A parameter calibration system and method for an excavator working mechanism

Abstract

This invention belongs to the field of mechanical automation technology, specifically a parameter calibration system and method for an excavator's working mechanism. It mainly relates to the automatic acquisition of multi-joint mechanical parameters; particularly, it relates to a kinematics-based calculation method. This invention is primarily applied in the field of automated operation of serial robots, especially for the positioning of the excavator's end point, to achieve the acquisition of machine data with unknown parameters.

Description

Technical Field

[0001] This invention relates to the field of mechanical automation technology, specifically to a parameter calibration system and method for an excavator's working mechanism. Background Technology

[0002] Excavators are widely used, but their operation is difficult, leading to the application of end-point positioning technology. End-point positioning can be applied to the automated and auxiliary operations of excavators. During end-point positioning, it is necessary to obtain the dimensional parameters of each joint of the excavator, such as the boom, stick, and bucket, in advance. After installing the sensors, the installation errors of each sensor need to be measured and calibrated. Measuring the dimensions of large excavators is difficult, and the methods for measuring sensor installation errors are complex and prone to significant errors. Taking the standard measurement and calibration process as an example, it is necessary to measure the position of each joint point using a total station and then calculate the angles of each joint. Since the boom and stick of an excavator are not on a single plane, the dimensional measurement has considerable errors, and the accuracy of the calculation results is even less guaranteed. Alternatively, using a measuring tape to measure the joint length is both dangerous and inefficient for large excavators.

[0003] Current calibration schemes suffer from several drawbacks. First, they can automatically annotate a limited number of parameters, resulting in insufficient automation. Second, their calibration methods are complex and inconvenient to operate. Furthermore, these methods require additional equipment, increasing costs. Summary of the Invention

[0004] This system presents a calibration method for excavators that utilizes mechanical geometry and inverse kinematics to obtain some dimensional parameters and all sensor installation errors. This method requires no additional parameters or sensors compared to traditional calibration and measurement methods, thus improving the convenience of measurement and calibration efficiency for on-site installation personnel.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a parameter calibration system and method for an excavator working mechanism.

[0006] A parameter calibration system for an excavator's working mechanism, the excavator comprising a single-arm boom connected in sequence, a linkage group connected to the single-arm boom, and a bucket connected to both the single-arm boom and the linkage group, wherein a GNSS high-precision positioning device is installed at the end point M of the bucket, and the single-arm boom and the linkage group are respectively equipped with posture sensors.

[0007] Preferably, the linkage assembly includes a primary moving link connected to the single lever arm, a bucket push rod connected to the primary moving link, and a bucket link connected to the bucket push rod, wherein a position sensor is provided on the primary moving link.

[0008] Preferably, the GNSS high-precision positioning device is an RTK positioning antenna.

[0009] Preferably, the pose sensor is any one of an IMU, a triaxial accelerometer, or an angle sensor.

[0010] Preferably, the excavator is either a three-arm excavator or a four-arm excavator.

[0011] A parameter calibration method for an excavator working mechanism, using the aforementioned parameter calibration system for an excavator working mechanism, includes the following steps:

[0012] Step a: Rotate the single-lever joint or bucket joint to obtain the trajectory trajn of point M in space via GNSS. Use approximation or equation fitting methods to transform trajn into a circular equation Cn, and record the attitude sensor readings throughout the process. Based on the circle equation of Cn, the spatial coordinates of the center point of the single-arm joint or bucket joint at the initial moment are calculated, thereby obtaining the spatial coordinates of the initial moment of rotation of the single-arm joint or bucket joint.

[0013] Step b: Take the side plane of the vehicle body and the coordinates during the rotation of the bucket joint. Let the coordinate system of this plane be XOY. Let the joint point connecting the bucket and the single arm be point D, the joint point connecting the bucket link and the bucket be point F, the joint point connecting the initial moving link and the single arm be point C, and the joint point connecting the initial moving link and the bucket push rod be point E. Measure the distance between points C and E as L4, the distance between points D and F as L5, the distance between points E and F as L6, and the distance between points F and M as L8.

[0014] Based on step a, the initial spatial coordinates of point D and point M are obtained. Based on the initial spatial coordinates of point D and point M, the distance between points D and M is calculated to be L7. The lengths of each segment in the bucket triangle L5, L7, and L8 are known, and the coordinates of points D and M are determined. Therefore, the coordinates of point F are determined.

[0015] Let the coordinates of point C be (X... C Y C Take the coordinates (X, Y) of point E at time 1. E1 Y E1 ), the coordinates of point F at time 1 (X F1 Y F1 At this moment, the attitude sensor reading on L4 is θ. 41 The installation error of the attitude sensor is e4.

[0016]

[0017] The distance between points E and F is L6, and the equation can be written as follows:

[0018] (x E1 -x F1 ) 2 +(y E1 -y F1 ) 2 =L6 2 (5)

[0019] Substituting equation 4 into equation 5, we get

[0020] (x c +L4 cos(θ 41 -e4)-x F1 ) 2 +(y c +L4 sin(θ 41 -e4)-y F1 ) 2 =L6 2 (6)

[0021] Equation 6 has three unknowns x c y c e4 is a quadratic equation in three variables. According to step a, we can arbitrarily select three points on the equation of the circle with center D, and list the system of equations:

[0022]

[0023] The equation was solved by computer to obtain the coordinates (x, y) of point C. c ,y c The distance between different joints and the corresponding installation error are calculated based on the coordinates of the joints, as well as the installation error of the attitude sensor on L4.

[0024] Preferably, in step a, the method of approximation or equation fitting is to use the least squares method to fit the point set to obtain the equation of the circle.

[0025] Preferably, in step b, the equation is solved by computer using methods such as the Newton-Raphson method or gradient descent.

[0026] Preferably, in step a, the equation of the circle is obtained by fitting the point set, and the values ​​of the parameters are solved using a numerical optimization algorithm.

[0027] Preferably, in step a, the numerical optimization algorithm used is either gradient descent or Newton's method.

[0028] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: In excavator end-point positioning technology, the lengths of each joint of the excavator need to be obtained in advance. For sensors installed on the excavator's working mechanism, their installation errors also need to be calibrated. Therefore, the overall process is cumbersome, affecting installation efficiency. Furthermore, traditional calibration processes require more additional measuring instruments or sensors, increasing costs and raising the barrier to entry. This solution only requires the operator to measure the dimensions of some joints and then perform a series of actions on the excavator to automatically obtain the calibration parameters of each sensor and the joint lengths. It also eliminates the need for additional instruments, lowering the barrier to entry and improving efficiency. Attached Figure Description

[0029] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0030] Figure 1 This is a schematic diagram of the specific structure of the excavator according to an embodiment of the present invention;

[0031] Figure 2 This is a flowchart illustrating the parameter calibration system and method of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] The excavator needs to be equipped with GNSS to obtain its position and attitude information in the world coordinate system. Tilt sensors need to be installed on the boom, stick, and bucket joints of the excavator, respectively, at L1, L3, and L4.

[0034] The ultimate goal of excavator bucket end-point positioning technology is to determine the position of the bucket teeth. This determined position can then be used as an auxiliary indicator for both unmanned and manual operation. End-point positioning is obtained through geometric derivation, calculating the excavator's position and attitude in the world coordinate system along with the state of each joint. Therefore, it requires the length information of the eight joint arms (L1 to L8) shown in the diagram, as well as the installation error information of the sensors mounted on each joint arm.

[0035] This solution obtains the length information of the articulated arm and the sensor installation error information through a series of actions. It also effectively reduces the negative impact of measurement errors on the system.

[0036] For the excavator's operating system, the various joints and booms can be divided into two categories:

[0037] 1. The actuator's movement drives the movement of a single lever and single joint (e.g.) Figure 1 Actuator 1 pushes lever L1;

[0038] 2. The actuator's movement drives the movement of multi-link joints (e.g.) Figure 1 Actuator 3 drives levers L4, L5, L6, L7, and L8.

[0039] For a type 1 articulated lever arm: a GNSS high-precision positioning device is installed at the end point M. When the corresponding actuator moves, the lever arm performs a circular motion around the joint. The aforementioned GNSS device acquires the discrete spatial coordinate set {(x1,y1,z1),(x2,y2,z2)...(x...} of the end point M during this circular motion. n ,y n ,z n To obtain the equation C of the spatial circle by fitting a circle to the point set (e.g., using the least squares method), the coordinates of the joint point, i.e., the center (x, y, z) of equation C, can be obtained. The length of the link arm is the distance between each joint point.

[0040]

[0041] For type 2 articulated links: a GNSS high-precision positioning device is installed at the end point M, and an IMU capable of measuring pitch and roll angles is installed on link L4 (initial moving link). The length information of some links needs to be measured in advance. Figure 1 L4, L5 (bucket linkage), L6 (bucket push rod), and L8 (among others). When the corresponding actuator moves, the joint linkages constrain each other, ultimately pushing the end effector to rotate around point D. The aforementioned GNSS equipment acquires the discrete spatial coordinate set {(x1,y1,z1),(x2,y2,z2)...(x...} of the end effector M during this circular motion. n ,y n ,z n By fitting a spatial circle to this set of points, we obtain the spatial circle equation C, from which the coordinates of the joint point, i.e., the center (x, y, z) of equation C, can be obtained. The length of the lever arm L7 at this point is the radius r of equation C. Furthermore, the installation error of the IMU can be obtained by solving the system of equations governing the motion process.

[0042] By observing the movement of the two types of joints described above, the positions of each joint point can be determined. The theoretical angles corresponding to these joints can then be calculated, and the difference between these theoretical angles and the actual sensor angles at that moment can be used to determine the installation error of each sensor.

[0043] Specifically, for example Figure 1Taking a three-joint excavator as an example. In this solution, the lengths of the four joint arms need to be measured manually, for illustration purposes. Figure 1 The lengths of joints L4, L5, L6, and L8 are specified. The automatic acquisition method for other joint lengths is as follows:

[0044] 1. Install the RTK positioning antenna at point M on the bucket;

[0045] 2. Actuator 1 pushes a type 1 articulated lever arm, rotating only joint A. The trajectory traj1 of point M in space is obtained via GNSS. Traj1 can be transformed into a circular equation C1 using approximation or equation fitting methods, and the attitude sensor readings throughout the process are recorded.

[0046] The same applies to points B and D.

[0047] 3. Actuator 2 pushes a type 1 articulated lever arm, rotating only joint B to obtain the circular equation C2 of point M in space, and records the attitude sensor readings throughout the process.

[0048] 4. Actuator 3 pushes a type 2 articulated lever arm, rotating only joint D to obtain the circular equation C3 of point M in space, and records the attitude sensor readings throughout the process.

[0049] (The order of steps 2, 3, and 4 in the above process is not important)

[0050] Based on the circle equations C1, C2, and C3, the initial spatial coordinates of points A, B, and D at the center of the circle can be calculated. The coordinates of point C can also be calculated. Furthermore, the installation errors of the three attitude sensors relative to each joint arm can be deduced.

[0051] This solution can efficiently and quickly obtain the length parameters of each boom of an excavator and the installation errors of the sensors. Compared with traditional methods that use total stations or other equipment for measurement and calibration, this solution requires no additional equipment or sensors, and eliminates cumbersome manual operation and measurement procedures, thereby improving calibration speed and reducing human error.

[0052] The ultimate goal of excavator bucket end-point positioning technology is to determine the position of the bucket teeth. End-point positioning is obtained through geometric derivation, calculating the excavator's position and attitude in the world coordinate system along with the states of each joint. Therefore, a schematic diagram is needed. Figure 1 The diagram shows the length information of the eight articulated arms L1 to L8, as well as the installation error information of the sensors mounted on each articulated arm.

[0053] This solution is an automatic calibration scheme for the physical parameters of each articulated arm and the sensor installation parameters of an excavator for end-point positioning of the working mechanism.

[0054] The following example uses an excavator operating mechanism with a three-joint positioning system for the boom, stick, and bucket. The automatic parameter calibration of the positioning system for single, double, three, and four joints of three-joint and four-joint excavators remains effective.

[0055] For three-joint excavators, the parameters that can be automatically calibrated include boom length (L1, L2, L3, L7) and installation errors of the three IMUs (e1, e2, e3).

[0056] The excavator needs to be equipped with GNSS to obtain its position and attitude information in the world coordinate system. Attitude sensors are installed on the relevant joints of the excavator's boom (L1), stick (L3), and bucket (L4).

[0057] For the excavator's operating system, the various joints and booms can be divided into two categories:

[0058] 1. The actuator's movement drives the movement of a single lever and single joint (e.g.) Figure 1 Actuator 1 pushes lever L1;

[0059] 2. The actuator's movement drives the movement of multi-link joints (e.g.) Figure 1 Actuator 3 drives levers L4, L5, L6, L7, and L8.

[0060] For type 1 articulated lever arms: Install a GNSS high-precision positioning device at the end point M. When the corresponding actuator moves, the lever arm performs a circular motion around the joint. At this time, the joint point is the center coordinate of the circle. That is, the spatial coordinates of the joint point are obtained.

[0061] For type 2 articulated levers: a GNSS high-precision positioning device is installed at the end point M, and an IMU capable of measuring pitch and roll angles is installed on lever L4. The linkage is continuously pushed, causing the end point to perform a circular motion around a key joint. The spatial coordinates of this key joint and the length of the lever performing the circular motion are known. Simultaneously, by using the lengths of other articulated levers, equations can be derived to determine the position of the other joint and the sensor installation error.

[0062] By observing the movement of the two types of joints described above, the positions of each joint point can be determined. The theoretical angles corresponding to these joints can then be calculated, and the difference between these theoretical angles and the actual sensor angles at that moment can be used to determine the installation error of each sensor.

[0063] This solution, through a series of actions, can obtain partial articulated arm length information and sensor installation error information. Furthermore, it can effectively reduce the negative impact of measurement errors on the system.

[0064] In this design, the lengths of four articulated arms need to be measured, namely L4, L5, L6, and L8 in the diagram. The calculation methods for the lengths of other joints are as follows:

[0065] 1. Install the RTK positioning antenna at point M on the bucket;

[0066] 2. Rotate only joint A, and obtain the trajectory traj1 of point M in space via GNSS. Traj1 can be transformed into a circular equation C1 using approximation or equation fitting methods, and the attitude sensor readings throughout the process are recorded.

[0067] The same applies to points B and D:

[0068] 3. Rotate only joint B to obtain the equation C2 of the circle at point M in space, and record the attitude sensor readings throughout the process.

[0069] 4. Rotate only joint D to obtain the equation C3 of the circle at point M in space, and record the attitude sensor readings throughout the process.

[0070] (The order of steps 2, 3, and 4 in the above process is not important)

[0071] Based on the circle equations C1, C2, and C3, the initial spatial coordinates of the circle center coordinate joints A, B, and D can be calculated.

[0072] For discrete points acquired by GNSS, the least squares method can be used to fit the point set to obtain the equation of the circle. The following example uses a planar circle: Let the point set be (x... i ,y i The general equation of a plane circle can be expressed as:

[0073] (xa) 2 +(yb) 2 =r 2 (0.1)

[0074] Where (a, b) are the coordinates of the center of the circle, and r is the radius of the circle. That is, we need to find the parameters a, b, and r that minimize the following sum of squared errors:

[0075]

[0076] The parameter values ​​can then be solved using numerical optimization algorithms, such as gradient descent or Newton's method.

[0077] From this, the length values ​​of L1, L3, and L7 can be obtained:

[0078]

[0079] From this, the joint angles θ1 and θ3 at the initial moment can be obtained.

[0080]

[0081] The installation error is:

[0082]

[0083] Where e1 and e3 are the installation errors of the sensors on L1 and L3, respectively, and θ 11 θ 31 This is the sensor reading at the initial moment.

[0084] Given that the lengths of each segment in the bucket triangle L5, L7, and L8 are known, and the coordinates of D and M are determined, the coordinates of point F are determined.

[0085] Take the following four steps: Points C and D are fixed. For now, consider only the side plane of the vehicle body, and let its coordinate system be XOY. The coordinates of points D and F are known. Let the coordinates of point C be (X... C Y C ), the coordinates of point E at time 1 (X E1 Y E1 At this moment, the tilt sensor reading on L4 is θ. 41 The sensor installation error is e4.

[0086]

[0087] The distance between points E and F is L6, and the equation can be written as follows:

[0088] (x E1 -x F1 ) 2 +(y E1 -y F1 ) 2 =L6 2 (5)

[0089] Substituting equation 4 into equation 5, we get

[0090] (x c +L4 cos(θ 41 -e4)-x F1 ) 2 +(y c +L4 sin(θ 41 -e4)-y F1 ) 2 =L6 2 (6)

[0091] Equation 6 has three unknowns x c y c e4 is a quadratic equation in three variables. According to step 4, three points on the equation of the circle centered at point D can be randomly selected to form a system of equations:

[0092]

[0093] The equations are solved by computer using methods such as the Newton-Raphson method or gradient descent. Therefore, the coordinates of point C (x...) c ,y c From this, we can obtain the installation error e4 of the tilt sensor on L4. At this point, the positions of each joint in space are clearly defined. Therefore, e2 can be calculated, and the length of L2 is the Cartesian distance between points B and C.

[0094]

[0095] Similarly, the length of L7 can be calculated.

[0096]

[0097] The above solutions achieve the following: automatic calibration of the parameters of each joint and boom of the excavator working mechanism; and automatic calibration of the installation error of the end point positioning equipment of the excavator working mechanism.

[0098] It is equally effective for excavators with three or four joints, as well as for different positioning schemes.

[0099] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0100] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for calibrating parameters of an excavator's working mechanism, characterized in that, A parameter calibration system for an excavator's working mechanism is used. The excavator includes a single boom connected in sequence, a linkage group connected to the single boom, and a bucket connected to both the single boom and the linkage group. A GNSS high-precision positioning device is installed at the end point M of the bucket, and a position and posture sensor is provided on the single boom and the linkage group respectively. The linkage assembly includes a primary moving link connected to the single lever arm, a bucket push rod connected to the primary moving link, and a bucket link connected to the bucket push rod. A position sensor is installed on the primary moving link. The calibration method includes the following steps: Step a: Rotate the single-lever joint or bucket joint to obtain the trajectory trajn of point M in space via GNSS. Use approximation or equation fitting methods to transform trajn into a circular equation Cn, and record the attitude sensor readings throughout the process. Based on the circle equation of Cn, the spatial coordinates of the center point of the single-arm joint or bucket joint at the initial moment are calculated, thereby obtaining the spatial coordinates of the initial moment of rotation of the single-arm joint or bucket joint. Step b: Take the side plane of the vehicle body and the coordinates during the rotation of the bucket joint. Let the plane coordinate system be XOY. Let the joint point connecting the bucket and the single arm be point D, the joint point connecting the bucket link and the bucket be point F, the joint point connecting the initial moving link and the single arm be point C, and the joint point connecting the initial moving link and the bucket push rod be point E. Measure the distance between points C and E as L4, the distance between points D and F as L5, the distance between points E and F as L6, and the distance between points F and M as L8. Based on step a, the initial spatial coordinates of point D and point M are obtained. The distance between points D and M is calculated to be L7. The lengths of each segment in the bucket triangle L5, L7, and L8 are known, and the coordinates of points D and M are determined. Therefore, the coordinates of point F are determined. Let the coordinates of point C be (X... C Y C ), take the coordinates (X) of point E at time 1. E1 Y E1 The coordinates of point F at time 1 (X) F1 Y F1 At this moment, the attitude sensor reading on L4 is... The installation error of the attitude sensor is ; ； The distance between points E and F is L6, and the equation can be written as follows: ； Substituting equation 4 into equation 5, we get ； Equation 6 has three unknowns. , , It is a quadratic equation in three variables. According to step a, we can arbitrarily select three points on the equation of the circle with center D, and list the system of equations: ； The system of equations was solved by computer, and the results were obtained. Point coordinates ( , and the installation error of the attitude sensor on L4 The distance between different joints and the corresponding installation error are obtained based on the coordinates of the joints.

2. The parameter calibration method for an excavator working mechanism according to claim 1, characterized in that, The GNSS high-precision positioning device is an RTK positioning antenna.

3. The parameter calibration method for an excavator working mechanism according to claim 1, characterized in that, The pose sensor can be any one of an IMU, a triaxial accelerometer, or an angle sensor.

4. The parameter calibration method for an excavator working mechanism according to claim 1, characterized in that, The excavator mentioned is either a three-arm excavator or a four-arm excavator.

5. The parameter calibration method for an excavator working mechanism according to claim 1, characterized in that, In step a, the approximation or equation fitting method is to use the least squares method to fit the point set to obtain the equation of the circle.

6. The parameter calibration method for an excavator working mechanism according to claim 1, characterized in that, In step b, the equation is solved by computer using the Newton-Raphson method or gradient descent method.

7. The parameter calibration method for an excavator working mechanism according to claim 5, characterized in that, In step a, the equation of the circle is obtained by fitting the point set, and the values ​​of the parameters are solved using a numerical optimization algorithm.

8. The parameter calibration method for an excavator working mechanism according to claim 5, characterized in that, In step a, the numerical optimization algorithm used is either gradient descent or Newton's method.

Images