CN109664295A - Robot belt sanding constant force control method and device based on one-dimensional force snesor - Google Patents
Robot belt sanding constant force control method and device based on one-dimensional force snesor Download PDFInfo
- Publication number
- CN109664295A CN109664295A CN201811523895.1A CN201811523895A CN109664295A CN 109664295 A CN109664295 A CN 109664295A CN 201811523895 A CN201811523895 A CN 201811523895A CN 109664295 A CN109664295 A CN 109664295A
- Authority
- CN
- China
- Prior art keywords
- force
- grinding
- robot
- sensor
- coordinate system
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000005498 polishing Methods 0.000 claims abstract description 46
- 230000003044 adaptive effect Effects 0.000 claims abstract description 24
- 238000013507 mapping Methods 0.000 claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 17
- 239000004576 sand Substances 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000007517 polishing process Methods 0.000 claims description 7
- 238000003754 machining Methods 0.000 claims description 5
- 238000013016 damping Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 2
- 230000003068 static effect Effects 0.000 claims description 2
- 230000003746 surface roughness Effects 0.000 description 7
- 229910000831 Steel Inorganic materials 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 6
- 238000006073 displacement reaction Methods 0.000 description 6
- 239000010959 steel Substances 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000007781 pre-processing Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007730 finishing process Methods 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/1633—Programme controls characterised by the control loop compliant, force, torque control, e.g. combined with position control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/005—Manipulators for mechanical processing tasks
- B25J11/0065—Polishing or grinding
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Finish Polishing, Edge Sharpening, And Grinding By Specific Grinding Devices (AREA)
Abstract
The invention discloses the robot belt sanding constant force control methods based on one-dimensional force snesor, comprising the following steps: according to force analysis, obtains the power mapping relations of each coordinate system；By polishing normal force and tangential force relationship, the power mapping relations of contact force and sensor coordinate system are established and simplified；By polishing deformation and grinding depth relationship, polishing kinetic model is established；Adaptive sliding mode iterative control algorithm is designed, and establishes power Controlling model；According to power mapping relations and power Controlling model, force controller is designed；By feedback force, application of force mapping relations calculate polishing normal force, and are input to force controller and calculate normal direction offset, then be transferred to control module；The present invention carries out the control of robot belt sanding power on one-dimensional force snesor, avoids the high cost problem of multi-dimension force sensor, while reducing control complexity；Using simple, parameter setting is convenient, effectively compensates for uncertainty bring error when belt sanding, is suitable for practical polishing.
Description
Technical Field
The invention relates to the research field of robot abrasive belt grinding force control, in particular to a robot abrasive belt grinding constant force control method and device based on a one-dimensional force sensor.
Background
As a finishing process, belt grinding can be used to improve the surface roughness of components as well as to achieve high material removal rates. When the abrasive belt is used for polishing and matched with an industrial robot with multiple degrees of freedom, a flexible manufacturing unit can be formed, the abrasive belt is particularly suitable for processing workpieces with complex surface geometric shapes, such as turbine blades or water taps, the personal health problem caused by the processing environment is also avoided, the processing efficiency is low, the labor cost is increased year by year, the stability is poor, the process consistency is not enough, and the like, which are frequently caused in the manual polishing and numerical control polishing processes.
As a result, a great deal of research has been conducted by many scholars on robotic belt sanding processes, some of which have been directed to the problem of sanding track planning. Although the track planning can improve the processing quality of the workpiece to a certain extent, the track planning of the robot by light cannot meet the processing requirement of the robot abrasive belt polishing, so that the force control needs to be carried out on the robot abrasive belt polishing, so that the high material removal rate is obtained, and the polishing quality is improved.
Robot belt sanding force control can be divided into passive force control and active force control. The passive force control mainly depends on auxiliary compliant mechanisms, so that the robot can naturally comply with the grinding force when contacting with the abrasive belt wheel. The control method can effectively improve the grinding quality, but reduces the dynamic range of force response and the precision of the tail end position.
To overcome these shortcomings of passive force control, active force control has been developed and is a major direction in the field of robot research today. Currently, research on the control of the active force of a robot can be basically classified into two types: force control based on a conventional strategy and force control based on an intelligent strategy, wherein the conventional control method is mainly divided into force/bit hybrid control and impedance control. Although these conventional force control strategies achieve some control effect, these control methods are difficult to achieve satisfactorily due to non-linearity and a large amount of uncertainty in the robot polishing. In order to overcome these problems, related researchers have proposed an intelligent control method that can calculate optimal control parameters in real time, but require a certain amount of training to achieve the control effect. Some researchers have proposed adaptive impedance control, which has better force tracking effect under the condition that the environmental parameters are unknown by using direct and indirect methods, but too many adaptive gain parameters are not beneficial to practical application in the adjusting process.
Disclosure of Invention
The invention mainly aims to overcome the defects in the prior art and provide a robot abrasive belt grinding constant force control method based on a one-dimensional force sensor. According to the stress condition of a workpiece during polishing, a contact force model between the tail end of a cutter and the workpiece during polishing is subjected to stress analysis to obtain a corresponding force mapping relation, and the force mapping relation is simplified by discussing and verifying the relation between a polishing normal force and a tangential force, so that the mapping relation between the polishing normal force and the force on the one-dimensional sensor is established; establishing a robot abrasive belt grinding dynamic model based on deformation by discussing the relation between deformation and grinding depth during grinding; and providing a self-adaptive sliding mode iterative algorithm, providing a corresponding force control model according to the algorithm, designing a controller according to the force control model and the force mapping relation, and obtaining a specific control flow, thereby achieving the purpose of controlling the polishing normal force of the tail end of the robot.
The invention also aims to provide a robot belt sanding constant-force control device based on the one-dimensional force sensor.
The main purpose of the invention is realized by the following technical scheme:
a robot abrasive belt grinding constant force control method based on a one-dimensional force sensor comprises the following steps:
s1, according to the stress condition of the polishing process, performing stress analysis on the contact force between a polishing workpiece at the tail end of the robot and the abrasive belt wheel to obtain a force mapping relation between a sensor coordinate system and an abrasive belt wheel coordinate system; establishing and simplifying a force mapping relation between the contact force and a sensor coordinate system by polishing the relation between the normal force and the tangential force;
s2, establishing a robot abrasive belt wheel grinding dynamic model based on deformation through the relation between grinding deformation and grinding depth;
s3, designing a self-adaptive sliding mode iterative control algorithm, and establishing a corresponding force control model according to the grinding dynamic model of the robot abrasive belt wheel;
s4, designing a corresponding adaptive sliding mode iteration constant force controller according to the force mapping relation between the contact force and the sensor coordinate system and a corresponding force control model;
and S5, calculating and feeding back the normal grinding force by the force received by the one-dimensional force sensor and applying the contact force and the force mapping relation on the sensor coordinate system, inputting the fed-back contact force into the corresponding adaptive sliding mode iteration constant force controller to obtain a normal offset, and transmitting the normal offset to the control module for control.
Further, in step S1, specifically, the method includes:
t1, establishing a pose relation between coordinate systems according to the model during grinding: the force sensor comprises a force sensor coordinate system and a belt wheel coordinate system, wherein the origin of the force sensor coordinate system is the geometric center of the sensor, the X-axis direction of the force sensor coordinate system is the axial direction of the sensor, the Y-axis direction of the force sensor coordinate system is the radial direction of the sensor, and the Z-axis direction of the force sensor coordinate system is determined according to the right-hand rule of the coordinate system; the origin of a sand belt wheel coordinate system is positioned on the surface of a sand belt of the center of the sand belt wheel along the radial direction, the X-axis direction of the sand belt wheel coordinate system is the radial direction of the sand belt wheel, the Y-axis direction of the sand belt wheel coordinate system is the axial direction of the sand belt wheel, the Z-axis direction of the sand belt wheel coordinate system is determined according to the right-hand rule of the coordinate systems, and the Z-axes of the two coordinate systems are parallel during grinding;
t2, analyzing the force in the grinding process, and setting FtAnd FnGrinding tangential force and normal force, F 'on a grinding wheel coordinate system'tAnd F'nIs shown astAnd FnForce transferred to the force sensor coordinate system, FxAnd FyRepresenting the forces in the X-axis and Y-axis, respectively, on a one-dimensional force sensor, then:
through the formula, solve the normal force and the tangential force of polishing on the abrasive band wheel coordinate system:
wherein theta is an included angle between Y axes of the two coordinate systems;
t3, establishing the relation between the grinding normal force and the grinding tangential force on the abrasive wheel coordinate system:
Fn＝ηFt，
η is the ratio of the normal force and the tangential force of the sanding on the abrasive belt wheel coordinate system;
t4, obtaining the force F of the X axis of the one-dimensional force sensor according to the step T2 and the step T3xGrinding normal force F on grinding belt wheel coordinate systemnThe relationship between:
Fx＝Fn(cosθ-sinθ/η)；
the final control objective is to make the force FnReach constant force, but FnCannot be obtained directly, and needs to pass through FxCalculating the force in the controller as FnTherefore, F obtained by a sensor is requiredxRequire conversion to F by this formulan；
Further, in step S2, specifically, the method includes:
u1, establishing a grinding force dynamic model:
wherein f isp(t) is a normal grinding force, m is a component of a system inertia matrix in the normal direction of the grinding force, c is a component of a system damping matrix in the normal direction of the grinding force, k is a component of the rigidity of the grinding process in the normal direction of the grinding force, and x (t) is the position of the cutter vertical to the surface of the workpiece, namely the grinding depth;first and second derivatives of x (t), respectively;
u2, analyzing the stress condition of the tail end of the robot in the polishing process, and establishing a stress model of the tail end of the robot:
f(t)＝fp(t)+fs(t)，
wherein f iss(t) is the deformation force of the robot during the polishing process;
u3, researching the relation between the grinding normal force and the grinding depth, and establishing the relation between the grinding deformation and the deformation force:
δx(t)＝x*(t)-x(t)，
fs(t)＝ksδx(t)，
wherein, deltax(t) grinding of the tool perpendicular to the surface of the workpieceDeflection, x (t) is the planned sanding depth, ksPolishing the static stiffness of the system for the robot;
u4, obtaining a robot belt sanding dynamic model based on deformation according to the relational expression of the steps U1, U2 and U3:
further, in step S3, specifically, the method includes:
v1, designing a sliding mode state according to the grinding force error:
ex(t)＝fx(t)-fxd(t)，
wherein λ is the sliding mode surface coefficient, Sx(t) slip form state at the time of grinding, ex(t)、Respectively the grinding force error and the first derivative of the grinding force error; f. ofx(t)、Respectively the moment actual polishing force and the first derivative of the moment actual polishing force; f. ofxd(t)、Respectively being the instant desired sanding force and the first derivative of the instant desired sanding force;
v2, the deformation-based robot belt sanding dynamics model established according to step S2, and the sliding mode state of step V1, then:
wherein m is the component of the system inertia matrix in the normal direction of the grinding force, c is the component of the system damping matrix in the normal direction of the grinding force, k is the component of the rigidity in the normal direction of the grinding force in the grinding process, SxiThe sliding mode state is the sliding mode state in the ith iteration;andrespectively forming a first derivative and a second derivative of a sliding mode surface by the processing depth;
and has the following components:
wherein S isxiFor actual grinding force of slip form surface, Sai(t) slip form surface formed by working depth, Sδi(t) is a slip form surface formed by working deformation, Sf(t) a slip form surface formed by ideal machining force; deltaxi(t) andthe component of the polishing deformation in the normal direction of the workpiece and the first derivative thereof are included;
v3, according to step V2, and the desired sliding mode state, i.e. SxiWhen 0, a force control model is obtained:
further, in step S4, specifically, the method includes:
w1, designing a control law of the corresponding adaptive sliding mode iterative constant force controller according to the simplified force mapping relation in the step S1 and the force control model in the step S3:
G(Δx(t))＝ksSδi(t)，
wherein,
ei(t)＝Sxd(t)-Sxi(t)，
wherein G () is the relationship between the normal offset and the slip form face formed by the robot deformation, ei(t) andfor slip form face error and its first derivative, Sxd(t) is an ideal grinding force slip form surface,for the adaptive term of the i-1 th iteration,is the adaptive term of the ith iteration, gamma is the iteration coefficient, i is the iteration coefficient, kp、kdGiven coefficients greater than zero;
w2, proposing relevant assumptions supporting the above algorithm stability and convergence, for proof of algorithm stability:
assume that the parameters of the system are unknown and that the system satisfies the following assumptions:
assume that 1: the initial state of the system being consistent and repeatable, i.e. Sa1(0)＝Sa2(0)＝…＝Sai(0)；
Assume 2: slip form surface SxiAnd SaiTheir first derivativesAndsecond derivative ofAndand SfIs bounded;
assume that 3:|kSaiand | ≧ epsilon, β and epsilon are constants greater than zero.
Further, in step S5, specifically, the method includes:
y1, receiving the force signal through the one-dimensional sensor, and transmitting the simulated force signal to the I/O module for processing through the signal amplifier;
the Y2 and the I/O module convert the analog signals into digital signals and transmit the digital signals to the upper computer;
y3, the upper computer is an upper computer real-time control system, and normal offset is calculated according to a control law;
and Y4, processing the calculated normal offset through the I/O module and transmitting the processed normal offset to a robot control cabinet, and controlling the robot to move by the robot control cabinet according to the received signal.
The other purpose of the invention is realized by the following technical scheme:
the robot abrasive belt grinding constant force control device based on the one-dimensional force sensor is characterized by comprising a control device and a grinding device;
the control device includes: the robot comprises a one-dimensional force sensor, a signal amplifier, an I/O module, an upper computer, a robot control cabinet and a six-degree-of-freedom robot;
the polishing device comprises: an abrasive belt sander and a workpiece machining clamp;
the six-degree-of-freedom robot and the I/O module are connected to an upper computer, and the six-degree-of-freedom robot and the robot control cabinet are connected to the I/O module;
the upper computer is used for receiving the transmission signal of the I/O module for processing, and then transmitting the processed signal to the robot control cabinet through the I/O module
The signal amplifier is respectively connected with the one-dimensional sensor and the I/O module and used for receiving the one-dimensional sensor signal and transmitting the received signal to the I/O module.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the invention enables the abrasive belt grinding force control of the robot to be carried out on the one-dimensional force sensor, avoids the problem of high cost caused by using a multi-dimensional force sensor, and simultaneously reduces the complexity of the control process;
2. the control method provided by the invention is simple to apply and set parameters, can effectively compensate errors caused by uncertainty in sanding of the abrasive belt, and is suitable for actual sanding.
Drawings
FIG. 1 is a flow chart of a method for controlling constant force of sanding belt grinding of a robot based on a one-dimensional force sensor according to the invention;
FIG. 2 is a signal transmission diagram of a robot belt sanding constant force control method based on a one-dimensional force sensor according to the invention;
FIG. 3 is a structural diagram of a robot belt sanding constant force control device based on a one-dimensional force sensor according to the invention;
FIG. 4 is a partial enlarged view of the end of a robot of the belt sanding constant force control device based on a one-dimensional force sensor according to the present invention;
FIG. 5 is an adaptive sliding mode iteration control map in the described embodiment of the invention;
FIG. 6 is a force analysis graph during grinding according to the embodiment of the present invention;
FIG. 7 is a flow chart of the ith iteration in the embodiment of the present invention;
FIG. 8 is an adaptation term in the embodiment of the present inventionThe calculation of (1).
In the figure, 1-abrasive belt sander, 2-processing workpiece, 3-processing workpiece clamp, 4-one-dimensional force sensor, 5-clamp of clamping sensor, 6-signal amplifier, 7-I/O module, 8-upper computer, 9-robot control cabinet, 10-six-freedom robot, 11-abrasive belt wheel.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Examples
A robot abrasive belt sanding constant force control method based on a one-dimensional force sensor is disclosed, as shown in figure 1,
FIG. 2 is a signal transmission diagram of a robot belt sanding constant force control algorithm based on a one-dimensional force sensor; FIG. 3 is a structural diagram of a robot belt sanding constant force control device based on a one-dimensional force sensor; fig. 4 is a partial enlarged view of the end of a robot in the constant force control device for sanding of a robot belt based on a one-dimensional force sensor. The device comprises the following specific application steps: firstly, operating an embedded real-time control system on a PC host and opening a relay; the robot clamps the workpiece and polishes the workpiece along a pre-planned track; converting analog signals collected on the force sensor into digital signals through an I/O module, and transmitting the signals to an embedded real-time control system in a PC (personal computer) through an Ethercat protocol; calculating the offset by using the self-adaptive sliding mode iterative control method provided by the invention; converting the offset of the digital signal into an analog signal through I/O (input/output) and transmitting the analog signal to a robot control cabinet; the robot receives-10V analog signals processed by an I/O module through built-in software in a control cabinet in the motion process and controls the robot to move, wherein the offset displacement direction is consistent with the analog signal sign, the offset displacement is in direct proportion to the absolute value of voltage, the frequency of an analog filter of the force sensor is 2500Hz, the sampling frequency of the embedded real-time control system is 1ms, the output voltage frequency of the control system is 100ms, and the connection between the Ethercat and the robot control cabinet is realized through the sensor function of the robot.
The specific control process diagram is shown in FIG. 5, and the robot end force f is measured by the sensorx(t) feedback to the controller with the desired sanding force fxd(t) calculating the grinding force error e by calculating the differencex(t) then calculating the first derivative thereofAnd the current polishing slip form state Sxi(t) and the desired sanding state Sxd(t) differencing to obtain an error ei(t) passing the error and combining the iteration coefficient gamma and the adaptive term of the last iterationCalculating an iteration step sizeThe normal offset amount delta x (t) is calculated, the offset amount of the robot is calculated through positive and negative kinematics and is input into a control system for control, and the grinding force is reflected on a force sensor through the action of G (delta x (t)) and the grinding environment.
According to the coordinate system pose relation and the stress analysis of fig. 6, the mapping relation between the force on the sensor coordinate system and the force on the abrasive belt wheel coordinate system can be obtained, and can be simplified by regarding the ratio between the normal force and the tangential force as a constant, wherein the ratio η can be obtained through preliminary experiments, the value is found to fluctuate within 2.5 +/-0.05, the error is about 2%, so η -2.5 can be obtained, and the estimation of the angle theta can be obtained through a force tracking method.
FIG. 7 shows the flow of the ith iteration, in which the workpiece is preprocessed before each start to eliminate unnecessary interference and ensure the consistency of each iteration condition as much as possible; when the robot is in a grinding state, the controller receives the force on the one-dimensional force sensor, calculates the grinding normal force in real time by using the simplified force mapping relation provided by the invention, and constructs a corresponding sliding mode surface by calculating the error between the actual grinding force and the expected grinding force; the actual sliding mode surface is compared with the ideal sliding mode surface to obtain a sliding mode surface error, the error is directly substituted into a grinding control law on one hand, and is substituted into the calculation process of an adaptive item on the other hand, wherein the adaptive item is calculated in real time by the sliding mode surface error at the current moment and the adaptive item in the previous iteration process, and the specific calculation process is shown in FIG. 8; the corresponding offset is obtained through the calculation of the control law and the robot is controlled, and meanwhile, the force on the one-dimensional force sensor is changed and continuously transmitted to an upper computer, so that a complete closed-loop control is formed.
In order to verify the effectiveness of the adaptive sliding mode iteration control method, the invention polishes the plane angle steel and the curved surface workpiece respectively. When the plane angle steel is polished, the included angle theta shown in fig. 6 is always zero, and the force on the one-dimensional sensor can be regarded as the polishing normal force, so that the errors caused by the force simplification relation and the included angle estimation provided by the invention are eliminated, and the effectiveness of the adaptive sliding mode iterative algorithm can be verified by a plane angle steel experiment. And then, designing a curved surface grinding experiment, and further verifying the effectiveness of the self-adaptive sliding mode iterative algorithm and the feasibility of the force simplification relation.
During plane grinding, taking Q235 angle steel with the thickness of 3mm and the size of 160mm x 40mm, and carrying out each iteration according to the flow shown in figure 7, wherein the specific steps are as follows:
step 1: and preprocessing each grinding and generating an initial grinding track, so as to ensure the consistency of conditions during each iteration as much as possible. The steps of preprocessing and generating the initial grinding track are as follows:
1) generating a preprocessed grinding track and a corresponding JOB program through an offline programming program;
2) copying the preprocessed JOB program into a robot demonstrator;
3) running a preprocessed JOB program without control;
4) generating an initial grinding track and a corresponding JOB program through an offline programming program;
5) copying the initial JOB program to a robot demonstrator;
step 2: starting debugging of a real-time control program, and receiving a force signal on a force sensor;
and step 3: let the robot carry out initial JOB procedure and polish, simultaneously, the offset is calculated to the force signal that the controller was returned through feeding back. The specific steps of the force feedback signal processing are as follows:
1) when the robot executes the JOB program, a workpiece at the tail end of the robot contacts with the abrasive belt wheel and generates grinding force;
2) a one-dimensional force sensor at the tail end of the robot receives the force signal and transmits the force signal to a signal amplifier;
3) the signal amplifier inputs the analog force signal to the I/O module for A/D conversion;
4) the I/O module transmits the force signal to a real-time control system of the upper computer;
5) in the control system, the desired sanding force was set to 20N and the initial parameter set to kp＝0.055，kdTaken together under the weight of 0.02The coefficient lambda of the sliding mode surface is 0.5, and the iteration coefficient gamma is 0.3;
6) the real-time control system calculates a corresponding sliding mode surface through the feedback force signal;
7) the real-time control system calculates the adaptive item at the current moment in real time through the sliding mode surface error and the adaptive item of the previous iteration, and the specific calculation flow is shown in fig. 8;
8) the real-time control system inputs the sliding mode surface error and the self-adaptive item of the current moment into a control law to calculate the offset which the robot should feed at the next moment;
and 4, step 4: and transmitting the calculated offset to a robot control cabinet, receiving a-10V analog signal processed by an I/O module by built-in software in the control cabinet, and controlling the robot to move, wherein the offset displacement direction is consistent with the analog signal sign, and the offset displacement is in direct proportion to the absolute value of voltage.
And 5: and after polishing is finished, removing the polished workpiece, replacing the same plane angle steel, and repeating the four steps to perform next self-adaptive iterative control.
When the plane is polished, polishing normal force in each iteration process is compared and the absolute value of the error is analyzed. When plane polishing is carried out under the self-adaptive sliding mode control algorithm, the polishing force is finally stabilized within 20 +/-2N, the average value, standard deviation and variance of the absolute value of errors basically fall, and compared with the non-iterative sliding mode control, the polishing force is respectively reduced by 46%, 38% and 62%, and effective polishing force control is realized. Meanwhile, the average value of the surface roughness of the polished workpiece is 0.2329 μm and the standard deviation is only 0.0360 μm by measuring the surface roughness of the polished workpiece, which shows that the surface roughness of the workpiece is distributed more uniformly and uniformly, and the effectiveness of the control method is reflected from the side surface.
When the curved surface is polished, a workpiece material is 45# steel, a curved surface workpiece with a curved surface contour line being a spline curve is taken, each iteration is carried out according to the flow shown in fig. 7, and the method comprises the following specific steps:
step 1: and preprocessing each grinding and generating an initial grinding track, so as to ensure the consistency of conditions during each iteration as much as possible. The steps of preprocessing and generating the initial grinding track are as follows:
1) generating a preprocessed grinding track and a corresponding JOB program through an offline programming program;
2) copying the preprocessed JOB program into a robot demonstrator;
3) running a preprocessed JOB program without control;
4) generating an initial grinding track and a corresponding JOB program through an offline programming program;
5) copying the initial JOB program to a robot demonstrator;
step 2: starting debugging of a real-time control program, and receiving a force signal on a force sensor;
and step 3: let the robot carry out initial JOB procedure and polish, simultaneously, the offset is calculated to the force signal that the controller was returned through feeding back. The specific steps of the force feedback signal processing are as follows:
1) when the robot executes the JOB program, a workpiece at the tail end of the robot contacts with the abrasive belt wheel and generates grinding force;
2) a one-dimensional force sensor at the tail end of the robot receives the force signal and transmits the force signal to a signal amplifier;
3) the signal amplifier inputs the analog force signal to the I/O module for A/D conversion;
4) the I/O module transmits the force signal to a real-time control system of the upper computer;
5) in the control system, the desired sanding force was set to 20N and the initial parameter set to kp＝0.04，kdTaken together (0.04)The coefficient lambda of the sliding mode surface is 0.5, and the iteration coefficient gamma is 0.4;
6) the real-time control system calculates a corresponding sliding mode surface through the feedback force signal;
7) the real-time control system calculates the adaptive item at the current moment in real time through the sliding mode surface error and the adaptive item of the previous iteration, and the specific calculation flow is shown in fig. 8;
8) the real-time control system inputs the sliding mode surface error and the self-adaptive item of the current moment into a control law to calculate the offset of the robot which should move at the next moment;
and 4, step 4: and transmitting the calculated offset to a robot control cabinet, receiving a-10V analog signal processed by an I/O module by built-in software in the control cabinet, and controlling the robot to move, wherein the offset displacement direction is consistent with the analog signal sign, and the offset displacement is in direct proportion to the absolute value of voltage.
And 5: and after polishing is finished, removing the polished workpiece, replacing the workpiece with the same curved surface workpiece, and repeating the four steps to perform next self-adaptive iterative control.
When the curved surface is polished, polishing normal force in each iteration process is compared and the absolute value of the error is analyzed. When the curved surface is polished under the self-adaptive sliding mode control algorithm, the polishing force is finally stabilized within 20 +/-2N, the average value, standard deviation and variance of the absolute value of errors basically fall, and compared with the non-iterative sliding mode control, the polishing force is respectively reduced by 51%, 45% and 70%, so that the effective polishing force control is realized. Meanwhile, the average value of the surface roughness of the polished workpiece is 0.2726 μm and the standard deviation is only 0.0512 μm by measuring the surface roughness of the polished workpiece, which shows that the surface roughness of the workpiece is distributed more uniformly and uniformly, and the effectiveness of the control method is reflected from the side surface.
The robot abrasive belt grinding constant force control device based on the one-dimensional force sensor comprises a control device and a grinding device;
the control device includes: the robot comprises a one-dimensional force sensor, a signal amplifier, an I/O module, an upper computer, a robot control cabinet and a six-degree-of-freedom robot;
the polishing device comprises: an abrasive belt sander and a workpiece machining clamp;
the six-degree-of-freedom robot and the I/O module are connected to an upper computer, and the six-degree-of-freedom robot and the robot control cabinet are connected to the I/O module;
the robot control cabinet further comprises a signal amplifier, the one-dimensional force sensor is connected with the signal amplifier through a signal lead, an analog signal output by the signal amplifier is connected with the I/O module, an A/D conversion function in the I/O module transmits a digital quantity signal to the embedded real-time control system, and the real-time control system transmits the calculated control quantity to the robot control cabinet through the D/A conversion function of the I/O module, so that the feedback control of the force signal is formed.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (8)
1. A robot abrasive belt grinding constant force control method based on a one-dimensional force sensor is characterized by comprising the following steps:
s1, according to the stress condition of the polishing process, performing stress analysis on the contact force between a polishing workpiece at the tail end of the robot and the abrasive belt wheel to obtain a force mapping relation between a sensor coordinate system and an abrasive belt wheel coordinate system; establishing and simplifying a force mapping relation between the contact force and a sensor coordinate system by polishing the relation between the normal force and the tangential force;
s2, establishing a robot abrasive belt wheel grinding dynamic model based on deformation through the relation between grinding deformation and grinding depth;
s3, designing a self-adaptive sliding mode iterative control algorithm, and establishing a corresponding force control model according to the grinding dynamic model of the robot abrasive belt wheel;
s4, designing a corresponding adaptive sliding mode iteration constant force controller according to the force mapping relation between the contact force and the sensor coordinate system and a corresponding force control model;
and S5, calculating and feeding back the normal grinding force by the force received by the one-dimensional force sensor and applying the contact force and the force mapping relation on the sensor coordinate system, inputting the fed-back contact force into the corresponding adaptive sliding mode iteration constant force controller to obtain a normal offset, and transmitting the normal offset to the control module for control.
2. The method for controlling constant force in sanding of abrasive belt of robot based on one-dimensional force sensor as claimed in claim 1, wherein said step S1 specifically comprises:
t1, establishing a pose relation between coordinate systems according to the model during grinding: a force sensor coordinate system, a belt wheel coordinate system; the original point of a force sensor coordinate system is the geometric center of the sensor, the X-axis direction of the force sensor coordinate system is the axial direction of the sensor, the Y-axis direction of the force sensor coordinate system is the radial direction of the sensor, and the Z-axis direction of the force sensor coordinate system is determined according to the right-hand rule of the coordinate system; the origin of a sand belt wheel coordinate system is positioned on the surface of a sand belt of the center of the sand belt wheel along the radial direction, the X-axis direction of the sand belt wheel coordinate system is the radial direction of the sand belt wheel, the Y-axis direction of the sand belt wheel coordinate system is the axial direction of the sand belt wheel, the Z-axis direction of the sand belt wheel coordinate system is determined according to the right-hand rule of the coordinate systems, and the Z-axes of the two coordinate systems are parallel during grinding;
t2, analyzing the force in the grinding process, and setting FtAnd FnGrinding tangential force and normal force on the abrasive belt wheel coordinate system respectively, Ft' and Fn' means that FtAnd FnForce transferred to the force sensor coordinate system, FxAnd FyRespectively representing one-dimensional force transmissionThe forces on the sensor in the X-axis and Y-axis are:
through the formula, solve the normal force and the tangential force of polishing on the abrasive band wheel coordinate system:
wherein theta is an included angle between Y axes of the two coordinate systems;
t3, establishing the relation between the grinding normal force and the grinding tangential force on the abrasive wheel coordinate system:
Fn＝ηFt，
η is the ratio of the normal force and the tangential force of the sanding on the abrasive belt wheel coordinate system;
t4, obtaining the force F of the X axis of the one-dimensional force sensor according to the step T2 and the step T3xGrinding normal force F on grinding belt wheel coordinate systemnThe relationship between:
Fx＝Fn(cosθ-sinθ/η)。
3. the method for controlling constant force in sanding of abrasive belt of robot based on one-dimensional force sensor as claimed in claim 1, wherein said step S2 specifically comprises:
u1, establishing a grinding force dynamic model:
wherein f isp(t) is a normal grinding force, m is a component of a system inertia matrix in the normal direction of the grinding force, c is a component of a system damping matrix in the normal direction of the grinding force, k is a component of the rigidity of the grinding process in the normal direction of the grinding force, and x (t) is the position of the cutter vertical to the surface of the workpiece, namely the grinding depth;first and second derivatives of x (t), respectively;
u2, analyzing the stress condition of the tail end of the robot in the polishing process, and establishing a stress model of the tail end of the robot:
f(t)＝fp(t)+fs(t)，
wherein f iss(t) is the deformation force of the robot during the polishing process;
u3, researching the relation between the grinding normal force and the grinding depth, and establishing the relation between the grinding deformation and the deformation force:
δx(t)＝x*(t)-x(t)，
fs(t)＝ksδx(t)，
wherein, deltax(t) is the grinding deformation of the tool perpendicular to the surface of the workpiece, x*(t) planned sanding depth, ksPolishing the static stiffness of the system for the robot;
u4, obtaining a robot belt sanding dynamic model based on deformation according to the relational expression of the steps U1, U2 and U3:
4. the method for controlling constant force in sanding of abrasive belt of robot based on one-dimensional force sensor as claimed in claim 3, wherein said step S3 specifically comprises:
v1, designing a sliding mode state according to the grinding force error:
ex(t)＝fx(t)-fxd(t)，
wherein λ is the sliding mode surface coefficient, Sx(t) slip form state at the time of grinding, ex(t)、Respectively the grinding force error and the first derivative of the grinding force error; f. ofx(t)、Respectively the moment actual polishing force and the first derivative of the moment actual polishing force; f. ofxd(t)、Respectively being the instant desired sanding force and the first derivative of the instant desired sanding force;
v2, the deformation-based robot belt sanding dynamics model established according to step S2, and the sliding mode state of step V1, then:
wherein m is the component of the system inertia matrix in the normal direction of the grinding force, c is the component of the system damping matrix in the normal direction of the grinding force, k is the component of the rigidity in the normal direction of the grinding force in the grinding process, SxiThe sliding mode state is the sliding mode state in the ith iteration;andrespectively forming a first derivative and a second derivative of a sliding mode surface by the processing depth;
and has the following components:
wherein S isxiFor actual grinding force of slip form surface, Sai(t) slip form surface formed by working depth，Sδi(t) is a slip form surface formed by working deformation, Sf(t) a slip form surface formed by ideal machining force; deltaxi(t) andthe component of the polishing deformation in the normal direction of the workpiece and the first derivative thereof are included;
v3, according to step V2, and the desired sliding mode state, i.e. SxiWhen 0, a force control model is obtained:
。
5. a robot belt sanding constant force control method based on one-dimensional force sensor according to claims 2 and 4, characterized in that the step S4 specifically comprises:
w1, designing a control law of the corresponding adaptive sliding mode iterative constant force controller according to the simplified force mapping relation in the step S1 and the force control model in the step S3:
G(Δx(t))＝ksSδi(t)，
wherein,
ei(t)＝Sxd(t)-Sxi(t)，
wherein G () is the relationship between the normal offset and the slip form face formed by the robot deformation, ei(t) andfor slip form face error and its first derivative, Sxd(t) is an ideal grinding force slip form surface,for the adaptive term of the i-1 th iteration,is the adaptive term of the ith iteration, gamma is the iteration coefficient, i is the iteration coefficient, kp、kdGiven coefficients greater than zero;
w2, proposing relevant assumptions supporting the above algorithm stability and convergence, for proof of algorithm stability:
assume that the parameters of the system are unknown and that the system satisfies the following assumptions:
assume that 1: the initial state of the system being consistent and repeatable, i.e. Sa1(0)＝Sa2(0)＝…＝Sai(0)；
Assume 2: slip form surface SxiAnd SaiTheir first derivativesAndsecond derivative ofAndand SfIs bounded;
assume that 3:|kSaiand | ≧ epsilon, β and epsilon are constants greater than zero.
6. The method for controlling constant force in sanding of abrasive belt of robot based on one-dimensional force sensor as claimed in claim 1, wherein said step S5 specifically comprises:
y1, receiving the force signal through the one-dimensional sensor, and transmitting the simulated force signal to the I/O module for processing through the signal amplifier;
the Y2 and the I/O module convert the analog signals into digital signals and transmit the digital signals to the upper computer;
y3, an upper computer real-time control system, and calculating the normal offset according to the control law;
and Y4, processing the calculated normal offset through the I/O module and transmitting the processed normal offset to a robot control cabinet, and controlling the robot to move by the robot control cabinet according to the received signal.
7. The robot abrasive belt grinding constant force control device based on the one-dimensional force sensor is characterized by comprising a control device and a grinding device;
the control device includes: the robot comprises a one-dimensional force sensor, a clamp for clamping the sensor, an I/O module, an upper computer, a robot control cabinet and a six-degree-of-freedom robot;
the polishing device comprises: an abrasive belt sander and a workpiece machining clamp;
the six-degree-of-freedom robot and the I/O module are connected to an upper computer, and the six-degree-of-freedom robot and the robot control cabinet are connected to the I/O module;
and the upper computer is used for receiving the I/O module transmission signal for processing and then transmitting the processed signal to the robot control cabinet through the I/O module.
8. The constant force control device for sanding with robot based on one-dimensional force sensor as claimed in claim 7, further comprising signal amplifier connected with the one-dimensional sensor and the I/O module respectively for receiving the one-dimensional sensor signal and transmitting the received signal to the I/O module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811523895.1A CN109664295B (en) | 2018-12-13 | 2018-12-13 | Robot abrasive belt grinding constant force control method and device based on one-dimensional force sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811523895.1A CN109664295B (en) | 2018-12-13 | 2018-12-13 | Robot abrasive belt grinding constant force control method and device based on one-dimensional force sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109664295A true CN109664295A (en) | 2019-04-23 |
| CN109664295B CN109664295B (en) | 2021-12-21 |
Family
ID=66143766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811523895.1A Active CN109664295B (en) | 2018-12-13 | 2018-12-13 | Robot abrasive belt grinding constant force control method and device based on one-dimensional force sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109664295B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110497423A (en) * | 2019-08-22 | 2019-11-26 | 泉州装备制造研究所 | A kind of manipulator adaptive machining method |
| CN110561237A (en) * | 2019-10-08 | 2019-12-13 | 华中科技大学 | Robot abrasive belt grinding method and system combining active and passive power control |
| CN112846640A (en) * | 2020-12-15 | 2021-05-28 | 贵州博泰自动化科技有限公司 | Pressure checking method and machining equipment |
| CN114131605A (en) * | 2021-12-06 | 2022-03-04 | 上海智能制造功能平台有限公司 | Automatic ship part polishing track registration adjusting device and method |
| CN116533237A (en) * | 2023-05-09 | 2023-08-04 | 浙江钧控智能科技有限公司 | Fuzzy variable impedance control method integrating attitude measurement and one-dimensional force sensor information |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5441437A (en) * | 1993-02-18 | 1995-08-15 | Hulstedt; Bryan A. | Compliant constant-force follower device for surface finishing tool |
| CN103009218A (en) * | 2012-12-17 | 2013-04-03 | 华南理工大学 | Robot polishing device and polishing method thereof |
| CN104723210A (en) * | 2015-02-14 | 2015-06-24 | 广东工业大学 | Electromagnetism one-dimensional constant force device and control method thereof |
| CN104858782A (en) * | 2015-04-03 | 2015-08-26 | 华南理工大学 | Constant pressure automatic grinding device and method based on fuzzy adaptive force control |
| CN106383495A (en) * | 2016-09-12 | 2017-02-08 | 华南理工大学 | Curved surface profile constant force tracking method and application based on non-linear double closed loop control |
| CN107962480A (en) * | 2017-11-28 | 2018-04-27 | 华中科技大学 | Force control method is processed in a kind of blade robot sbrasive belt grinding |
-
2018
- 2018-12-13 CN CN201811523895.1A patent/CN109664295B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5441437A (en) * | 1993-02-18 | 1995-08-15 | Hulstedt; Bryan A. | Compliant constant-force follower device for surface finishing tool |
| CN103009218A (en) * | 2012-12-17 | 2013-04-03 | 华南理工大学 | Robot polishing device and polishing method thereof |
| CN104723210A (en) * | 2015-02-14 | 2015-06-24 | 广东工业大学 | Electromagnetism one-dimensional constant force device and control method thereof |
| CN104858782A (en) * | 2015-04-03 | 2015-08-26 | 华南理工大学 | Constant pressure automatic grinding device and method based on fuzzy adaptive force control |
| CN106383495A (en) * | 2016-09-12 | 2017-02-08 | 华南理工大学 | Curved surface profile constant force tracking method and application based on non-linear double closed loop control |
| CN107962480A (en) * | 2017-11-28 | 2018-04-27 | 华中科技大学 | Force control method is processed in a kind of blade robot sbrasive belt grinding |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110497423A (en) * | 2019-08-22 | 2019-11-26 | 泉州装备制造研究所 | A kind of manipulator adaptive machining method |
| CN110497423B (en) * | 2019-08-22 | 2022-08-16 | 泉州装备制造研究所 | Self-adaptive machining method for manipulator |
| CN110561237A (en) * | 2019-10-08 | 2019-12-13 | 华中科技大学 | Robot abrasive belt grinding method and system combining active and passive power control |
| CN110561237B (en) * | 2019-10-08 | 2020-07-24 | 华中科技大学 | Robot abrasive belt grinding method and system combining active and passive power control |
| CN112846640A (en) * | 2020-12-15 | 2021-05-28 | 贵州博泰自动化科技有限公司 | Pressure checking method and machining equipment |
| CN114131605A (en) * | 2021-12-06 | 2022-03-04 | 上海智能制造功能平台有限公司 | Automatic ship part polishing track registration adjusting device and method |
| CN114131605B (en) * | 2021-12-06 | 2024-03-22 | 上海智能制造功能平台有限公司 | Automatic registration adjustment device and method for ship part polishing track |
| CN116533237A (en) * | 2023-05-09 | 2023-08-04 | 浙江钧控智能科技有限公司 | Fuzzy variable impedance control method integrating attitude measurement and one-dimensional force sensor information |
| CN116533237B (en) * | 2023-05-09 | 2024-08-27 | 浙江钧控智能科技有限公司 | Fuzzy variable impedance control method integrating attitude measurement and one-dimensional force sensor information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109664295B (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109664295B (en) | Robot abrasive belt grinding constant force control method and device based on one-dimensional force sensor | |
| CN105643399B (en) | The complex-curved automatic grinding-polishing system of robot and processing method based on Shared control | |
| Cen et al. | A wireless force-sensing and model-based approach for enhancement of machining accuracy in robotic milling | |
| Ke et al. | Review on robot-assisted polishing: Status and future trends | |
| Zhang et al. | Machining with flexible manipulator: toward improving robotic machining performance | |
| US9110456B2 (en) | Robotic machining with a flexible manipulator | |
| CN110948504B (en) | Normal constant force tracking method and device for robot machining operation | |
| CN104626168B (en) | Robot Force position based on intelligent algorithm Shared control method | |
| Trigatti et al. | A new path-constrained trajectory planning strategy for spray painting robots-rev. 1 | |
| CN111975579B (en) | Robot constant-force polishing system based on polishing model and iterative algorithm | |
| CN108115705A (en) | A kind of robot polishing control system and method | |
| Lakshminarayanan et al. | An adaptive framework for robotic polishing based on impedance control | |
| Fan et al. | The integral sliding mode control of a pneumatic force servo for the polishing process | |
| CN110315396B (en) | Industrial robot constant-force grinding and polishing method based on big data | |
| Yin et al. | Research on machining error prediction and compensation technology for a stone-carving robotic manipulator | |
| CN109927028A (en) | A kind of force-location mix control method of power man-controlled mobile robot polishing processing | |
| Zhang et al. | A review of surface quality control technology for robotic abrasive belt grinding of aero-engine blades | |
| Jin et al. | Effect of downward depth and inflation pressure on contact force of gasbag polishing | |
| Shi et al. | NC polishing of aspheric surfaces under control of constant pressure using a magnetorheological torque servo | |
| CN113156819A (en) | Self-adaptive neural network synchronous impedance control method for coordinated grinding mechanical arm system | |
| CN114454060A (en) | Robot self-adaptive curved surface tracking constant-force grinding and polishing method and system | |
| CN115369223B (en) | Double-sided ultrasonic rolling cooperative reinforcement system and control method thereof | |
| CN110281152B (en) | Robot constant-force polishing path planning method and system based on online touch test | |
| Iglesias et al. | Robotic path compensation training method for optimizing face milling operations based on non-contact CMM techniques | |
| CN114274047B (en) | Efficient precise polishing track optimization method based on force sensing measurement |
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 |