posture at a certain moment.

[0013] In one embodiment, the method further includes: by Riemann manifold mapping, the position and posture of the end of the excavator in the Cartesian coordinate system are represented as the Riemann manifold double quaternion of the Riemann manifold tangent space as follows: .

[0014] In one embodiment, the method further includes: constructing a non-parametric Bayesian Gaussian process model based on the projected behavior data of the excavator onto the Riemann manifold tangent space; the square exponential covariance function of the non-parametric Bayesian Gaussian process model is: .

[0015] in, , and is the hyperparameter of the Gaussian process model, and are the timestamps corresponding to the excavator behavior data components, is the Kronecker symbol; The timestamp of the given query , get the mean of the output mining behavior data component and variance They are: ; ; in, and are the covariance matrices of the timestamps corresponding to the excavator behavior data and the query data timestamps, is the corresponding cross covariance matrix. A set of cloned data vectors for the operator's behavior; The final behavior cloning result obtained by the inverse mapping of the Riemann manifold mapping is: ; in, Represents the behavior cloning result.

[0016] A Riemannian manifold excavator behavior cloning device based on improved dual quaternion, the device comprising: A forward dynamics model building module is used to build a forward dynamics model of the excavator by using the DH parameter method, and obtain the transfer matrix in the Cartesian coordinate system of the excavator end, the translation variable and the rotation variable in the Cartesian coordinate system of the excavator according to the forward dynamics model; A quaternion representation module, used to obtain a quaternion representation of the rotating part of the excavator end according to the transfer matrix and the rotation variable, and to obtain a dual quaternion representation of the position and attitude of the entire excavator according to the translation variable and the quaternion representation; A Riemann mapping module, used to represent the position and posture of the end of the excavator in the Cartesian coordinate system as a Riemann manifold dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping according to the dual quaternion representation; The behavior cloning module is used to construct a non-parametric Bayesian Gaussian process model based on the projection of the excavator's behavior data onto the Riemann manifold tangent space. According to the non-parametric Bayesian Gaussian process model, given a query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

[0017] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: A forward dynamics model of the excavator is constructed by the DH parameter method, and a transfer matrix in the Cartesian coordinate system of the excavator end, and a translation variable and a rotation variable in the Cartesian coordinate system of the excavator are obtained according to the forward dynamics model; According to the transfer matrix and the rotation variable, a quaternion representation of the rotating part of the excavator end is obtained, and according to the translation variable and the quaternion representation, a dual quaternion representation of the position and attitude of the entire excavator is obtained; According to the dual quaternion representation, the position and attitude representation of the end of the excavator are obtained, and the position and attitude of the end of the excavator in the Cartesian coordinate system are represented as the dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping; According to the projection of the excavator's behavior data onto the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed. According to the non-parametric Bayesian Gaussian process model, given the query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

[0018] The above-mentioned Riemann manifold excavator behavior cloning method and device based on improved dual quaternions can more accurately and comprehensively describe the posture state of the excavator by improving the dual quaternion method, while the traditional Euler angle method is easily affected by the universal joint deadlock problem and a single quaternion is difficult to fully characterize the complex posture. This avoids possible mathematical singularity problems during operation. The method based on Riemann manifold can use the geometric structure of the manifold to accurately model in nonlinear space and capture complex behaviors. In addition, the introduction of improved dual quaternions can effectively describe multi-degree-of-freedom motion, and the Riemann manifold theory can better characterize the connection between the behavior trajectory in time and space, thereby improving the model's ability to express the excavator's operating behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a behavior cloning method of a Riemannian manifold excavator based on improved dual quaternions in one embodiment; Figure 2 A structural block diagram of a behavior cloning device of a Riemannian manifold excavator based on improved dual quaternions in one embodiment; Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0021] In one embodiment, Figure 1 As shown, a Riemann manifold excavator behavior cloning method based on improved dual quaternion is provided, comprising the following steps: Step 102, constructing a forward dynamics model of the excavator by using the DH parameter method, and obtaining a transfer matrix in the Cartesian coordinate system of the excavator end, and translation variables and rotation variables in the Cartesian coordinate system of the excavator according to the forward dynamics model.

[0022] The DH (Denavit-Hartenberg) parameter method is a standard method widely used in robot kinematic modeling. It mathematically describes the relative positions and directions of the joints and links of a robot (here, an excavator) by defining a series of link and joint parameters. For a mechanical system such as an excavator that consists of multiple links (boom, arm, bucket, etc.) and joints, the DH parameter method can systematically establish its kinematic model.

[0023] Through the above steps, the forward dynamics model constructed according to the DH parameter method can be used to obtain the transfer matrix in the Cartesian coordinate system of the excavator end, the translation variables and rotation variables in the Cartesian coordinate system, so as to accurately describe the position and posture of the excavator end effector in the Cartesian space.

[0024] Step 104, obtaining the quaternion representation of the rotating part of the excavator end according to the transfer matrix and the rotation variable, and obtaining the dual quaternion representation of the position and posture of the entire excavator according to the translation variable and the quaternion representation.

[0025] Dual quaternion is an extension of quaternion. It can represent rotation and translation at the same time and is suitable for describing the complete position and posture of a rigid body in three-dimensional space.

[0026] Step 106, obtaining the position and attitude representation of the end of the excavator according to the dual quaternion representation, and expressing the position and attitude of the end of the excavator in the Cartesian coordinate system as the dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping.

[0027] Riemannian manifold is a smooth manifold with Riemannian metric, which provides a mathematical framework for studying curved space. In robot kinematics, the space formed by the position and posture of the end of the excavator is usually nonlinear, and Riemannian manifold can model this nonlinear space well. Tangent space is a linear approximate space at a point on the manifold, which facilitates linear operations and calculations on nonlinear manifolds. It can be mapped from dual quaternion space (a specific form of Riemannian manifold) to tangent space through logarithmic mapping. Logarithmic mapping converts the nonlinear dual quaternion representation into a linear tangent space representation, which is convenient for subsequent mathematical analysis and processing. After logarithmic mapping, the position and posture of the end of the excavator, which was originally represented by dual quaternions in the Cartesian coordinate system, is converted into a vector in the tangent space of the Riemannian manifold. This vector is the dual quaternion representation of the Riemannian manifold. In the tangent space, operations such as interpolation and derivation can be performed more conveniently, which also helps to use methods such as linear algebra and machine learning to analyze and process the motion data of the excavator.

[0028] Step 108, based on the projection of the excavator's behavior data onto the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed. Based on the non-parametric Bayesian Gaussian process model, given the query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

[0029] In the above-mentioned Riemann manifold excavator behavior cloning method based on improved dual quaternions, the traditional Euler angle method is easily affected by the universal joint deadlock problem, and a single quaternion is difficult to fully characterize the complex posture. This application can more accurately and comprehensively describe the posture state of the excavator through the improved dual quaternion method, avoiding possible mathematical singularity problems during operation. The method based on Riemann manifold can use the geometric structure of the manifold to accurately model in nonlinear space and capture complex behaviors. In addition, the introduction of improved dual quaternions can effectively describe multi-degree-of-freedom motion, and the Riemann manifold theory can better characterize the connection between the behavior trajectory in time and space, thereby improving the model's ability to express the excavator's operating behavior.

[0030] In one of the embodiments, the excavator is modified for electronic control; the modification includes: setting IMU sensors on the excavator's boom, dipper arm, bucket and body rotation mechanism, installing a proportional valve on the excavator's manual mechanical drive, and connecting a data acquisition card.

[0031] Specifically, IMU and other attitude acquisition sensors are placed on the excavator's boom (big arm), dipper (small arm), bucket and body slewing mechanism, and transmitted back to the Jetson Orin Nano industrial computer through the CAN bus to realize the behavior cloning data and the attitude data collection function of the Cartesian space coordinate point at the end of the excavator. In addition, for the excavator's drive system, it is necessary to install a proportional valve on the traditional manual mechanical drive, and then connect the data acquisition card to realize the mutual conversion from analog to digital, and then realize the electric control of the excavator.

[0032] In one embodiment, the forward dynamics model of the excavator is constructed by the DH parameter method; the DH parameters include: the distance between two adjacent axes along the x-axis , the angle of rotation of two adjacent axes along the x-axis , the distance between two adjacent axes translated along the z-axis , the angle of rotation of two adjacent axes along the z axis ; According to the forward dynamics model, the transfer matrix in the Cartesian coordinate system of the excavator end is obtained. for: ; in, , , , . , , , , , , , The attitude data collected by the IMU sensor; The translation variables of the excavator in the Cartesian coordinate system are: ; The rotation variable is expressed in Euler angles as: .

[0033] The forward dynamics model of the excavator is a spatial four-bar kinematic model, which can be obtained by DH ( ) parameter methods are shown in the following table:

[0034] Through the data of four IMUs placed on each body of the excavator, the expression of the excavator's position and posture in the Cartesian coordinate system can be fully obtained.

[0035] In one embodiment, according to the transfer matrix, the translation variable and the rotation variable, the quaternion representation of the rotation part of the end of the excavator is obtained as follows: ; in, , , and It is expressed in Cartesian coordinate system.

[0036] Furthermore, according to the quaternion representation, plus the translation part, a dual quaternion representation of the position and attitude of the entire excavator is obtained, including: .

[0037] In addition, according to the dual quaternion representation, the end position and attitude of the excavator are expressed as: ; in, is the position of the excavator end in the Cartesian coordinate system, It is the corner mark of the excavator's position and posture at a certain moment.

[0038] Through Riemannian manifold mapping, the position and posture of the excavator's end in the Cartesian coordinate system are expressed as a dual quaternion of the Riemannian manifold tangent space: .

[0039] In one embodiment, a non-parametric Bayesian Gaussian process model is constructed based on the projection of the behavior data of the excavator onto the Riemann manifold tangent space; the square exponential covariance function of the non-parametric Bayesian Gaussian process model is: ; in, , and is the hyperparameter of the Gaussian process model, and are the timestamps corresponding to the excavator behavior data components, is the Kronecker symbol; The timestamp of the given query , get the mean of the output mining behavior data component and variance They are: ; ; in, and are the covariance matrices of the timestamps corresponding to the excavator behavior data and the query data timestamps, is the corresponding cross covariance matrix. A set of cloned data vectors for the operator's behavior; The final behavior cloning result obtained by the inverse mapping of the Riemann manifold mapping is: ; in, Represents the behavior cloning result.

[0040] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0041] In one embodiment, Figure 2 As shown, a Riemann manifold excavator behavior cloning device based on improved dual quaternions is provided, comprising: a forward dynamics model building module 202, a quaternion representation module 204, a Riemann mapping module 206 and a cloning module 208, wherein: The forward dynamics model building module 202 is used to build the forward dynamics model of the excavator by using the DH parameter method, and obtain the transfer matrix in the Cartesian coordinate system of the excavator end, the translation variable and the rotation variable in the Cartesian coordinate system of the excavator according to the forward dynamics model.

[0042] The quaternion representation module 204 is used to obtain the quaternion representation of the rotating part of the excavator end according to the transfer matrix and the rotation variable, and to obtain the dual quaternion representation of the position and posture of the entire excavator according to the translation matrix and the quaternion representation.

[0043] The Riemann mapping module 206 is used to obtain the position and attitude representation of the end of the excavator according to the dual quaternion representation, and to represent the position and attitude of the end of the excavator in the Cartesian coordinate system as the Riemann manifold dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping.

[0044] The behavior cloning module 208 is used to construct a non-parametric Bayesian Gaussian process model based on the projection of the excavator's behavior data onto the Riemann manifold tangent space. According to the non-parametric Bayesian Gaussian process model, given a query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

[0045] In one of the embodiments, a transformation module is also included for performing electronic control transformation on the excavator; the transformation includes: setting IMU sensors on the excavator's boom, dipper arm, bucket and body rotation mechanism, installing a proportional valve on the excavator's manual mechanical drive, and connecting a data acquisition card.

[0046] In one embodiment, the forward dynamics model building module 202 is further used to build the forward dynamics model of the excavator by using the DH parameter method; the DH parameters include: the distance between two adjacent axes along the x-axis , the angle of rotation of two adjacent axes along the x-axis , the distance between two adjacent axes translated along the z-axis , the angle of rotation of two adjacent axes along the z axis ; According to the forward dynamics model, the transfer matrix in the Cartesian coordinate system of the excavator end is obtained. for: ; in, , , , . , , , , , , , The attitude data collected by the IMU sensor; The translation variables of the excavator in the Cartesian coordinate system are: ; The rotation variable is expressed in Euler angles as: .

[0047] In one embodiment, the quaternion representation module 204 is further used to obtain the quaternion representation of the rotating part of the excavator end according to the transfer matrix, the translation variable and the rotation variable: ; in, , , and It is expressed in Cartesian coordinate system.

[0048] In one embodiment, the quaternion representation module 204 is further used to obtain the dual quaternion representation of the position and attitude of the entire excavator according to the quaternion representation: .

[0049] In one embodiment, the quaternion representation module 204 is further used to obtain the end position and posture representation of the excavator according to the dual quaternion representation as follows: ; in, is the position of the excavator end in the Cartesian coordinate system, It is the corner mark of the excavator's position and posture at a certain moment.

[0050] In one embodiment, the Riemann mapping module 206 is further used to represent the position and posture of the end of the excavator in the Cartesian coordinate system as a Riemann manifold double quaternion of the Riemann manifold tangent space through Riemann manifold mapping: .

[0051] In one embodiment, the behavior cloning module 208 is further used to construct a non-parametric Bayesian Gaussian process model based on the projection of the behavior data of the excavator onto the Riemann manifold tangent space; the square exponential covariance function of the non-parametric Bayesian Gaussian process model is: ; in, , and is the hyperparameter of the Gaussian process model, and are the timestamps corresponding to the excavator behavior data components, is the Kronecker symbol; The timestamp of the given query , get the mean of the output mining behavior data component and variance They are: ; ; in, and are the covariance matrices of the timestamps corresponding to the excavator behavior data and the query data timestamps, is the corresponding cross covariance matrix. A set of cloned data vectors for the operator's behavior; The final behavior cloning result obtained by the inverse mapping of the Riemann manifold mapping is: ; in, Represents the behavior cloning result.

[0052] For the specific definition of the Riemann manifold excavator behavior cloning device based on improved dual quaternions, please refer to the definition of the Riemann manifold excavator behavior cloning method based on improved dual quaternions in the above text, which will not be repeated here. Each module in the above-mentioned Riemann manifold excavator behavior cloning device based on improved dual quaternions can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0053] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a behavior cloning method of a Riemann manifold excavator based on an improved dual quaternion is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0054] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0055] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0056] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0057] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A Riemannian manifold excavator behavior cloning method based on improved dual quaternion, characterized in that: The method comprises: A forward dynamics model of the excavator is constructed by the DH parameter method, and a transfer matrix in a Cartesian coordinate system at the end of the excavator, a translation variable and a rotation variable in a Cartesian coordinate system at the end of the excavator are obtained according to the forward dynamics model; According to the transfer matrix and the rotation variable, a quaternion representation of the rotating part of the excavator end is obtained, and according to the translation variable and the quaternion representation, a dual quaternion representation of the position and attitude of the entire excavator is obtained; According to the dual quaternion representation, the position and posture of the end of the excavator in the Cartesian coordinate system are represented as a Riemann manifold dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping; According to the projection of the behavior data of the excavator operator onto the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed. According to the non-parametric Bayesian Gaussian process model, given the query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

2. The method according to claim 1, characterized in that The method further comprises: The excavator is modified for electronic control; the modification includes: setting IMU sensors on the excavator's boom, dipper arm, bucket and body rotation mechanism, installing a proportional valve on the excavator's manual mechanical drive, and connecting a data acquisition card.

3. The method according to claim 2, characterized in that The forward dynamics model of the excavator is constructed by the DH parameter method. According to the forward dynamics model, the transfer matrix in the Cartesian coordinate system of the excavator end, the translation variable and the rotation variable in the Cartesian coordinate system of the excavator are obtained, including: The forward dynamics model of the excavator is constructed by the DH parameter method; the DH parameters include: the distance between two adjacent axes along the x-axis , the angle of rotation of two adjacent axes along the x-axis , the distance between two adjacent axes translated along the z-axis , the angle of rotation of two adjacent axes along the z axis ; According to the forward dynamics model, the transfer matrix in the Cartesian coordinate system of the excavator end is obtained. for: in, , , , , , , , , , , , These are the posture data collected by the IMU sensors placed on the excavator's boom, arm, bucket, and body slewing mechanism; The translation variables in the Cartesian coordinate system of the excavator are: The rotation variable is expressed in Euler angles as: 。 4. The method according to claim 3, characterized in that According to the transfer matrix and the rotation variable, the quaternion representation of the rotating part of the excavator end is obtained, including: According to the transfer matrix, translation variables and rotation variables, the quaternion expression of the rotation part of the excavator end is obtained as follows: in, , , and It is expressed in Cartesian coordinate system.

5. The method according to claim 4, characterized in that According to the translation variable and the quaternion representation, the dual quaternion representation of the position and attitude of the entire excavator is obtained as follows: 。 6. The method according to claim 5, characterized in that According to the dual quaternion representation, the end position and attitude representation of the excavator are obtained, including: According to the dual quaternion representation, the end position and attitude of the excavator are expressed as: in, is the position of the excavator end in the Cartesian coordinate system, It is the corner mark of the excavator's position and posture at a certain moment.

7. The method according to claim 6, characterized in that Through Riemannian manifold mapping, the position and posture of the excavator's end in the Cartesian coordinate system are expressed as a dual quaternion representation in the Riemannian manifold tangent space, including: Through Riemannian manifold mapping, the position and posture of the excavator's end in the Cartesian coordinate system are expressed as a dual quaternion in the Riemannian manifold tangent space: 。 8. The method according to claim 7, characterized in that According to the projection of the behavior data of the excavator to the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed. According to the non-parametric Bayesian Gaussian process model, given the query timestamp, the mean and variance of the output behavior data component are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping, including: According to the projection of the excavator's behavior data onto the Riemann manifold tangent space, a non-parametric Bayesian Gaussian process model is constructed; the square exponential covariance function of the non-parametric Bayesian Gaussian process model is: in, , and is the hyperparameter of the Gaussian process model, and are the timestamps corresponding to the excavator behavior data components, is the Kronecker symbol; The timestamp of the given query , get the mean of the output mining behavior data component and variance They are: in, and are the covariance matrices of the timestamps corresponding to the excavator behavior data and the query data timestamps, is the corresponding cross-covariance matrix, A set of cloned data vectors for the operator's behavior; The final behavior cloning result obtained by the inverse mapping of the Riemann manifold mapping is: in, Represents the behavior cloning result.

9. A Riemannian manifold excavator behavior cloning device based on improved dual quaternions, characterized in that: The device comprises: A forward dynamics model building module is used to build a forward dynamics model of the excavator by using a DH parameter method, and obtain a transfer matrix in a Cartesian coordinate system of the excavator end, a translation variable and a rotation variable in a Cartesian coordinate system of the excavator according to the forward dynamics model; A quaternion representation module, used to obtain a quaternion representation of the rotating part of the excavator end according to the transfer matrix and the rotation variable, and to obtain a dual quaternion representation of the position and attitude of the entire excavator according to the translation variable and the quaternion representation; A Riemann mapping module is used to obtain the position and attitude representation of the end of the excavator according to the dual quaternion representation, and to represent the position and attitude of the end of the excavator in the Cartesian coordinate system as a Riemann manifold dual quaternion representation of the Riemann manifold tangent space through Riemann manifold mapping; The behavior cloning module is used to construct a non-parametric Bayesian Gaussian process model based on the projection of the excavator's behavior data onto the Riemann manifold tangent space. According to the non-parametric Bayesian Gaussian process model, given a query timestamp, the mean and variance of the output behavior data components are calculated, and the final behavior cloning result is obtained through the inverse mapping of the Riemann manifold mapping.

Images