CN108058188B - Control method of robot health monitoring and fault diagnosis system - Google Patents

Control method of robot health monitoring and fault diagnosis system Download PDF

Info

Publication number
CN108058188B
CN108058188B CN201711188384.4A CN201711188384A CN108058188B CN 108058188 B CN108058188 B CN 108058188B CN 201711188384 A CN201711188384 A CN 201711188384A CN 108058188 B CN108058188 B CN 108058188B
Authority
CN
China
Prior art keywords
robot
joint
torque
state
current
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711188384.4A
Other languages
Chinese (zh)
Other versions
CN108058188A (en
Inventor
李方硕
曹俊
何理
张泽庞
徐健
李朝阳
贾云龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Linkhou Robot Co ltd
Original Assignee
Suzhou Linkhou Robot Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Suzhou Linkhou Robot Co ltd filed Critical Suzhou Linkhou Robot Co ltd
Priority to CN201711188384.4A priority Critical patent/CN108058188B/en
Publication of CN108058188A publication Critical patent/CN108058188A/en
Application granted granted Critical
Publication of CN108058188B publication Critical patent/CN108058188B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
    • B25J19/0095—Means or methods for testing manipulators

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Manipulator (AREA)

Abstract

The invention belongs to the technical field of robots, and relates to a control method of a robot health monitoring and fault diagnosis system. According to the invention, the running state and the health state of the robot are judged in real time by establishing the robot joint torque prediction model and the robot mechanical parameter identification model, and various abnormal states are processed in time, so that mechanical faults possibly occurring in the robot can be effectively found in an early stage, and the economic loss is reduced.

Description

Control method of robot health monitoring and fault diagnosis system
Technical Field
The invention relates to the technical field of robots, in particular to a control method of a robot health monitoring and fault diagnosis system.
Background
Robots have been widely used in various fields of automated production, and currently, there is no effective means for monitoring the health status of robots. Sudden damage to the robot may cause the entire production line to be stalled, resulting in significant economic losses. Therefore, the health monitoring and fault diagnosis of the robot are of great significance to avoid loss.
Therefore, it is necessary to provide a new control method to solve the above problems.
Disclosure of Invention
The invention mainly aims to provide a control method of a robot health monitoring and fault diagnosis system.
The invention realizes the purpose through the following technical scheme: a control method of a robot health monitoring and fault diagnosis system,
the method comprises the following steps:
s1, obtaining a mechanical parameter vector P of the robot when leaving the factory, wherein the dimension is mx 1, and the mechanical parameters comprise the load, the mass, the centroid, the inertia and the friction coefficient of each joint;
s2, acquiring the motion quantity of each joint of the robot: joint angle q, joint velocity
Figure GDA0002766679080000011
Joint acceleration, all three are Nx 1 vectors;
s3, acquiring actual torque T of each joint of the robotr；
S4, establishing a prediction model T of robot joint torquepB is a known parameter matrix;
s5, substituting the mechanical parameters P of S1 and the joint motion quantity obtained by S2 into the model established by S4, and solving the predicted value T of the joint torquep；
S6, comparing the actual joint torque T obtained in S3rAnd joint predicted torque T obtained at S5pJudging the motion state of each joint of the robot at the current moment;
s7, establishing an identification model P of mechanical parameters of the robotk＝(BTB)-1BTTr；
S8, the motion data obtained in S2 and the joint actual torque T obtained in S3rSolving the robot mechanical parameter P at the current moment by substituting the model established in S7k；
S9, comparing the current time robot mechanical parameter P obtained in S8kAnd the factory parameters P of the robot0Judging the health state of the robot body at the current moment;
s10, recording the motion state of each joint of the robot obtained in S6 and the health state of the robot body obtained in S9;
s11, judging the working state of the robot according to the judgment result obtained in the S6 and the judgment result obtained in the S9;
s12, according to the result obtained in S11, if the state of the robot is abnormal, performing abnormal processing;
s13, according to the result obtained in S11, if the robot state is normal, according to the formula Pk+1＝(1-λ)Pk+ λ P updating the mechanical parameter to obtain a new value P S1k+1λ is the update rate, and the value interval is [0.01,0.1]]；
The system comprises:
the data acquisition part is positioned at the motion joint and used for acquiring the state parameters of the current robot in real time, wherein the state parameters comprise mechanical parameters, motion quantity of each joint and torque measured value of each joint;
the running state monitoring part is used for establishing a robot joint torque prediction model, substituting the joint motion amount and the mechanical parameters obtained by the data acquisition part to calculate a current joint torque predicted value, comparing the joint torque predicted value with an actual value to calculate a torque deviation, and judging the current running state of the robot;
the health state monitoring part is used for establishing a robot mechanical parameter identification model, substituting the joint motion amount and the joint torque obtained by the data acquisition part to calculate the current robot mechanical parameter, and comparing the current robot mechanical parameter with the factory mechanical parameter to judge the current health state of the robot;
the state processing part is used for performing corresponding processing according to the current running state and the health state of the robot, and updating the mechanical parameters of the robot if the state is normal; if the state is abnormal, performing exception handling in a mode of deceleration movement and movement stop;
a state recording part for recording the state parameters of the robot;
wherein S1-S3 are completed by the data acquisition part, S4-S6 are completed by the running state detection part, S7-S9 are completed by the health state detection part, S10 is completed by the state recording part, and S11-S13 are completed by the state processing part.
Specifically, the actual torque T of each joint of the robotrCalculated by the following formula
Figure GDA0002766679080000021
Wherein tau ismiFor the i-th shaft motor to output torque, GiIs the reduction ratio of the ith joint,
Figure GDA0002766679080000022
is the angular velocity of the motor of the ith axis,
Figure GDA0002766679080000023
angular acceleration of the i-th axis, JmiIs the moment of inertia of the rotor of the i-th axis motor, cmiIs the i-th axis motor viscous damping coefficient, fmiIs the friction torque of the i-th shaft motor rotor. The output torque of the motor is equal to the current of the motor multiplied by a torque constant, and the actual torque T of the robotrIs a column vector of Nx 1 dimension, N represents the degree of freedom of the robot, and the ith element is the ith axis joint torque tauiI.e. has Tr＝[τ1，τ2，…，τN]T。
Specifically, the step S6 includes: and subtracting the predicted joint torque from the actual joint torque to obtain a torque deviation, and judging the current running state of the robot according to the torque deviation.
Further, the torque deviation signal is high-pass filtered, if the absolute value of the filtered signal is larger than a second threshold value T2Determining that the robot body is subjected to strong external high-frequency disturbance; if the absolute value of the filtered signal is greater than a first threshold value T1And is less than a second threshold value T2If the robot body receives the common external high-frequency disturbance, the robot body is determined to receive the common external high-frequency disturbance; second threshold value T2Greater than a first threshold value T1。
Further, the torque deviation signal is low-pass filtered, if the absolute value of the filtered signal is larger than a second threshold value T2Determining that the robot body is subjected to strong external low-frequency disturbance; if the absolute value of the filtered signal is greater than a first threshold value T1And is less than a second threshold value T2If the robot body receives the common external low-frequency disturbance, the robot body is determined to receive the common external low-frequency disturbance; second threshold value T2Greater than a first threshold value T1。
Specifically, the abnormality processing manner in step S12 includes deceleration movement and movement stop.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the running state and the health state of the robot are judged in real time by establishing the robot joint torque prediction model and the robot mechanical parameter identification model, and various abnormal states are processed in time, so that mechanical faults possibly occurring in the robot can be effectively found in an early stage, and the economic loss is reduced.
Drawings
FIG. 1 is a schematic diagram of a structure and a flowchart corresponding to the system for health monitoring and fault diagnosis of a robot according to an embodiment;
FIG. 2 is a logic block diagram of an embodiment of a robot health monitoring and fault diagnosis system during operation.
Fig. 3 is a comparison graph of the actual value and the predicted value of the torque of the first axis joint of the Scara robot in the normal operation state.
Detailed Description
The present invention will be described in further detail with reference to specific examples.
Example (b):
as shown in fig. 1, the present invention provides a method for controlling a robot health monitoring and fault diagnosis system, which comprises: the data acquisition part is positioned at the motion joint and used for acquiring the state parameters of the current robot in real time, such as mechanical parameters, the motion amount of each joint and the torque measured value of each joint;
the running state monitoring part is used for establishing a robot joint torque prediction model, substituting the joint motion amount and the mechanical parameters obtained by the data acquisition part to calculate a current joint torque predicted value, comparing the joint torque predicted value with an actual value to calculate a torque deviation, and judging the current running state of the robot;
the health state monitoring part is used for establishing a robot mechanical parameter identification model, substituting the joint motion amount and the joint torque obtained by the data acquisition part to calculate the current robot mechanical parameter, and comparing the current robot mechanical parameter with the factory mechanical parameter to judge the current health state of the robot;
the state processing part is used for performing corresponding processing according to the current running state and the health state of the robot, and updating the mechanical parameters of the robot if the state is normal; if the state is abnormal, performing exception handling in a mode of deceleration movement and movement stop;
and a state recording part for recording the robot state parameters.
As shown in fig. 2, the control method using the robot health monitoring and fault diagnosis system includes the following steps:
s1, obtaining a mechanical parameter vector P of the robot when leaving the factory, wherein the dimension is mx 1, and the mechanical parameters comprise the load, the mass, the centroid, the inertia and the friction coefficient of each joint;
s2, acquiring the motion quantity of each joint of the robot: joint angle q, joint velocity
Figure GDA0002766679080000041
Acceleration of joint
Figure GDA0002766679080000042
All three are Nx 1 vectors;
s3, acquiring actual torque T of each joint of the robotr；
S4, establishing a prediction model T of robot joint torquepB is a known parameter matrix;
s5, substituting the mechanical parameters P of S1 and the joint motion quantity obtained by S2 into the model established by S4, and solving the predicted value T of the joint torquep；
S6, comparing the actual joint torque T obtained in S3rAnd joint predicted torque T obtained at S5pJudging the motion state of each joint of the robot at the current moment;
s7, establishing an identification model P of mechanical parameters of the robotk＝(BTB)-1BTTr；
S8, the motion data obtained in S2 and the joint actual torque T obtained in S3rSolving the robot mechanical parameter P at the current moment by substituting the model established in S7k；
S9, comparisonCurrent time robot mechanical parameter P obtained in S8kAnd the factory parameters P of the robot0Judging the health state of the robot body at the current moment;
s10, recording the motion state of each joint of the robot obtained in S6 and the health state of the robot body obtained in S9;
s11, judging the working state of the robot according to the judgment result obtained in the S6 and the judgment result obtained in the S9;
s12, according to the result obtained in S11, if the state of the robot is abnormal, performing abnormal processing;
s13, according to the result obtained in S11, if the robot state is normal, according to the formula Pk+1＝(1-λ)Pk+ λ P updating the mechanical parameter to obtain a new value P S1k+1λ is the update rate, and the value interval is [0.01,0.1]]；
Wherein S1-S3 are completed by the data acquisition part, S4-S6 are completed by the running state detection part, S7-S9 are completed by the health state detection part, S10 is completed by the state recording part, and S11-S13 are completed by the state processing part.
For S1, the mechanical parameters of the robot body comprise load, mass center, inertia, friction coefficient and the like of each joint, and the initial values of the mechanical parameters indicated by S1 are determined before factory shipment.
The motion amount of the joint designated at S2 includes a joint angle q and a joint speed
Figure GDA00027666790800000516
Acceleration of joint
Figure GDA00027666790800000510
All three are Nx1 vectors, q ═ q1，q2，...，qN]T，
Figure GDA00027666790800000511
qi、
Figure GDA00027666790800000512
And
Figure GDA00027666790800000513
the angle, angular velocity and angular acceleration of the ith joint, respectively.
The actual torque of the joint designated at S3 is the output torque of the drive mechanism, and is calculated by the following equation
Figure GDA0002766679080000051
The index i represents the ith axis of the robot, where λiIs the motor torque constant, IiFor motor current, τmiFor the output of torque of the motor, GiIn order to reduce the joint speed by a predetermined ratio,
Figure GDA00027666790800000514
in order to determine the angular velocity of the motor,
Figure GDA00027666790800000515
for angular acceleration of the motor, JmiIs the moment of inertia of the rotor of the motor, cmiIs the viscous damping coefficient of the motor, fmiIs the friction torque of the motor rotor. The output torque of the motor is equal to the current of the motor multiplied by a torque constant, and the actual torque T of the robotrIs a column vector of Nx 1 dimension, N represents the degree of freedom of the robot, and the ith element is the ith axis joint torque tauiI.e. has Tr＝[τ1,τ2,…,τN]T。
The actual torque prediction model of the robot indicated by S4 is obtained by the following method: a joint torque prediction model is a robot inverse dynamics model, a force and torque balance equation of a single joint is established by using a Newton Euler method, the angular velocity, the angular acceleration, the mass center velocity and the mass center acceleration of a rod piece are obtained through forward iteration of kinematics, and the force and the torque applied to the robot joint can be obtained through reverse iteration of dynamics.
The above-mentioned pushing process is described in detail below by taking the first two joints of the Scara robot as an example, wherein the forward kinematics iterative formula and the reverse dynamics iterative formula are as follows:
Figure GDA0002766679080000052
Figure GDA0002766679080000053
Figure GDA0002766679080000054
Figure GDA0002766679080000055
Figure GDA0002766679080000056
Figure GDA0002766679080000057
the parameters are defined as follows:
Figure GDA0002766679080000058
Figure GDA0002766679080000061
the forward derivation equation of the kinematics of the front two joints of Scara is
Figure GDA0002766679080000062
Figure GDA0002766679080000063
Figure GDA0002766679080000064
Figure GDA0002766679080000065
Wherein b is1＝[0，0，1]T，b2＝[0，0，1]T，r0，1＝[l1，0，0]，r0，c1＝[l1c，0，0]，r1，2＝[l2，0，0]，r1，c2＝[l2c，0，0]Geometric parameter l1，licAre all known quantities, the rotation matrix R1 0And R1 2Respectively as follows:
Figure GDA0002766679080000066
the inverse derivation formula of the Scara anterior two-joint dynamics is as follows:
f2＝m2ac，2-m2g2
τ2＝-f2×r1，c2+ω2×(J2ω2)+J2α2
Figure GDA0002766679080000067
Figure GDA0002766679080000068
wherein g is1＝[0，0，-9.8]，g2＝[0，0，-9.8]，m1Is the mass of the first rod member, m2Mass of the second rod member, J1The moment of inertia of the first rod piece relative to the first rotating shaft is a 3 x 3 matrix; j. the design is a square2The moment of inertia of the second rod with respect to the second rotation axis is also a 3 x 3 matrix. Because Scara one-axis and two-axisParallel, without affecting the results of the calculations, one can assume J1And J2Has the following forms:
Figure GDA0002766679080000069
derivation J according to parallel axis theorem of moment of inertia1zzAnd J2zzThe value of (c). The moment of inertia of the first rod relative to the center of mass is known as H1The second rod piece has a mass inertia moment H relative to the mass center2Then there is J1zz＝H1+m1*l1c*l1c，J2zz＝H2+m2*l2c*l2c. Push to the result of
Figure GDA00027666790800000610
Figure GDA00027666790800000611
Considering the friction of joints, let f1And f2The friction torques of the front two joints of the Scara robot are respectively obtained, and the joint torque equation is modified to
Figure GDA0002766679080000071
Figure GDA0002766679080000072
The friction torque comprises three parts of coulomb friction, linear viscous damping and square damping, and the friction torque expressions of the front two joints of the Scara robot are as follows
Figure GDA0002766679080000073
Figure GDA0002766679080000074
Wherein f isdiDamping the torque coefficient for dry friction, ci1Is a viscous damping torque coefficient, ci2The inverse dynamics model expression of the Scara robot is arranged for squaring the damping torque coefficient, and can be written into the following matrix form.
F＝B*P
Wherein the expressions of the matrix B, the vector P and the vector F are respectively
Figure GDA0002766679080000075
Figure GDA0002766679080000076
F＝[τ1，τ2]T
c2＝cos(q2) s2＝sin(q2)
For a single sample, the dimension of the B matrix is 2 × 9, the dimension of the P matrix is 9 × 1, and the dimension of the F matrix is 2 × 1. For Y sets of sample data, the dimension of the B matrix is 2Y × 9, the dimension of the P matrix is 9 × 1, and the dimension of the F matrix is 2Y × 1.
Let the predicted value of joint torque be TpReplacing F with TpAn expression of the joint torque prediction model can be obtained:
Tp＝B*P
the predicted joint torque value at S5 can be calculated by the prediction model at S4, where the robot mechanical parameter vector P is known and designated by S1, the model input is the joint motion variable q at S2,
Figure GDA0002766679080000077
Figure GDA0002766679080000078
bring the joint exerciseEntering a B matrix, and calculating a joint torque predicted value T by matrix point multiplicationp。
The motion state determination method of S6 is as follows: definition of TdiIs the i-th axis joint torque deviation
Tdi＝Tri-Tpi
Wherein T isriIs TrIs the actual value of the i-th axis joint torque calculated at S3, TpiIs TpThe ith element of (2) is the predicted value of the i-th axis joint torque calculated at S5.
Defining a first decision threshold T1Second determination threshold T2And has T2>T1>0。
Will TdiHigh-pass filtering is carried out to obtain a filtered signal Thd. If T is1<|Thd|<T2Then the robot is determined to be subjected to general external high-frequency disturbance if Thd|>T2The robot is deemed to be subjected to strong external high frequency disturbances, where | is in absolute sign. The disturbance may be caused by the following factors:
1) the transmission mechanisms such as the speed reducer, the motor and the like have faults.
2) The robot body is impacted.
3) Looseness and even part falling-off of connection among the load, the clamp and the robot flange occur.
4) Other factors.
Will TdiLow-pass filtering to obtain filtered signal Tsd. If T is1<|Tsd|<T2Then the robot is determined to be subjected to general external high-frequency disturbance if Tsd|>T2The robot is deemed to be subjected to strong external low frequency disturbances, where | is in absolute sign. The disturbance may be caused by the following factors:
1) the load mechanics parameters are incorrectly filled.
2) The transmission lacks lubrication.
3) The transmission mechanism ages.
4) The robot body is subjected to additional forces.
5) The outside environment is either too cold or too hot.
6) Other factors.
The mechanical parameter identification model indicated by S7 can be obtained by using least square method, and the solving formula is as follows
Pk＝(BTB)-1BTTr
The mechanical parameters at the current moment indicated by S8 can be calculated through a model shown by S7, and the model inputs the joint movement amount obtained by S2 and the actual torque of the joint obtained by S3, wherein the joint movement amount can initialize a B matrix, and the actual torque of the joint can initialize a vector TrAnd the model output is the mechanical parameter P of the robot at the current momentk。
The robot health state judgment method of S9 is as follows: i.e. P0For the mechanical parameter vector, P, of the robot leaving the factorykDefining a new variable delta P | | | P for the mechanical parameter vector at the current moment obtained by calculation of S8k-P0I, where the symbol Pk-P0I represents the vector Pk-P0The two norms of (a).
If Δ P>P1And considering the health state of the robot to be abnormal, otherwise, considering the health state of the robot to be normal. Wherein P is1>0 is a judgment threshold, and the abnormal state may be caused by the following factors:
1) problems arise with the motor.
2) The transmission wears or ages.
Here, the details are also described using Scara robot as an example, and the parameter vector P is written in two parts, Pk＝[P1，P2]Similarly, P is0Written in two parts, P0＝[P10，P20]In which P is10Is a subset of vectors, P, related to a quality parameter of the robot20Is a subset of vectors related to the robot friction torque, wherein
Figure GDA0002766679080000091
P1＝[c11，c12，c21，c22，fd1，fd2]，ΔP1＝||P1-P10||，ΔP2＝||P2-P20If Δ P |, if | | |1If the value of | | is abnormal, two possibilities exist, namely, the motor fails, and the robot body has additional mass. If | | | Δ P2If the value of | | is abnormal, the transmission mechanism has problems, such as the lubrication of the speed reducer is not ideal enough or the speed reducer is damaged.
And S10 recording the motion state of each joint of the robot obtained in S6 and the health state of the robot body obtained in S9.
And S11, making a decision according to judgment results of S6 and S9, and if the running state and the health state of the robot are normal, considering that the robot state is normal, or else, considering that the robot state is abnormal.
And S12, responding to the abnormal working state of the robot obtained in S11, switching the robot to a stop mode when the robot is subjected to strong external disturbance, and otherwise switching the robot to a low-speed running mode.
S13 responds to the normal working state of the robot obtained in S11, the specific action is to update the mechanical parameters of the robot, the mechanical parameters S1 of the robot at the current time are assumed to be P, and the mechanical parameters of the robot calculated in S8 at the current time are assumed to be PkThen the robot mechanical parameter S1 is P at the next momentk+1:
Pk+1＝(1-λ)Pk+λP
In the formula, λ is an update rate, a value interval is [0.01,0.1], and the larger the value of λ is, the faster the mechanical parameter of the robot indicated by S1 is updated.
In order to verify the correctness of the method, the motion amount data and the current data of the front two joints of the Scara robot are collected to calculate the actual torque of the first joint, the current mechanical parameters of the robot are calculated by utilizing an S7 mechanical parameter identification model, then the mechanical parameters and the resampled motion amount are brought into an S4 joint torque Prediction model to calculate the predicted value of the first joint torque, and the related result is shown in figure 3, wherein a solid line Prediction represents the predicted value of the joint torque, and a dotted line Real represents the actual value of the joint torque, so that the predicted value and the actual value of the joint torque are highly consistent under the normal working state, and the accuracy of the S4 joint torque Prediction model is verified.
What has been described above are merely some embodiments of the present invention. It will be apparent to those skilled in the art that various changes and modifications can be made without departing from the inventive concept thereof, and these changes and modifications can be made without departing from the spirit and scope of the invention.

Claims (6)

1. A control method of a robot health monitoring and fault diagnosis system is characterized by comprising the following steps:
s1, obtaining a mechanical parameter vector P of the robot when leaving the factory, wherein the dimension is mx 1, and the mechanical parameters comprise the load, the mass, the centroid, the inertia and the friction coefficient of each joint;
s2, acquiring the motion quantity of each joint of the robot: joint angle q, joint velocity
Figure FDA0002766679070000011
Acceleration of joint
Figure FDA0002766679070000012
All three are Nx 1 vectors;
s3, acquiring actual torque T of each joint of the robotr；
S4, establishing a prediction model T of robot joint torquepB is a known parameter matrix;
s5, substituting the mechanical parameters P of S1 and the joint motion quantity obtained by S2 into the model established by S4, and solving the predicted value T of the joint torquep；
S6, comparing the actual joint torque T obtained in S3rAnd joint predicted torque T obtained at S5pJudging the motion state of each joint of the robot at the current moment;
s7, establishing an identification model P of mechanical parameters of the robotk＝(BTB)-1BTTr；
S8, the motion data obtained in S2 and the joint actual torque T obtained in S3rSolving the robot mechanical parameter P at the current moment by substituting the model established in S7k；
S9, comparing the current time robot mechanical parameter P obtained in S8kAnd the factory parameters P of the robot0Judging the health state of the robot body at the current moment;
s10, recording the motion state of each joint of the robot obtained in S6 and the health state of the robot body obtained in S9;
s11, judging the working state of the robot according to the judgment result obtained in the S6 and the judgment result obtained in the S9;
s12, according to the result obtained in S11, if the state of the robot is abnormal, performing abnormal processing;
s13, according to the result obtained in S11, if the robot state is normal, according to the formula Pk+1＝(1-λ)Pk+ λ P updating the mechanical parameter to obtain a new value P S1k+1λ is the update rate, and the value interval is [0.01,0.1]]；
The system comprises:
the data acquisition part is positioned at the motion joint and used for acquiring the state parameters of the current robot in real time, wherein the state parameters comprise mechanical parameters, motion quantity of each joint and torque measured value of each joint;
the running state monitoring part is used for establishing a robot joint torque prediction model, substituting the joint motion amount and the mechanical parameters obtained by the data acquisition part to calculate a current joint torque predicted value, comparing the joint torque predicted value with an actual value to calculate a torque deviation, and judging the current running state of the robot;
the health state monitoring part is used for establishing a robot mechanical parameter identification model, substituting the joint motion amount and the joint torque obtained by the data acquisition part to calculate the current robot mechanical parameter, and comparing the current robot mechanical parameter with the factory mechanical parameter to judge the current health state of the robot;
the state processing part is used for performing corresponding processing according to the current running state and the health state of the robot, and updating the mechanical parameters of the robot if the state is normal; if the state is abnormal, performing exception handling in a mode of deceleration movement and movement stop;
a state recording part for recording the state parameters of the robot;
wherein S1-S3 are completed by the data acquisition part, S4-S6 are completed by the running state detection part, S7-S9 are completed by the health state detection part, S10 is completed by the state recording part, and S11-S13 are completed by the state processing part.
2. The control method according to claim 1, characterized in that: actual torque T of each joint of the robotrCalculated by the following formula
Figure FDA0002766679070000021
Wherein tau ismiFor the i-th shaft motor to output torque, GiIs the reduction ratio of the ith joint,
Figure FDA0002766679070000022
is the angular velocity of the motor of the ith axis,
Figure FDA0002766679070000023
angular acceleration of the i-th axis, JmiIs the moment of inertia of the rotor of the i-th axis motor, cmiIs the i-th axis motor viscous damping coefficient, fmiFriction torque of the ith shaft motor rotor; the output torque of the motor is equal to the current of the motor multiplied by a torque constant, and the actual torque T of the robotrIs a column vector of Nx 1 dimension, N represents the degree of freedom of the robot, and the ith element is the ith axis joint torque tauiI.e. has Tr＝[τ1，τ2，…，τN]T。
3. The control method according to claim 1, characterized in that: the step S6 includes: and subtracting the predicted joint torque from the actual joint torque to obtain a torque deviation, and judging the current running state of the robot according to the torque deviation.
4. The control method according to claim 3, characterized in that: high-pass filtering the torque deviation signal, if the absolute value of the filtered signalGreater than a second threshold value T2Determining that the robot body is subjected to strong external high-frequency disturbance; if the absolute value of the filtered signal is greater than a first threshold value T1And is less than a second threshold value T2If the robot body receives the common external high-frequency disturbance, the robot body is determined to receive the common external high-frequency disturbance; second threshold value T2Greater than a first threshold value T1。
5. The control method according to claim 3, characterized in that: low-pass filtering the torque deviation signal, if the absolute value of the filtered signal is greater than a second threshold value T2Determining that the robot body is subjected to strong external low-frequency disturbance; if the absolute value of the filtered signal is greater than a first threshold value T1And is less than a second threshold value T2If the robot body receives the common external low-frequency disturbance, the robot body is determined to receive the common external low-frequency disturbance; second threshold value T2Greater than a first threshold value T1。
6. The control method according to claim 1, characterized in that: the abnormality processing manner in step S12 includes deceleration movement and movement stop.
CN201711188384.4A 2017-11-24 2017-11-24 Control method of robot health monitoring and fault diagnosis system Active CN108058188B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711188384.4A CN108058188B (en) 2017-11-24 2017-11-24 Control method of robot health monitoring and fault diagnosis system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711188384.4A CN108058188B (en) 2017-11-24 2017-11-24 Control method of robot health monitoring and fault diagnosis system

Publications (2)

Publication Number Publication Date
CN108058188A CN108058188A (en) 2018-05-22
CN108058188B true CN108058188B (en) 2021-04-30

Family

ID=62136040

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711188384.4A Active CN108058188B (en) 2017-11-24 2017-11-24 Control method of robot health monitoring and fault diagnosis system

Country Status (1)

Country Link
CN (1) CN108058188B (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108908345B (en) * 2018-08-31 2023-07-14 上海大学 Under-actuated dexterous hand transmission system state sensing system
CN109465823B (en) * 2018-11-06 2022-03-18 泰康保险集团股份有限公司 Intelligent robot control method and device, electronic equipment and storage medium
CN113165172B (en) * 2018-12-24 2024-09-03 Abb瑞士股份有限公司 Method, device and server for diagnosing robot
CN109571549A (en) * 2018-12-29 2019-04-05 上海新时达机器人有限公司 The friction force monitoring methods and system and equipment of a kind of robot body
CN112123371A (en) * 2019-06-25 2020-12-25 株式会社日立制作所 Robot fault prediction device and system, and robot fault prediction method
CN110988526B (en) * 2019-11-21 2021-01-29 珠海格力电器股份有限公司 Robot assembly inspection method and device and storage medium
CN113021411B (en) * 2019-12-24 2023-06-20 株式会社日立制作所 Robot failure prediction device and system, and robot failure prediction method
CN111086025A (en) * 2019-12-25 2020-05-01 南京熊猫电子股份有限公司 Multi-fault-cause diagnosis system and method applied to industrial robot
CN111283731A (en) * 2020-03-17 2020-06-16 安徽智训机器人技术有限公司 Industrial robot operation fault determination method and system
CN111532988B (en) * 2020-04-26 2021-07-30 成都见田科技有限公司 Remote intelligent monitoring method and monitoring computer applied to elevator
CN111761576A (en) * 2020-06-15 2020-10-13 上海高仙自动化科技发展有限公司 Health monitoring method and system, intelligent robot and readable storage medium
WO2022041064A1 (en) * 2020-08-27 2022-03-03 Rethink Robotics Gmbh Method and apparatus for robot joint status monitoring
CN113776791A (en) * 2021-08-04 2021-12-10 深圳优地科技有限公司 Method and device for monitoring health state of robot, robot and storage medium
CN114323718B (en) * 2021-12-14 2023-12-15 合肥欣奕华智能机器股份有限公司 Robot fault prediction method and device
CN114770607B (en) * 2022-06-20 2022-09-02 深圳希研工业科技有限公司 Robot health monitoring method and system based on big data

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1284414A2 (en) * 2001-08-13 2003-02-19 Siemens Aktiengesellschaft Diagnosis of robot transmissions
CN101059697A (en) * 2006-04-17 2007-10-24 发那科株式会社 Device and method for controlling electric motor
CN101200066A (en) * 2006-12-11 2008-06-18 Abb研究有限公司 A method and a control system for monitoring the condition of an industrial robot
CN105095918A (en) * 2015-09-07 2015-11-25 上海交通大学 Multi-robot system fault diagnosis method
CN106020116A (en) * 2015-03-24 2016-10-12 发那科株式会社 Robot controller capable of performing fault diagnosis of robot
CN106814701A (en) * 2016-12-26 2017-06-09 武汉华中数控股份有限公司 Management and control digital control platform system and its construction method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1284414A2 (en) * 2001-08-13 2003-02-19 Siemens Aktiengesellschaft Diagnosis of robot transmissions
CN101059697A (en) * 2006-04-17 2007-10-24 发那科株式会社 Device and method for controlling electric motor
CN101200066A (en) * 2006-12-11 2008-06-18 Abb研究有限公司 A method and a control system for monitoring the condition of an industrial robot
CN106020116A (en) * 2015-03-24 2016-10-12 发那科株式会社 Robot controller capable of performing fault diagnosis of robot
CN105095918A (en) * 2015-09-07 2015-11-25 上海交通大学 Multi-robot system fault diagnosis method
CN106814701A (en) * 2016-12-26 2017-06-09 武汉华中数控股份有限公司 Management and control digital control platform system and its construction method

Also Published As

Publication number Publication date
CN108058188A (en) 2018-05-22

Similar Documents

Publication Publication Date Title
CN108058188B (en) Control method of robot health monitoring and fault diagnosis system
CN111788040B (en) Kinetic parameter identification method of robot, robot and storage device
CN106483964B (en) Robot compliance control method based on contact force observer
CN111496791B (en) Integral dynamic parameter identification method based on serial robots
CN112743541B (en) Soft floating control method for mechanical arm of powerless/torque sensor
CN109583093A (en) A kind of industrial robot dynamic parameters identification method considering joint elasticity
CN108466289A (en) A kind of dynamic modeling method for the parallel robot considering joint-friction
CN107391861A (en) Industrial robot loading kinetics parameter identification method independent of body kinetic parameter
CN113977578B (en) Soft measurement method for end force of hydraulic mechanical arm
CN113748597B (en) Motor control device
CN112677156B (en) Robot joint friction force compensation method
CN111267105A (en) Kinetic parameter identification and collision detection method for six-joint robot
CN103728988B (en) SCARA robot trajectory tracking control method based on internal model
CN111965976B (en) Robot joint sliding mode control method and system based on neural network observer
CN110103222A (en) A kind of industrial robot collision checking method
CN105045103A (en) Servo manipulator friction compensation control system based on LuGre friction model and method
CN104199291B (en) Dissipative structure theory based TORA (Translation oscillators with a rotating actuator) system self-adaption control method
CN115890735B (en) Mechanical arm system, mechanical arm, control method of mechanical arm system, controller and storage medium
CN114260892B (en) Elastic joint moment control method and device, readable storage medium and robot
CN115946131A (en) Flexible joint mechanical arm motion control simulation calculation method and device
CN108227493B (en) Robot trajectory tracking method
CN108445778B (en) Dynamics modeling method for space non-cooperative target non-complete constraint assembly
Han et al. External force estimation method for robotic manipulator based on double encoders of joints
CN113246137A (en) Robot collision detection method based on external moment estimation model
CN113650014A (en) Redundant mechanical arm tracking control method based on echo state network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant