A method for detecting the deformation of the end effector of a robotic arm, and a robotic arm controller.
By installing IMU sensors on the robotic arm, fitting the deflection curve, and constructing a decoupling matrix, the real-time and accuracy problems of deformation detection at the end of the heavy-duty robotic arm were solved, realizing high-frequency and accurate deformation calculation, and providing a reliable basis for subsequent control.
Patent Information
- Application Number
- CN202611237807.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-14
- Publication Date
- 2026-11-17
AI Technical Summary
Existing robotic arm control systems neglect the impact of structural flexibility under heavy load and long reach conditions, resulting in significant deviations between the actual and theoretical poses of the end effector, affecting operational accuracy and safety. Existing detection methods involve large computational loads and poor real-time performance, making it difficult to meet the requirements of high-frequency, real-time deformation detection.
By installing inertial measurement unit (IMU) sensors on the robotic arm, attitude information is collected in real time, deflection curves are fitted, a decoupling matrix is constructed, the deformation of the upper and lower arms is calculated, and the decoupling calculation is performed through joint transmission effects, thus achieving rapid and accurate online detection of end-effector deformation.
It enables rapid and accurate online detection of deformation at the end of the robotic arm, providing a high-precision basis for deformation compensation, and is suitable for real-time deformation monitoring and control of large, heavy-duty robotic arms.
Smart Images

Figure CN123111176A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a method for detecting the deformation of the end effector of a robotic arm and a robotic arm controller, belonging to the field of robotic arm technology. Background Technology
[0002] With the development of mining, metallurgy, and heavy equipment manufacturing towards large-scale and automated operations, large, heavy-duty robotic arms are widely used in high-load tasks within complex spaces, such as handling and assembling heavy components. These robotic arms typically feature long reach, high load-bearing capacity, and large structural mass. During actual operation, the upper and lower arms undergo significant elastic deformation under their own weight and end-effector loads. Existing robotic arm control systems mostly establish kinematic and dynamic models based on the rigid body assumption (here, establishing kinematic and dynamic models based on the rigid body assumption means assuming that each component of the robotic arm is an absolutely rigid "rigid body" and will not deform under force, and using purely mathematical formulas to calculate the position of the end effector based on the joint rotation angle and arm length), neglecting the influence of structural flexibility. Under heavy loads and with a long reach, the bending deformation of the upper and lower arms of the robotic arm leads to a significant deviation between the actual and theoretical poses of the end effector, seriously affecting operational accuracy and safety.
[0003] Current research methods for detecting the deformation of robotic arms mainly include offline compensation methods based on finite element analysis (which involves dividing the robotic arm structure into countless tiny networks using digital simulation technology and calculating the deformation of each network under different forces using complex physical equations) or indirect estimation methods based on force sensors (which involves measuring the magnitude of the external force currently being applied by installing force / torque sensors). However, these methods generally suffer from problems such as large computational load, poor real-time performance, or difficulty in directly reflecting the overall deformation state, making it difficult to meet the needs of online, high-frequency, and real-time deformation detection for large, heavy-duty robotic arms.
[0004] Chinese invention patent application CN119795177A discloses a pose compensation method and system for the end effector of a heavy-duty robotic arm. The method involves constructing finite element models for several link units that make up the robotic arm; solving each finite element model to obtain the deformation at the load end under different forces and link lengths; and constructing a mapping model of the deformation corresponding to the deformation of each link unit under different forces based on the deformation. The actual force on each link unit is calculated based on the actual load of the actuator and the link structure parameters. For each link unit, the link deformation is calculated using the mapping model of deformation and the actual force. Based on the homogeneous transformation matrix of the forward kinematics, the end effector offset caused by the deformation of each link when it deforms individually is obtained. All actuator offsets are summed to generate a compensation amount for end effector pose compensation, thereby achieving pose compensation for the end effector of the robotic arm and improving the operating efficiency and accuracy of the robotic arm. In this patent (CN119795177A), when calculating the deformation of the link, each link unit is regarded as an independent deformable body. Only the independent deformation caused by the force on each link itself is considered. The coupling effect of the deformation of the robot arm transmitted through the joint on the adjacent links and the end effector is not fully considered. Under actual heavy load conditions, the deformation of each link does not exist in isolation. The bending deformation of the upper arm will change the position reference of the lower arm through the joint. The deformation of the lower arm will further affect the spatial position of the end effector. This deformation coupling response is particularly significant in the structure of heavy load and large length-to-diameter ratio robot arm. If this coupling effect is ignored, the calculation accuracy of the end effector position compensation will be insufficient, making it difficult to meet the requirements of high-precision operation. Summary of the Invention
[0005] The purpose of this invention is to provide a method for detecting the deformation of the end effector of a robotic arm and a robotic arm controller, so as to solve the problem of inaccurate measurement of end effector deformation under heavy load and long boom extension conditions.
[0006] To address the aforementioned technical problems, the first aspect of this invention provides a method for detecting the deformation of the end effector of a robotic arm, comprising the following steps:
[0007] 1) When the robotic arm under test moves to any shape, the deflection curve of the robotic arm under test is fitted based on the position data and offset angle data of several measuring points pre-set on the robotic arm under test.
[0008] 2) Based on the deflection curve and the theoretical data obtained in advance under the rigid body configuration of the robot arm under test, the initial deformation of the robot arm under test is obtained.
[0009] 3) Based on the pre-acquired geometric parameters of the robotic arm under test and the current shape of the robotic arm under test, construct a decoupling matrix to represent the coupling effect of the robotic arm under test on the end effector through joint transmission.
[0010] 4) Based on the initial deformation and decoupling matrix, the actual deformation of the end of the robotic arm under test is obtained.
[0011] In one possible implementation, when the robotic arm under test comprises a large arm fixed to the base end, a small arm fixed to the end of the large arm, and an end effector fixed to the end of the small arm, the initial deformation of the robotic arm under test is obtained by the following method based on the deflection curve and pre-acquired theoretical data under the rigid body configuration of the robotic arm under test:
[0012] At the same measuring point, the measuring point value on the deflection curve is compared with the theoretical value of the measuring point in the theoretical data to calculate the initial deformation of the measuring point, and the initial deformation of all measuring points is obtained.
[0013] Based on the location of the measuring points, all measuring points are divided into measuring points belonging to the upper arm and measuring points belonging to the lower arm;
[0014] The initial deformation of the boom is obtained based on the measuring points belonging to the boom and their initial deformation; the initial deformation of the forearm is obtained based on the measuring points belonging to the forearm and their initial deformation.
[0015] The initial deformation of the upper arm and the initial deformation of the lower arm are used as the initial deformation of the robotic arm under test.
[0016] In one possible implementation, step 3) includes: constructing a decoupling matrix to represent the coupling effect of the upper arm on the lower arm and the end effector through joint transmission, based on the pre-acquired geometric parameters of the upper arm of the robot under test and the current shape of the robot under test.
[0017] In one possible implementation, the decoupling matrix is represented by the following formula:
[0018] ;
[0019] in, This is the decoupling matrix; for Compared to The homogeneous transformation matrix; for Compared to The homogeneous transformation matrix; The end-effector posture of the upper arm under the theoretical rigid body model; Use the base coordinate system; This is the end position of the boom after deformation.
[0020] In one possible implementation, the actual deformation amount is calculated using the following formula:
[0021] ;
[0022] Where U is the actual deformation; This is the decoupling matrix; This represents the initial deformation of the upper arm; This represents the initial deformation of the forearm.
[0023] In one possible implementation, based on the position data and offset angle data of several measuring points on the robotic arm under test, a relationship function between position and offset angle is established by cubic spline interpolation or least squares polynomial fitting; the relationship function is integrated to obtain a continuous deflection curve, and the continuous deflection curve is used as the deflection curve of the robotic arm under test.
[0024] In one possible implementation, the offset angle data of several measuring points pre-set on the robotic arm under test are obtained by the following method:
[0025] Attitude data for each measuring point is collected by sensors pre-set at each measuring point;
[0026] Based on the pre-constructed coordinate system and attitude data of each sensor, the offset angle data between the cantilever beam axis and the gravity direction of the robot arm under test at each sensor position is calculated, and the offset angle data is used as the offset angle data of the corresponding sensor's measurement point.
[0027] In one possible implementation, the position data of several measuring points pre-set on the robotic arm under test are obtained by the following method:
[0028] A base coordinate system is constructed for the robot arm under test, wherein the cantilever beam axis of the robot arm under test is used as the coordinate axis of the base coordinate system;
[0029] Based on the base coordinate system, the coordinate data of each measuring point on the robot arm under test is determined, and the coordinate value of the cantilever beam axis in the coordinate data of each measuring point is used as the position data of the corresponding measuring point.
[0030] To address the aforementioned technical problems, a second aspect of the present invention provides a robotic arm controller, including a processor, the processor being configured to execute a computer program to implement the steps of the method described below:
[0031] 1) When the robotic arm under test moves to any shape, the deflection curve of the robotic arm under test is fitted based on the position data and offset angle data of several measuring points pre-set on the robotic arm under test.
[0032] 2) Based on the deflection curve and the theoretical data obtained in advance under the rigid body configuration of the robot arm under test, the initial deformation of the robot arm under test is obtained.
[0033] 3) Based on the pre-acquired geometric parameters of the robotic arm under test and the current shape of the robotic arm under test, construct a decoupling matrix to represent the coupling effect of the robotic arm under test on the end effector through joint transmission.
[0034] 4) Based on the initial deformation and decoupling matrix, the actual deformation of the end of the robotic arm under test is obtained.
[0035] In one possible implementation, when the robotic arm under test comprises a large arm fixed to the base end, a small arm fixed to the end of the large arm, and an end effector fixed to the end of the small arm, the initial deformation of the robotic arm under test is obtained by the following method based on the deflection curve and pre-acquired theoretical data under the rigid body configuration of the robotic arm under test:
[0036] At the same measuring point, the measuring point value on the deflection curve is compared with the theoretical value of the measuring point in the theoretical data to calculate the initial deformation of the measuring point, and the initial deformation of all measuring points is obtained.
[0037] Based on the location of the measuring points, all measuring points are divided into measuring points belonging to the upper arm and measuring points belonging to the lower arm;
[0038] The initial deformation of the boom is obtained based on the measuring points belonging to the boom and their initial deformation; the initial deformation of the forearm is obtained based on the measuring points belonging to the forearm and their initial deformation.
[0039] The initial deformation of the upper arm and the initial deformation of the lower arm are used as the initial deformation of the robotic arm under test.
[0040] In one possible implementation, step 3) includes: constructing a decoupling matrix to represent the coupling effect of the upper arm on the lower arm and the end effector through joint transmission, based on the pre-acquired geometric parameters of the upper arm of the robot under test and the current shape of the robot under test.
[0041] In one possible implementation, the decoupling matrix is represented by the following formula:
[0042] ;
[0043] in, This is the decoupling matrix; for Compared to The homogeneous transformation matrix; for Compared to The homogeneous transformation matrix; The end-effector posture of the upper arm under the theoretical rigid body model; Use the base coordinate system; This is the end position of the boom after deformation.
[0044] In one possible implementation, the actual deformation amount is calculated using the following formula:
[0045] ;
[0046] Where U is the actual deformation; This is the decoupling matrix; This represents the initial deformation of the upper arm; This represents the initial deformation of the forearm.
[0047] In one possible implementation, based on the position data and offset angle data of several measuring points on the robotic arm under test, a relationship function between position and offset angle is established by cubic spline interpolation or least squares polynomial fitting; the relationship function is integrated to obtain a continuous deflection curve, and the continuous deflection curve is used as the deflection curve of the robotic arm under test.
[0048] In one possible implementation, the offset angle data of several measuring points pre-set on the robotic arm under test are obtained by the following method:
[0049] Attitude data for each measuring point is collected by sensors pre-set at each measuring point;
[0050] Based on the pre-constructed coordinate system and attitude data of each sensor, the offset angle data between the cantilever beam axis and the gravity direction of the robot arm under test at each sensor position is calculated, and the offset angle data is used as the offset angle data of the corresponding sensor's measurement point.
[0051] In one possible implementation, the position data of several measuring points pre-set on the robotic arm under test are obtained by the following method:
[0052] A base coordinate system is constructed for the robot arm under test, wherein the cantilever beam axis of the robot arm under test is used as the coordinate axis of the base coordinate system;
[0053] Based on the base coordinate system, the coordinate data of each measuring point on the robot arm under test is determined, and the coordinate value of the cantilever beam axis in the coordinate data of each measuring point is used as the position data of the corresponding measuring point.
[0054] The beneficial effects of this invention are as follows: When the robotic arm under test moves to any shape, the deflection curve of the robotic arm under test is fitted based on the position data and offset angle data of several measuring points pre-set on the robotic arm under test; the initial deformation of the robotic arm under test is obtained according to the deflection curve and the theoretical data of the rigid body configuration of the robotic arm under test obtained in advance; a decoupling matrix is constructed to represent the coupling effect of the robotic arm under test on the end effector through joint transmission based on the geometric parameters of the robotic arm under test obtained in advance and the current shape of the robotic arm under test; the actual deformation of the end effector of the robotic arm under test is obtained according to the initial deformation and the decoupling matrix, realizing online detection that can acquire structural deformation information in real time during the movement of the robotic arm and accurately calculate the deformation of the end effector. Attached Figure Description
[0055] Figure 1 This is a flowchart of a method for detecting the deformation of the end effector of a robotic arm proposed in this invention;
[0056] Figure 2 This is a flowchart illustrating the calculation of end-effector deformation in a practical application scenario using a method for detecting end-effector deformation proposed in this invention.
[0057] Figure 3 This is a schematic diagram of the arrangement of several sensors in a practical application scenario for the method for detecting the deformation of the end effector of a robotic arm proposed in this invention.
[0058] Figure 4 This is a schematic diagram of the cantilever beam bending curve fitting and end deformation calculation in a practical application scenario of the method for detecting the deformation at the end of a robotic arm proposed in this invention.
[0059] Figure 5 This is a structural diagram of a robotic arm controller proposed in this invention;
[0060] The reference numerals in the figures include:
[0061] 1-20: IMU sensor. Detailed Implementation
[0062] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.
[0063] The inventive concept of this invention lies in: by real-time sensing and decoupled calculation of the deformation state of the upper and lower arms of the robotic arm, the deformation of the robotic arm end effector can be calculated quickly and accurately online, providing a reliable basis for subsequent deformation compensation and high-precision control.
[0064] It is important to emphasize that the basic principle of this invention is as follows: the upper arm and lower arm of the robotic arm are respectively equivalent to cantilever beam structures (the cantilever beam structure refers to a beam structure that is fixed at one end and free at the other end). Multiple IMU (Inertial Measurement Unit) sensors are evenly arranged along the length of the upper arm and lower arm to collect attitude information at different positions in real time. Using the offset angle between the gravity direction and the link axis measured by the IMU, the actual bending curves of the upper arm and lower arm are fitted to calculate their vertical deformation relative to the theoretical rigid body configuration. Furthermore, a decoupling matrix is established to determine the influence of the upper arm deformation on the lower arm and end-effector deformation. Finally, by combining the upper arm deformation, lower arm deformation, and decoupling compensation, the real-time deformation of the robotic arm end-effector is calculated online.
[0065] Detailed implementation method 1:
[0066] like Figure 1 The diagram shows a flowchart of a method for detecting the deformation of the end effector of a robotic arm proposed in this invention, which includes steps S11, S12, S13, and S14, specifically:
[0067] Step S11: When the robotic arm under test moves to an arbitrary shape, the deflection curve of the robotic arm is fitted based on the position data and offset angle data of several measuring points pre-set on the robotic arm. Here, "the robotic arm moves to an arbitrary shape" means that after each joint of the robotic arm rotates to any specific angle according to the preset motion plan, the end effector and each stage of the linkage present their pose in space. It should be noted that in actual application scenarios, the deformation of the robotic arm comes from the combined effect of its own gravity and the end load. The deflection curve refers to the smooth curve programmed along the axis after the beam deforms. The several measuring points refer to several measurement points pre-set on the robotic arm under test, preferably the measuring points are evenly distributed on the cantilever beam axis of the robotic arm under test.
[0068] The position data and offset angle data of several measuring points pre-set on the robotic arm under test are obtained by the following method:
[0069] The posture data of each measuring point is collected by sensors that are pre-set on each measuring point; based on the coordinate system and posture data of each sensor that are pre-constructed for each sensor, the offset angle data between the cantilever beam axis of the robot arm under test and the direction of gravity at each sensor position is calculated, and the offset angle data is used as the offset angle data of the corresponding measuring point of the sensor.
[0070] The position data of several measuring points pre-set on the robotic arm under test are obtained by the following method: a base coordinate system is constructed for the robotic arm under test, wherein the cantilever beam axis of the robotic arm under test is used as the coordinate axis of the base coordinate system; based on the base coordinate system, the coordinate data of each measuring point on the robotic arm under test is determined, and the coordinate value of the cantilever beam axis in the coordinate data of each measuring point is used as the position data of the corresponding measuring point.
[0071] Based on the position data and offset angle data of several measuring points on the robotic arm under test, a relationship function between position and offset angle is established by cubic spline interpolation or least squares polynomial fitting; the relationship function is integrated to obtain a continuous deflection curve, and the continuous deflection curve is used as the deflection curve of the robotic arm under test.
[0072] Step S12: Based on the deflection curve and the theoretical data of the rigid body configuration of the robot arm under test, the initial deformation of the robot arm under test is obtained. Here, the deformation value of all measuring points on the robot arm under test is obtained by comparing the data value of each measuring point on the deflection curve with the theoretical data value of that measuring point, thereby obtaining the initial deformation value of the robot arm under test. Preferably, the maximum deformation is taken as the initial deformation of the robot arm under test.
[0073] Step S13: Based on the pre-acquired geometric parameters of the robotic arm under test and the current shape of the robotic arm under test, construct a decoupling matrix to represent the coupling effect of the robotic arm under test on the end effector through joint transmission.
[0074] Step S14: Based on the initial deformation and decoupling matrix, obtain the actual deformation of the end effector of the robotic arm under test.
[0075] Through the above steps S11-14, real-time sensing and decoupled calculation of the robotic arm under test are achieved, enabling rapid and accurate online calculation of the deformation at the end of the robotic arm, providing a reliable basis for subsequent deformation compensation and high-precision control.
[0076] Method Detailed Implementation 2:
[0077] The following section will discuss practical application scenarios, in conjunction with the accompanying diagrams in the instruction manual. Figure 2-4 The present invention provides an explanation of a method for detecting the deformation of the end effector of a robotic arm.
[0078] like Figure 2 The diagram shows a flowchart of a method for detecting the deformation of the end effector of a robotic arm proposed in this invention, applied in a practical scenario. The flowchart includes steps one, two, three, four, and five, specifically:
[0079] Step 1: Equivalent Modeling of the Cantilever Beam of the Robotic Arm and Arrangement of Inertial Measurement Units
[0080] The large, heavy-duty robotic arm is modeled as a cantilever beam structure, with the upper arm fixed to the base end and the lower arm fixed to the end of the upper arm. N1 and N2 IMU sensors are mounted along the length of the upper and lower arms, respectively (see...). Figure 3 As shown, IMU sensors 1 to 20 are installed along the length of the upper arm and forearm, respectively. Base coordinate systems are established for each. (i.e. Figure 4 The rightmost coordinate system ) and the local coordinate systems of each IMU Where i=n (n is the total number of installed IMU sensors, and n is a positive integer greater than 1), the length sequence of each IMU sensor's installation position relative to the base coordinate system is defined. (n is the total number of installed IMU sensors, and n is a positive integer greater than 1). Assume that in the IMU sensor's own coordinate system, the -Y axis is aligned with the direction of gravity, and the Z axis is aligned with the axial direction of the corresponding cantilever beam, providing a unified reference for subsequent attitude measurement and deformation calculation.
[0081] Step 2: Attitude data acquisition and angle feature extraction of the inertial measurement unit
[0082] When the robotic arm moves to any position and bears the actual workload, the attitude data of each IMU sensor is collected in real time. Based on the attitude data measured by the IMU sensors, the angular components of each IMU sensor position in the direction of gravity caused by the elastic deformation of the robotic arm are calculated. This yields the offset angle between the cantilever beam axis and the direction of gravity at each IMU sensor position, that is, the rotation angle of each IMU sensor around its own X-axis before and after deformation. This is used as an angular feature quantity to construct a discrete angle dataset of the upper and lower arms along the axial direction of the robotic arm. This discrete angle dataset contains the rotation angle of all IMU sensors around their own X-axis, that is, the dataset contains all offset angles.
[0083] In addition, it should be emphasized that in scenarios where the robotic arm moves to any shape and bears actual work loads, the deformation state mainly manifests in the plane formed by the cantilever beam axis and the direction of gravity.
[0084] Step 3: Fitting the bending curves of the upper arm and forearm and calculating the deformation.
[0085] The host computer system acquires attitude data from each IMU sensor in real time via wireless or wired transmission, and performs bending curve fitting on the upper arm and forearm respectively. Specifically, based on the discrete angle dataset and the attitude data from each IMU sensor, cubic spline interpolation or least-squares polynomial fitting is used to establish a function of angle with respect to length L. Then through integration... A continuous deflection curve is obtained. The calculated actual bending curve is compared with the theoretical value under the rigid body configuration of the robotic arm, and the maximum deformation of the upper arm and lower arm along the direction of gravity is calculated, denoted as the deformation of the upper arm itself. and the deformation of the forearm itself The self-deformation mentioned above is also the initial deformation. See also Figure 4 As shown, in the coordinate system of the n IMU sensors themselves, the deformation of the measuring point where each IMU sensor is located is obtained. The largest deformation among the measuring points where all IMU sensors are installed on the upper arm is taken as the deformation of the upper arm itself, and the largest deformation among the measuring points where all IMU sensors are installed on the lower arm is taken as the deformation of the lower arm itself.
[0086] Step 4: Constructing the decoupling relationship between boom deformation and end-effector deformation
[0087] Considering that the deformation of the upper arm will have a coupling effect on the forearm and end effector through joint transmission, a coordinate mapping matrix before and after the deformation of the upper arm is established. This method decouples the calculation of displacement changes caused by unit deformation of the boom in the end-effector coordinate system. The decoupling matrix is... It can be obtained from the geometric parameters of the robotic arm and the current joint configuration, and the calculation formula is as shown in equation (1):
[0088]
[0089] Among them, coordinate system Given that represents the end-effector posture under a theoretical rigid body model, which is the end-effector posture under ideal / undeformed conditions, in practical applications, the theoretical rigid body model is also the robotic arm model constructed using a three-dimensional digital model. Additionally, see [link to relevant documentation]. Figure 4 The theoretical rigid body model shows that the end posture of the boom is on the theoretical cantilever beam. It is the end posture of the boom after deformation, measured by an IMU sensor installed at the end of the boom; It is a known base coordinate system. Based on this, establish respectively and Compared to homogeneous transformation matrix and This is to determine the relationship between the end effector's pose before and after the boom deformation, and to decouple the displacement of the robotic arm's end effector caused by the boom deformation. That is, This is the decoupling matrix; for Compared to The homogeneous transformation matrix; for Compared to The homogeneous transformation matrix; The end-effector posture of the upper arm under the theoretical rigid body model; Use the base coordinate system; This is the end position of the boom after deformation.
[0090] Step 5: Calculation of real-time deformation at the robotic arm end effector
[0091] The deformation of the boom obtained in step three Forearm deformation and the decoupling matrix established in step four The real-time deformation U of the robotic arm's end effector under the current working conditions is calculated. The calculation method simultaneously considers the superposition of linear displacements of the upper and lower arms and the coupling of the upper arm deformation, and its expression is:
[0092] Where U is the real-time deformation of the robotic arm end effector relative to the theoretical rigid body model, which can be directly used for online monitoring or feedback control. This is the decoupling matrix; This refers to the deformation of the boom (i.e., the initial deformation of the boom); operator Used to find right The magnitude of the influence; This is the forearm deformation amount (i.e., the initial forearm deformation amount).
[0093] Through steps one through five above, the deformation state of the upper and lower arms of the robotic arm can be perceived and decoupled in real time, enabling rapid and accurate online calculation of the deformation at the end of the robotic arm, providing a reliable basis for subsequent deformation compensation and high-precision control.
[0094] Detailed implementation of the robotic arm controller:
[0095] The present invention provides a robotic arm controller, which can be any control unit in the original robotic arm system that meets the computer performance requirements, such as an embedded motion controller, an industrial computer, or a dedicated DSP, or it can be a newly added independent controller, used to execute the method for measuring the end effector of a robotic arm proposed in this invention. If a new controller is used, it needs to be connected to the encoders, sensors, end effectors, etc. of each joint of the robotic arm in order to collect the data required by the method for measuring the end effector of the robotic arm.
[0096] A schematic diagram of the robotic arm controller is shown below. Figure 5As shown, the system includes a memory, a processor, a system bus, and a computer program stored in the memory. The processor and memory communicate and exchange data via the system bus. The processor executes the computer program to implement the steps of the method for measuring the phase change of the robotic arm end effector according to the present invention. The processor can be a microprocessor (MCU) or other processing device; the memory can be any type of memory that stores information using electrical energy, such as non-volatile storage media (including computer programs, databases), or other types of memory.
[0097] In summary, this invention proposes a method for detecting the deformation of the end effector of a robotic arm and a robotic arm controller, which are applied in the field of heavy-duty robotic arm state perception and control technology. Based on an inertial measurement unit (IMU) sensor, it enables online detection of the deformation of the end effector of a large, heavy-duty robotic arm, achieving the following beneficial effects:
[0098] 1) High real-time performance: Based on IMU sensor online data acquisition, without the need for complex finite element solution, it can realize real-time detection of the deformation of the robotic arm.
[0099] 2) Good engineering adaptability: The boom and forearm are equivalent to a cantilever beam model, with clear structural assumptions, making it suitable for a variety of large and heavy-duty robotic arms.
[0100] 3) High detection accuracy: By fitting the bending curve with multi-point IMU attitude information, the overall deformation state of the connecting rod can be accurately reflected.
[0101] 4) Simple algorithm: It transforms complex finite element nonlinear calculations into decoupled operations based on geometric relationships, which significantly reduces the computational load on the vehicle controller.
[0102] 5) High system integration: The system does not rely on external optical measurement equipment (such as laser trackers) and can operate independently in complex and obstructed environments such as open-pit mines and metallurgical workshops.
Claims
1. A method for detecting the deformation of the end effector of a robotic arm, characterized in that, Includes the following steps: 1) When the robotic arm under test moves to any shape, the deflection curve of the robotic arm under test is fitted based on the position data and offset angle data of several measuring points pre-set on the robotic arm under test. 2) Based on the deflection curve and the theoretical data obtained in advance under the rigid body configuration of the robot arm under test, the initial deformation of the robot arm under test is obtained. 3) Based on the pre-acquired geometric parameters of the robotic arm under test and the current shape of the robotic arm under test, construct a decoupling matrix to represent the coupling effect of the robotic arm under test on the end effector through joint transmission. 4) Based on the initial deformation and decoupling matrix, the actual deformation of the end of the robotic arm under test is obtained.
2. The method for detecting the deformation of the robotic arm end effector according to claim 1, characterized in that, When the robotic arm under test includes a main arm fixed to the base, a forearm fixed to the end of the main arm, and an end effector fixed to the end of the forearm, the initial deformation of the robotic arm under test is obtained by the following method based on the deflection curve and the theoretical data of the rigid body configuration of the robotic arm under test obtained in advance: At the same measuring point, the measuring point value on the deflection curve is compared with the theoretical value of the measuring point in the theoretical data to calculate the initial deformation of the measuring point, and the initial deformation of all measuring points is obtained. Based on the location of the measuring points, all measuring points are divided into measuring points belonging to the upper arm and measuring points belonging to the lower arm; The initial deformation of the boom is obtained based on the measuring points belonging to the boom and their initial deformation; the initial deformation of the forearm is obtained based on the measuring points belonging to the forearm and their initial deformation. The initial deformation of the upper arm and the initial deformation of the lower arm are used as the initial deformation of the robotic arm under test.
3. The method for detecting the deformation of the robotic arm end effector according to claim 2, characterized in that, Step 3) includes: constructing a decoupling matrix based on the pre-acquired geometric parameters of the upper arm of the robot under test and the current shape of the robot under test, to represent the coupling effect of the upper arm on the lower arm and the end effector through joint transmission.
4. The method for detecting the deformation of the robotic arm end effector according to claim 3, characterized in that, The decoupling matrix is represented by the following formula: ; in, This is the decoupling matrix; for Compared to The homogeneous transformation matrix; for Compared to The homogeneous transformation matrix; The end-effector posture of the upper arm under the theoretical rigid body model; Use the base coordinate system; This is the end position of the boom after deformation.
5. The method for detecting the deformation of the robotic arm end effector according to claim 4, characterized in that, The actual deformation amount is calculated using the following formula: ; Where U is the actual deformation; This is the decoupling matrix; This represents the initial deformation of the upper arm; This represents the initial deformation of the forearm.
6. The method for detecting the deformation of the robotic arm end effector according to claim 1, characterized in that, Based on the position data and offset angle data of several measuring points on the robotic arm under test, cubic spline interpolation or least squares polynomial fitting is used to establish a function relating position and offset angle. The relationship function is integrated to obtain a continuous deflection curve, which is then used as the deflection curve of the robotic arm under test.
7. The method for detecting the deformation of the robotic arm end effector according to claim 6, characterized in that, The offset angle data of several measuring points pre-set on the robotic arm under test are obtained by the following method: Attitude data for each measuring point is collected by sensors pre-set at each measuring point; Based on the pre-constructed coordinate system and attitude data of each sensor, the offset angle data between the cantilever beam axis and the gravity direction of the robot arm under test at each sensor position is calculated, and the offset angle data is used as the offset angle data of the corresponding sensor's measurement point.
8. The method for detecting the deformation of the robotic arm end effector according to claim 6, characterized in that, The position data of several measuring points pre-set on the robotic arm under test are obtained using the following method: A base coordinate system is constructed for the robot arm under test, wherein the cantilever beam axis of the robot arm under test is used as the coordinate axis of the base coordinate system; Based on the base coordinate system, the coordinate data of each measuring point on the robot arm under test is determined, and the coordinate value of the cantilever beam axis in the coordinate data of each measuring point is used as the position data of the corresponding measuring point.
9. A robotic arm controller, comprising a processor, characterized in that, The processor is used to execute a computer program to implement the steps of the method according to any one of claims 1-8.
Citation Information
Patent Citations
Pose compensation method and system for end effector of heavy-load mechanical arm
CN119795177A
