WO2015055085A1 - 基于虚拟现实技术的无损检测机器人智能检测方法 - Google Patents

基于虚拟现实技术的无损检测机器人智能检测方法 Download PDF

Info

Publication number
WO2015055085A1
WO2015055085A1 PCT/CN2014/087770 CN2014087770W WO2015055085A1 WO 2015055085 A1 WO2015055085 A1 WO 2015055085A1 CN 2014087770 W CN2014087770 W CN 2014087770W WO 2015055085 A1 WO2015055085 A1 WO 2015055085A1
Authority
WO
WIPO (PCT)
Prior art keywords
destructive testing
motion
robot
axis
simulation model
Prior art date
Application number
PCT/CN2014/087770
Other languages
English (en)
French (fr)
Inventor
王可庆
黄晓辰
朱雪宏
陈建
李明
林戈
陈怀东
林忠元
丁承君
张明路
Original Assignee
中广核检测技术有限公司
苏州热工研究院有限公司
中国广核集团有限公司
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 中广核检测技术有限公司, 苏州热工研究院有限公司, 中国广核集团有限公司 filed Critical 中广核检测技术有限公司
Priority to BR112016008313A priority Critical patent/BR112016008313B8/pt
Priority to EP14854066.9A priority patent/EP3059738B1/en
Publication of WO2015055085A1 publication Critical patent/WO2015055085A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
    • G21C—NUCLEAR REACTORS
    • G21C17/00—Monitoring; Testing ; Maintaining
    • G21C17/003—Remote inspection of vessels, e.g. pressure vessels
    • G—PHYSICS
    • G21—NUCLEAR PHYSICS; NUCLEAR ENGINEERING
    • G21D—NUCLEAR POWER PLANT
    • G21D3/00—Control of nuclear power plant
    • G21D3/001—Computer implemented control
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E30/00—Energy generation of nuclear origin
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E30/00—Energy generation of nuclear origin
    • Y02E30/30—Nuclear fission reactors

Definitions

  • the invention belongs to the technical field of surface detection, and particularly relates to an intelligent detection method for a non-destructive testing robot based on virtual reality technology in a nuclear power plant reactor pressure vessel.
  • Non-destructive testing uses the characteristics of sound, light, magnetism and electricity of a substance to detect defects or inhomogeneities in the object to be inspected without impairing or affecting the performance of the object to be inspected, giving the size and location of the defect. , nature and quantity of information.
  • Non-destructive testing mainly includes four types: radiographic inspection (RT), ultrasonic inspection (UT), magnetic particle inspection (MT) and liquid permeation detection (PT).
  • Non-destructive testing methods include eddy current testing (ET), acoustic emission testing (AT), thermal imaging/infrared (TIR), leakage testing (LT), AC field measurement (ACFMT), magnetic flux leakage testing (MFL), and far field testing.
  • Detection method RFT
  • TOFD ultrasonic diffraction time difference method
  • non-destructive testing has the following characteristics: (1) non-destructive, because it does not damage the performance of the tested object when it is tested; (2) comprehensive, because the detection is non-destructive, it is necessary 100% comprehensive inspection of the object to be tested, which is not possible for destructive testing; (3) full-scale, destructive testing is generally only suitable for testing raw materials, such as stretching commonly used in mechanical engineering , compression, bending, etc., destructive testing is carried out for the production of raw materials, for finished products and supplies, unless it is not ready to continue to serve, otherwise it can not be destructive testing, and non-destructive testing is not detected due to damage The performance of the object. Therefore, it can not only test the raw materials for manufacturing, the intermediate process links, and the final finished products, but also the equipment in service.
  • Nuclear power plant reactor pressure vessels have a large number of irregular shaped components. At present, the surface ultrasonic/eddy current testing of such components is mostly manual mode.
  • the non-destructive testing robot control software for reactor pressure vessel developed and used by non-destructive testing companies in the field of nuclear power at home and abroad adopts no-graphic or two-dimensional graphical display control methods.
  • the disadvantages of this display control method are as follows: Due to the importance of the reactor pressure vessel for the safe operation of nuclear power plants, its non-destructive testing process must also be safe and controllable, due to the environment in which the reactor pressure vessel is located and the reactor pressure vessel itself. Shape characteristics, after the installation of the detection robot, each axis will be in an environment where manual monitoring and camera detection are difficult to comprehensively monitor.
  • the no-line display control mode checks all the robots in the case of the robot's own attitude and the positional relationship with respect to the pressure vessel. The safety of sports will be difficult to guarantee.
  • the two-dimensional graphic display control can display the positional relationship between the linear motion axis and the reactor pressure vessel in a two-dimensional figure in one plane, but it cannot display the posture and positional relationship of all the axes of the actual equipment. In order to make the non-destructive testing robot reach the target inspection position, the operator needs to repeatedly adjust according to experience, excessive adjustment causes the motion efficiency to be low, increases the equipment failure and the probability of misoperation, and reduces the motion control efficiency.
  • the object of the present invention is to provide a non-destructive detection robot intelligent detection method based on virtual reality technology, which solves the problem that the non-destructive detection robot performs non-destructive detection through no-line display control or two-dimensional graphic control in the prior art, and requires an operator to have higher
  • the technical level and rich adjustment experience are repeated adjustment and motion control, and the non-destructive testing efficiency is low, and the detection needs to be adjusted according to experience and other technical problems.
  • An intelligent detection method for non-destructive testing robot based on virtual reality technology characterized in that before the detection, a virtual reality technology is needed to construct a non-destructive testing robot simulation model and a reactor pressure vessel simulation model, and the method comprises the following steps:
  • Non-destructive testing robot simulation model based on real-time non-destructive testing robots
  • the position and attitude information feedback value of the motion axis is used to perform position and attitude transformation in the three-dimensional virtual environment, and virtual display and control of the non-destructive detection robot synchronous motion for non-destructive detection.
  • the non-destructive testing robot simulation model and the reactor pressure vessel simulation model are constructed by the three-dimensional modeling software, the non-destructive testing robot simulation model and the reactor pressure vessel simulation model are started, and the non-destructive testing is started with the reactor pressure vessel as an application scenario.
  • the robot moves the axes of motion to the initial position.
  • mapping between the simulation model and the simulated hardware device in the method is performed according to the following steps:
  • the non-destructive testing robot simulation model and the reactor pressure vessel simulation model are started for simultaneous processing.
  • the reactor pressure vessel simulation model is used as the application scenario, and the non-destructive testing robot simulation is performed.
  • the motion axes of each degree of freedom of the model are synchronized to the position coordinates after calibration;
  • the step of performing the calibration of each degree of freedom motion axis position in the method step (2) comprises:
  • the motion axial negative mechanical limit moves to detect whether the motion axis reaches the negative mechanical limit.
  • the specific steps of controlling the synchronous motion of the non-destructive testing robot in the method step (4) include:
  • Non-destructive testing robots are moved to the control card according to a predetermined period after the movement of each degree of freedom motion axis or The control box sends the encoder feedback values of the motion axes of the degrees of freedom of the non-destructive testing robot;
  • the control card or the control box reads the encoder return value of each degree of freedom motion axis of the non-destructive detection robot, and converts the movement distance of the linear axis and the movement angle of the rotation axis in each of the degrees of freedom of the non-destructive detection robot;
  • Non-destructive testing robot simulation model and reactor pressure vessel simulation model Through the determination of three-dimensional coordinates and the graph transformation algorithm, the actual physical quantity represented by the encoder return value is solved, and the solution for driving the non-destructive testing robot simulation model is obtained.
  • the non-destructive testing robot simulation model transformation realizes the synchronous motion of the non-destructive testing robot simulation model and the actual non-destructive testing robot in virtual reality.
  • the non-destructive testing robot in the method is provided with a mounting arm that cooperates with a specific installation station of a nuclear power plant reactor pressure vessel, and a pneumatic centering device is arranged at each end of the mounting arm to be inserted and matched with a specific installation station of the nuclear power plant reactor pressure vessel;
  • a mounting post is disposed between the mounting arms, the mounting arm is fixedly connected to the column, and the lower end of the column is connected with a plurality of mechanical arms for performing non-destructive testing; the end of the mechanical arm is fixed to the probe tool holder.
  • the method assumes that the mechanical arm of the non-destructive detecting robot is regarded as a mechanism connecting a plurality of intermediate links between two terminal links, and the connecting rods are connected by joints; assuming from the non-actuator end to the actuator end
  • the links are numbered 0, 1, 2, ..., n, and the motion pair connecting the i-1th link and the i-th link is recorded as the i-th joint, and the i-th is established at the i-th joint.
  • the coordinate system Fi defined by the origin O i and the coordinate axes X i , Y i , Z i is fixed to the i-1th link, and the coordinate system Fi is defined as follows:
  • Z i is defined as the axis of the i-th joint; the positive direction of the axis has two possibilities on a straight line, the Z i axis of the moving joint is at any position, and the joint defines its direction;
  • X i is defined as the common perpendicular of Z i-1 and Z i , the direction is Z i-1 points to Z i ; if Z i-1 and Z i intersect, the positive direction of X i is arbitrarily specified, if unit vector i i , k i-1 and k i are respectively fixed on X i , Z i-1 and Z i , then i i is defined as k i-1 ⁇ k i ; if Z i-1 and Z i are parallel, X is defined i passes the origin of the i-1th coordinate system;
  • An overall global coordinate system is constructed for each robotic arm of the non-destructive testing robot as defined above.
  • the graph transformation parameter table used for the position and posture transformation is:
  • the method step (4) performs position and posture transformation, if there is interference between the non-destructive testing robot simulation model and the reactor pressure vessel simulation model, the motion of each motion axis of the non-destructive testing robot is stopped; otherwise, the running detection is continued.
  • the motion of the motion axis in the method can be determined by predetermining the motion trajectory, and then pressing Perform motion detection by moving according to the motion trajectory.
  • the method for detecting the motion trajectory includes the following steps:
  • the non-destructive testing robot performs non-destructive testing along the running track according to the detection instruction.
  • Another object of the present invention is to provide a non-destructive detection robot intelligent detection system, characterized in that the system comprises a virtual reality technology for constructing a non-destructive detection robot simulation model and a reactor pressure vessel simulation model, and is used for non-destructive testing.
  • the non-destructive testing robot is installed at a predetermined position in the reactor pressure vessel to be tested; after returning each motion axis of the non-destructive detecting robot to the initial state, and calibrating the end position of the motion axis of one degree of freedom, determining the working zero point; and zeroing the working zero point
  • the coordinates are used to construct a global coordinate system and an axis coordinate system; the non-destructive testing robot simulation model is associated with the simulated non-destructive testing robot and the corresponding relationship between the reactor pressure vessel simulation model and the simulated post-calibration reactor pressure vessel is established.
  • 3D virtual display module and,
  • the non-destructive testing robot simulation model performs the position and attitude transformation in the three-dimensional virtual environment according to the real-time non-destructive detection of the position and attitude information feedback values of the various degrees of freedom motion axes of the robot, and virtually displays and controls the non-destructive testing robot for non-destructive testing.
  • the three-dimensional virtual display module includes a view operation sub-module for performing a standard view operation and an arbitrary view adjustment operation on the non-destructive detection robot simulation model and the reactor pressure vessel simulation model in the use scenario, for the non-destructive detection robot simulation model and the reactor pressure
  • the model blanking module for displaying and hiding operations in the environment in which the container simulation model is located, the initial pose data for each part of the non-destructive testing robot simulation model and the reactor pressure vessel simulation model, traversing the current position of all parts in the model assembly And storing the location information into the database for updating the initial location setting module, for reading the initial position information of all the parts in the model assembly in the database, and moving the corresponding model to the initial position of the reset module for use in the model
  • the current position is a reference, a motion capture start module for calculating the next working position of the model, a capture end module for suspending the model motion, and a clearing error module for clearing the model motion error and realizing the position of the simulation model and the actual object.
  • the non-destructive detection robot When the user sends the detection command, the non-destructive detection robot receives the user's instruction, and by analyzing the motion track of the instruction and the motion axis, it is transmitted to the motor driver of the non-destructive detection robot and the motor controlled by the motor driver, and the motor drives each motion axis to move along the motion track. Check to the target coordinates.
  • the amount of motion required for each motion axis depends on the relationship between the current position and the target position, and may be affected by other specific points or specific conditions. If the non-destructive testing robotic device does not have a control algorithm to solve each The motion process and motion trajectory planning of the motion axis cannot complete intelligent motion control according to the target position, current position, objective existing constraints and specific position arrival requirements.
  • the invention realizes the three-dimensional real-time display control detection through the virtual reality technology, and reflects the posture of the movement axis of the reactor and the relative positional relationship between the robot and the reactor pressure vessel by means of the full-scale full-scale three-dimensional display.
  • a WYSIWYG display control method for simulating the actual field working environment and working conditions, and to provide an all-dimensional display and monitoring of the entire non-destructive inspection activity through view change, component blanking, and interference detection. Monitoring all inspection movements and motion control of the robot plays an extremely important safety guarantee and display control function.
  • the surface ultrasonic detecting probe or the eddy current detecting probe fixed on the movement axes of the respective degrees of freedom of the robot is non-destructively detected.
  • a preferred technical solution is: the non-destructive testing robot setting There are 2 to 20 degrees of freedom of motion axes.
  • the non-destructive detecting robot is provided with at least two motion axes; each of the motion axes has at least 2 degrees of freedom. More preferably, the non-destructive testing robot is provided with two motion axes; each motion axis has 4 degrees of freedom.
  • the motion axes of the four degrees of freedom are sequentially X, Y, and Z axes that can be accurately moved in a predetermined space (three-dimensional coordinate system).
  • These axes of motion are connected to the servo motor and control the motion stepping or angle of rotation via a control card or control box.
  • the motion of each motion axis of the non-destructive testing robot can be performed by moving the motion axes of each degree of freedom in sequence, or simultaneously.
  • the sequential motion according to the present invention means that when the motion axis of one degree of freedom is moved to the stop by the control command and the corresponding control parameter, the motion axis of the other degree of freedom starts to move.
  • the simultaneous movement according to the present invention means that the motion axes of multiple degrees of freedom perform simultaneous motion according to the respective received control commands and corresponding control parameters, which may be stopped at the same time or may not be stopped at the same time; The path of the respective axes of motion is determined.
  • the X-axis motion is a linear motion in the horizontal direction of the X-axis in the reactor pressure vessel of the nuclear power plant;
  • the Y-axis motion is a linear motion in the horizontal direction of the Y-axis in a three-dimensional space coordinate system (predetermined nuclear power plant reactor pressure vessel space) And perpendicular to the X-axis motion direction;
  • the Z-axis motion is a linear motion in the vertical direction in the three-dimensional coordinate system (predetermined nuclear power plant reactor pressure vessel space), which is composed of the X-axis motion direction and the X-axis motion direction.
  • the W-axis motion is a 360-degree rotational motion of the W-axis in a three-dimensional space coordinate system (predetermined nuclear power plant reactor pressure vessel space).
  • the motion scheme of the present invention obtains the motion trajectory
  • the specific method for forming the control command of the motion axes of the respective degrees of freedom and the corresponding control parameters by the motion trajectory is not described in detail.
  • Commercially available programmable control cards or control boxes are available for reference. After the corresponding motion trajectory is programmed according to the corresponding operating procedures of the nuclear power plant, the control command or the control box can be formed to form the control commands and corresponding control parameters of the motion axes of the respective degrees of freedom.
  • These control cards or control boxes can control the servo motors of the various degrees of freedom motion axes for corresponding motion processing.
  • the non-destructive testing robot intelligent detecting system and the intelligent detecting method of the invention effectively simplify the detection process, improve the detection precision and speed, and reduce the irradiation dose level of the inspector in service.
  • the robot realizes the all-dimensional stereoscopic WYSIWYG display control mode, which greatly improves the safety of on-site project implementation, improves project implementation efficiency, improves robot control efficiency and control performance, and reduces Unnecessary operations and reduced probability of human error occur, improving the overall performance stability, safety and operability of the robot control system.
  • FIG. 1 is a structural block diagram of a non-destructive testing robot intelligent detecting system
  • FIG. 2 is a schematic diagram of a connection when a non-destructive testing robot intelligent detecting system performs a detecting operation
  • FIG. 3 is a schematic structural view of a non-destructive testing robot
  • FIG. 4 is a flow chart of simulating reality in a three-dimensional environment when the non-destructive testing robot intelligent detecting system performs a detecting operation
  • FIG. 5 is a schematic diagram of coordinate transformation between a non-destructive testing robot simulation model and an actual non-destructive testing robot
  • FIG. 6 is a flow chart showing a specific operation of the non-destructive testing robot intelligent detection system for detecting operations
  • Figure 7 is a flow chart of the synchronous movement of the non-destructive testing robot simulation model and the actual non-destructive testing robot.
  • Figure 8 is a calibration flow chart of an actual non-destructive testing robot
  • Figure 9 is a schematic diagram of the feedback and coordinate transformation process of the non-destructive testing robot simulation model and the actual non-destructive testing robot motion.
  • FIG. 10 is an overall flow chart of the non-destructive testing robot intelligent detection system performing the detecting operation.
  • the non-destructive testing robot intelligent detection system comprises a virtual reality technology for constructing a non-destructive testing robot simulation model and a reactor pressure vessel simulation model, and is used for installing a non-destructive testing nondestructive testing robot to the reactor to be tested.
  • a three-dimensional virtual display module that establishes a correspondence between the non-destructive testing robot simulation model and the simulated non-destructive testing robot and a correspondence between the reactor pressure vessel simulation model and the simulated post-calibration reactor pressure vessel; for non-destructive testing
  • the robot simulation model performs position and attitude transformation in the three-dimensional virtual environment based on the real-time acquired non-destructive detection robot position and attitude information feedback values of each degree of freedom motion axis, and virtually displays and controls the non-destructive detection robot for non-destructive detection. Module.
  • the three-dimensional virtual display module includes a view operation sub-module for performing a standard view operation and an arbitrary view adjustment operation on the non-destructive detection robot simulation model and the reactor pressure vessel simulation model in the use scenario, for the non-destructive detection robot simulation model and the reactor pressure
  • the model blanking module for displaying and hiding operations in the environment in which the container simulation model is located, the initial pose data for each part of the non-destructive testing robot simulation model and the reactor pressure vessel simulation model, traversing the current position of all parts in the model assembly And storing the location information into the database for updating the initial location setting module, for reading the initial position information of all the parts in the model assembly in the database, and moving the corresponding model to the initial position of the reset module for use in the model
  • the current position is a reference, a motion capture start module for calculating the next working position of the model, a capture end module for suspending the model motion, and a clearing error module for clearing the model motion error and realizing the position of the simulation model and the actual object.
  • the non-destructive testing robot intelligent detection system is installed on the upper computer, and is connected with the control card to control the non-destructive detecting robot.
  • the non-destructive testing robot has a surface ultrasonic detecting probe or an eddy current detecting probe fixed on each movement axis.
  • the non-destructive testing robot is provided with 11 motion axes, of which at least 2 motion axes are provided with 4 degrees of freedom.
  • the four degrees of freedom are, in order, X-axis motion, Y-axis motion, Z-axis motion, and W-axis motion for performing a 360° range of rotational motion in a predetermined space in a predetermined space.
  • These motion axes and servo motors can be A single servo motor can also be connected to multiple servo motors for motion axis drive.
  • the motion step or rotation angle can be controlled by a control card or control box.
  • the X-axis motion is a linear motion of the X-axis in a nuclear reactor reactor pressure vessel;
  • the Y-axis motion is a Y-axis linear motion in a nuclear power plant reactor pressure vessel, and is perpendicular to the X-axis motion direction;
  • the axis motion is the linear motion of the Z-axis in the vertical direction of the reactor pressure vessel of the nuclear power plant.
  • the plane perpendicular to the X-axis movement direction and the X-axis movement direction is the W-axis movement.
  • the W-axis is 360° in the reactor pressure vessel of the nuclear power plant. Rotational motion within range. The principle of motion process will be described in detail in the process of obtaining the motion trajectory.
  • the non-destructive testing robot is shown in FIG. 3, and is provided with a mounting arm 1 (three for supporting the non-destructive testing robot and being installed on the installation station), each of which is matched with a specific installation station of the nuclear power plant reactor pressure vessel.
  • the pneumatic centering device 2 is arranged at the end of the mounting arm and is inserted and matched with the specific installation station of the nuclear power plant reactor pressure vessel.
  • the mounting arm is fixedly connected to the body of the non-destructive detecting robot, the column 3, and the lower end of the column 3 is provided with a plurality of mechanical arms 4 for performing non-destructive testing according to the condition of the component to be detected in the reactor of the nuclear power plant reactor.
  • a flange scanning robot arm, a nozzle scanning robot arm, a lower head scanning robot arm, etc. may be provided; each of the robot arms is provided with a plurality of degrees of freedom motion axes. .
  • the end position of the moving shaft can be set according to the actual situation.
  • the front end rotating pair 5 is fixed on the front end rotating pair, and the probe tool holder 6 connected to the detecting probe is fixed to the target position through the moving shaft.
  • the specific installation process is not detailed.
  • a number of installation stations are preset in the nuclear power plant reactor pressure vessel, and the nuclear power plant operator can install the non-destructive testing robot to the predetermined station according to the prompt of installing the guide column.
  • the motion axes of the non-destructive testing robot are restored to the initial state, and the motion axis position calibrated to one degree of freedom is the working zero point.
  • the end position of the motion axis of the vertical degree of freedom of the column is zero-point coordinates (0, 0, 0).
  • the position axes of each degree of freedom motion axis are then calibrated, and the end position of the motion axis of one degree of freedom is determined as the working zero point, and the axis coordinate system of the global coordinate system and the motion axes of the respective degrees of freedom is constructed.
  • the steps of performing position calibration of each degree of freedom motion axis include:
  • the motion axial negative mechanical limit moves to detect whether the motion axis reaches the negative mechanical limit.
  • the motion axis enable, speed, software limit, and controller limit are initialized. Then, the software starts calibration, the calibration program releases the control card, and the motion axis is negatively mechanically limited. Movement, always detecting whether the motion axis reaches the negative mechanical limit; after reaching the negative mechanical limit, it moves in the positive direction. The amount of motion depends on the characteristics of each axis and the predetermined position to be reached, and the calibration action is completed.
  • the calibrated robot establishes a corresponding relationship with the non-destructive testing robot and the reactor pressure vessel in the three-dimensional virtual reality environment.
  • mapping between the simulation model and the simulated hardware device is performed according to the following steps:
  • the non-destructive testing robot is equipped with basic motion control function test and safety function test after assembly. test. After the system forms the software, the software controls the non-destructive testing robot to perform the position calibration with eleven degrees of freedom, confirms that the position calibration is completed and the position calibration is correct, then opens the three-dimensional virtual display interface for synchronization, and after the synchronization is completed, the hardware device and the virtual reality are established. The correspondence between the posture and position of the simulation device.
  • the non-destructive testing robot simulation model and the reactor pressure vessel simulation model have been constructed.
  • the construction method can use SolidWorks software as the display software, and on the basis of the secondary development, the API provided by the SolidWorks software calls the function of the three-dimensional software itself, and directly develops the software module of the specific function based on the existing function. For example, using Visual C# software for secondary development, writing plug-ins in SolidWorks, and using plug-ins to control 3D models in SolidWorks software.
  • the simulation model of various non-destructive testing robots and reactor pressure vessels is constructed by using the API provided by SolidWorks software, and a three-dimensional real-time display interface is provided to display the conditions on site.
  • Consistent operating environment, real-time display of non-destructive testing robot operating in the reactor pressure vessel can adjust the virtual model to the standard view and arbitrary angle view, the graphics do not appear distortion, deformation. And to facilitate the expression of the model, increase the display and hiding operations of the environment in which the model is located. These constitute sub-modules of the three-dimensional virtual display module.
  • the view operation sub-module can adjust the standard view and the arbitrary angle view of the virtual model without distortion or distortion.
  • the virtual model can be adjusted at any angle.
  • the model is zoomed in and out in real time, and the adjustment view is mainly divided into two adjustment modes. The first one is that the arbitrary angle adjustment mode can adjust any angle, input the angle to be rotated, and realize the model angle adjustment of any angle by clicking the corresponding adjustment button; the second is the standard view operation mode by calling the SolidWorks API function. Add buttons to the programming interface to implement standard view operations such as virtual model front view, bottom view, and right view.
  • the model blanking module is used to facilitate the expression of the model.
  • the nuclear reactor pressure vessel simulation model is displayed and hidden by clicking the corresponding display and hidden buttons.
  • the control software controls the movement of the robot, and obtains the feedback value of each axis encoder of the robot in real time.
  • the feedback value represents the position and attitude information of the respective motion axes of the robot.
  • the three-dimensional model passes through the global coordinate system, the motion axis degree of freedom coordinate system, and the robot.
  • the calculation of the motion model and other elements, the actual motion value and the current pose of the robot are calculated and the 3D model is driven to perform matrix calculation to complete the robot.
  • the model is accurately transformed with the reactor pressure vessel model in a three-dimensional virtual environment to perform the function of reflecting the virtual display control of the robot and the detected device.
  • the specific steps of controlling the synchronous motion of the non-destructive testing robot include:
  • Non-destructive testing robot The motion feedback axes of each degree of freedom motion axis of the non-destructive testing robot are sent to the control card or control box according to a predetermined period after the movement of each of the degrees of freedom motion axes;
  • the control card or the control box reads the encoder return value of each degree of freedom motion axis of the non-destructive detection robot, and converts the movement distance of the linear axis and the movement angle of the rotation axis in each of the degrees of freedom of the non-destructive detection robot;
  • Non-destructive testing robot simulation model and reactor pressure vessel simulation model Through the determination of three-dimensional coordinates and the graph transformation algorithm, the actual physical quantity represented by the encoder return value is solved, and the solution for driving the non-destructive testing robot simulation model is obtained.
  • the non-destructive testing robot simulation model transformation realizes the synchronous motion of the non-destructive testing robot simulation model and the actual non-destructive testing robot in virtual reality.
  • the operator can perform view transformation, component blanking, interference detection and other functions on the three-dimensional virtual reality interface according to the operating habits, the trajectory of the movement of the operating robot, and the monitoring procedure.
  • the non-destructive testing robot simulation model is a three-dimensional virtual display for the motion process of the non-destructive testing robot.
  • the reactor pressure vessel simulation model is a three-dimensional virtual display of the reactor pressure vessel. Since the reactor pressure vessel is in a stationary state during the detection process, the reactor pressure vessel simulation model can be used as a usage scenario.
  • the reactor pressure vessel simulation model can realize model blanking by displaying and hiding parts, which is convenient for users to observe. In addition to these functions, the non-destructive testing robot simulation model needs to capture and calculate the motion of the motion axis to facilitate detection.
  • the virtual display of these simulation models requires a display algorithm.
  • the overall model can be constructed according to the physical model with reference to the physical model; while the display algorithm is modeled by computer graphics, and the rationality of the algorithm is verified by simulation experiments.
  • the essence of computer graphics is coordinate transformation. According to the rules of graphic transformation in computer graphics, the coordinates of the non-destructive detection robot are moved and the joints are rotated. Then the corresponding coordinate transformation is performed to obtain the joint variable parameter table, which is brought into the equation to obtain the joint changes. Regular, write the program.
  • the establishment of the coordinate system in computer graphics can be divided into the coordinate system pre-method and the coordinate system post-method.
  • the non-destructive detecting robot can be regarded as a set of connecting rods, and the connecting rods are connected by a moving pair. Connect the joints of the two links to constrain the relative motion between them.
  • the kinematic chain of the non-destructive testing robot consists of two types of links: the intermediate link and the terminal link. A terminal link is used as the base link, then the joint is connected, the joint is connected to the intermediate link, then the joint is connected, the intermediate link is articulated, and so on, and finally the other joint link is connected.
  • the connecting rod is numbered 0, 1, 2, ..., n from the base link to the end effector, and the motion pair connecting the i-1th link and the i-th link is recorded as the i-th joint, such that the machine
  • the body can be thought of as consisting of n+1 links and n joints.
  • the i-th coordinate system is established at the i-th joint, and the coordinate system Fi defined by the origin O i and the coordinate axes X i , Y i , Z i is fixed to the i-th link, and the coordinate system is as follows Rule definition:
  • 1Z i is the axis of the i-th joint.
  • the positive direction of the axis is made possible by two possibilities, and the Z i axis of the moving joint can be located at any position because the motion pair only defines its direction.
  • 2X i is defined as the common perpendicular of Z i-1 and Z i , with the direction pointing to the latter. If Z i-1 and Z i intersect, the positive direction of X i is uncertain and can be arbitrarily specified. In this case, it is prescribed that if the unit vectors i i , k i-1 and k i are fixed on X i , Z i-1 and Z i , respectively, i i is defined as k i-1 ⁇ k i . If Z i-1 and Z i are parallel, the position of X i is uncertain. To ensure that the definition is unique, X i is specified to pass through the origin of the i-1th coordinate system.
  • the distance between 3Z i and Z i+1 is defined as a i , which is a non-negative value.
  • the Z i of the intersection of 4Z i and X i+1 is labeled as b i . Since this amount is a coordinate, it can be positive or negative. Its absolute value is the distance between X i and X i+1 , called the offset between consecutive male vertical lines.
  • the angle between 5Z i and Z i+1 is defined as ⁇ i , and the positive direction of the measurement is defined with respect to X i+1 . This amount is referred to as the twist angle of a continuous pair of axes.
  • the angle between 6X i and X i+1 is defined as ⁇ i , and the positive direction of the measurement is defined with respect to Z i .
  • the global coordinates of the non-destructive testing robot are established as a whole.
  • Figure 5 shows the global coordinates of the established non-destructive testing robot.
  • the corresponding parameter table, as shown in Table 1, can further calculate the graph transformation equation.
  • the non-destructive testing robot When the non-destructive testing robot has eleven degrees of freedom motion axes, it is necessary to obtain eleven degrees of freedom encoder return values in real time.
  • Connected to the non-destructive testing robot through Industrial Ethernet read the return value of each encoder by the control card command, and obtain the linear axis motion distance and the rotation axis motion angle through conversion, and determine the 3D module coordinates and realize the graphics transformation algorithm.
  • the actual physical quantity represented by the encoder return value is solved, and the solution for driving the three-dimensional model is obtained by the algorithm solution, and the three-dimensional model transformation is driven to realize the synchronous movement of the robot and the actual non-destructive detection robot in the virtual reality.
  • the operator rotates and zooms the three-dimensional graphics through input devices such as a keyboard and a mouse, and the image does not have error images such as distortion, distortion, and disappearance.
  • error images such as distortion, distortion, and disappearance.
  • the specific indicators are as follows: 1. Translation error ⁇ 0.5mm, rotation angle error ⁇ 0.1 °; 2. Data acquisition rate: ⁇ 95%; 3. Dynamic information timely rate: ⁇ 95%; 4. Static information comprehensive rate: ⁇ 95% ; 5. Information accuracy rate: ⁇ 95%.
  • the non-destructive detection robot When the user sends the detection command, the non-destructive detection robot receives the user's instruction, and by analyzing the motion track of the instruction and the motion axis, it is transmitted to the motor driver of the non-destructive detection robot and the motor controlled by the motor driver, and the motor drives each motion axis to move along the motion track. Check to the target coordinates.
  • the target position to be detected may be determined, the installation position of the non-destructive detecting robot is determined, the spatial position coordinates of the target to be detected, and the motion trajectory of each motion axis of the non-destructive detecting robot have been determined, and the predetermined trajectory may be determined at this time.
  • the trajectory runs (it can be described as a positive solution).
  • the motion trajectory needs to be calculated according to the spatial position coordinates of the target to be detected and the coordinates of the end positions of the respective motion axes of the non-destructive detection robot through the above-described three-dimensional algorithm (that is, the inverse solution).
  • the motion trajectory has been determined in advance in the positive solution process, and is also calculated according to the spatial position coordinates of the target to be detected and the coordinates of the end positions of the respective motion axes of the non-destructive detection robot through the above-described three-dimensional algorithm.
  • each motion axis moves in accordance with the motion trajectory.
  • a scan of the motion axis is required.
  • Non-destructive testing of the probe is performed only if all motion axes are in place. Judging whether each motion axis is in motion can be performed by setting several path points on the motion track.
  • a pulse prompt signal is sent to the user through the control card that controls the motion of the motion axis to confirm that the motion axis moves to the determined position.
  • the upper computer is a PC, configuration: Dell Precision T3500, 1 Xeon W3503CPU processor, 16GB RAM, 2048GB hard drive, running on Windows XP SP3 operating system.
  • Non-destructive testing robot The range of travel of each degree of freedom motion axis can be preset.
  • the x-axis travel range is 0-1200mm
  • the y-axis travel range is 0-1000mm
  • the z-axis travel range is 0-600mm
  • the w-axis is the circumferential rotary axis, which can be rotated 360°.
  • Axial positioning accuracy ⁇ 0.05mm
  • repeated positioning accuracy 0.05mm The circumferential positioning accuracy is ⁇ 0.1°
  • the repeat positioning accuracy is ⁇ 0.05°.
  • the control card adopts Galil's four-axis motion control card, which can complete the double closed-loop (speed loop and position loop) servo control of four motion axes. After photoelectric isolation protection, it can receive real-time position signal and zero-point signal feedback.
  • the motion control card and the PC constitute a master-slave control structure: the PC is responsible for the management of the human-computer interaction interface and the real-time monitoring of the control system (such as keyboard and mouse management, system status display, motion trajectory planning, Control command transmission, external signal monitoring, etc.; control card completes all details of motion control (including pulse and direction signal output, automatic lifting speed processing, detection of origin and limit signals, etc.).
  • the built-in library functions include S-type, T-type acceleration, linear interpolation and circular interpolation, multi-axis linkage function, and so on.
  • the underlying software and hardware for motion control are integrated, with various speed and position control functions required for servo motor control. These functions can be easily called up by computer, simplifying the programming of control software.
  • the servo motor is driven by a servo drive.
  • the servo drive can be powered by a high-power PWM driver from MAXON for DC brushed servo motors.
  • the output power range is 10-250W, which meets the design needs. Because the high-frequency component of the PWM type driver has great interference to the encoder signal and the ultrasonic/eddy current signal, in order to suppress the electromagnetic interference, a filter circuit is designed on the hardware, and a partial filtering algorithm is added to the software.
  • Taiwan Mingwei power supply with high system reliability, mature technology and high cost performance. At present, such products have been widely used in various types of testing equipment.
  • the intelligent detection system and the intelligent detection method effectively simplify the detection process, improve the detection accuracy and speed, and reduce the irradiation dose level of the inspectors in service.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Plasma & Fusion (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Manipulator (AREA)
  • Monitoring And Testing Of Nuclear Reactors (AREA)

Abstract

一种基于虚拟现实技术的无损检测机器人智能检测方法，包括以下步骤：（1）将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；（2）将无损检测机器人各个自由度运动轴恢复到初始状态后，然后对各个自由度运动轴进行位置标定，构建全局坐标系和各个自由度运动轴的轴坐标系；（3）使仿真模型与实际设备建立对应关系；（4）无损检测机器人仿真模型根据实时获取的无损检测机器人各个自由度运动轴的位置姿态信息反馈值在三维虚拟环境中进行位置姿态变换，虚拟显示并控制无损检测机器人同步运动进行无损检测。该方法实现了全方位立体化的所见即所得的显示控制模式，大大提高了现场项目实施的安全性、提高了项目实施效率。

Description

基于虚拟现实技术的无损检测机器人智能检测方法 技术领域
本发明属于表面检测技术领域，具体涉及一种核电站反应堆压力容器内基于虚拟现实技术的无损检测机器人智能检测方法。
背景技术
目前，在核电检测领域，为了减少对检测人员和被检测对象的影响，一般采用无损检测。无损检测是利用物质的声、光、磁和电等特性，在不损害或不影响被检测对象使用性能的前提下，检测被检对象中是否存在缺陷或不均匀性，给出缺陷大小，位置，性质和数量等信息。无损检测主要有射线检验(RT)、超声检测(UT)、磁粉检测(MT)和液体渗透检测(PT)四种。其他无损检测方法有涡流检测(ET)、声发射检测(AT)、热像/红外(TIR)、泄漏试验(LT)、交流场测量技术(ACFMT)、漏磁检验(MFL)、远场测试检测方法(RFT)、超声波衍射时差法(TOFD)等。
无损检测与破坏性检测相比，具有以下特点：(1)非破坏性，因为它在做检测时不会损害被检测对象的使用性能；(2)全面性，由于检测是非破坏性，因此必要时可对被检测对象进行100％的全面检测，这是破坏性检测办不到的；(3)全程性，破坏性检测一般只适用于对原材料进行检测，如机械工程中普遍采用的拉伸、压缩、弯曲等，破坏性检验都是针对制造用原材料进行的，对于产成品和在用品，除非不准备让其继续服役，否则是不能进行破坏性检测的，而无损检测因不损坏被检测对象的使用性能。所以，它不仅可对制造用原材料，各中间工艺环节、直至最终产成品进行全程检测，也可对服役中的设备进行检测。
核电站反应堆压力容器内存在数量众多的不规则形状部件，目前对该类部件的表面超声/涡流检测多采用手动模式。现有国内外核电领域无损检测专业公司所开发并使用的反应堆压力容器无损检测机器人控制软件均采用无图形或二维图形化显示控制方法，该显示控制方法的弊端主要有以下几点：1、由于反应堆压力容器对于核电站安全运行的重要性，它的无损检测过程必须也是安全可控的，由于反应堆压力容器所处环境及反应堆压力容器自身 形状特点，检测机器人安装后各轴将处于人工监测及摄像头检测难以全面监测的环境中，无图行显示控制方式在对机器人自身姿态以及相对于压力容器位置关系不确定的情况下检查机器人的所有运动的安全性将很难以保证。2、二维图形显示控制虽可以在一个平面内用二维的图形显示直线运动轴与反应堆压力容器的位置关系，但却无法显示实际设备所有轴的姿态以及位置关系。为了使无损检测机器人到达目标检查位置，操作人员需要根据经验反复调整，过多的调整造成运动效率低下，增加了设备故障及误操作概率，降低了运动控制效率。现有控制技术在机器人各轴运动中，若要规避或达到某些特定位置时，需要操作人员根据经验进行人为不停调整，进行无损检测的效率低，不能在线操作和进行运动控制。本发明因此而来。
发明内容
本发明的目的是提出一种基于虚拟现实技术的无损检测机器人智能检测方法，解决了现有技术中无损检测机器人通过无图行显示控制或者二维图形控制进行无损检测，需要操作人员具有较高的技术水平以及较为丰富的调整经验进行反复调整和运动控制，无损检测效率低、检测需要根据经验频繁调整等技术问题。
为了解决现有技术中的这些问题，本发明提供的技术方案是：
一种基于虚拟现实技术的无损检测机器人智能检测方法，其特征在于在检测前需要预先采用虚拟现实技术构建无损检测机器人仿真模型以及反应堆压力容器仿真模型，所述方法包括以下步骤：
(1)将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；
(2)将无损检测机器人各个自由度运动轴恢复到初始状态后，然后对各个自由度运动轴进行位置标定，并确定一个自由度的运动轴末端位置为工作零点，构建全局坐标系和各个自由度运动轴的轴坐标系；
(3)使无损检测机器人仿真模型与被仿真的标定后无损检测机器人建立对应关系以及使反应堆压力容器仿真模型与被仿真的标定后反应堆压力容器之间建立对应关系；即使仿真模型与实际设备建立对应关系；
(4)无损检测机器人仿真模型根据实时获取的无损检测机器人各个自 由度运动轴的位置姿态信息反馈值在三维虚拟环境中进行位置姿态变换，虚拟显示并控制无损检测机器人同步运动进行无损检测。
优选的，所述方法中无损检测机器人仿真模型以及反应堆压力容器仿真模型通过三维建模软件构建，启动无损检测机器人仿真模型以及反应堆压力容器仿真模型后，以反应堆压力容器为应用场景开始标定无损检测机器人各个自由度运动轴到初始位置。
优选的，所述方法中使仿真模型与被仿真的硬件设备建立对应关系是按照如下步骤进行：
(1)将无损检测机器人各部件组装后进行运动控制功能、安全功能测试，确保硬件设备功能正常运行；
(2)将无损检测机器人安装至反应堆压力容器规定位置，然后对无损检测机器人各个自由度运动轴进行位置标定；
(3)各个自由度运动轴位置标定完成后，并确认标定准确后，启动无损检测机器人仿真模型、反应堆压力容器仿真模型进行同步处理，以反应堆压力容器仿真模型为应用场景，将无损检测机器人仿真模型各个自由度运动轴同步到标定后的位置坐标；
(4)同步完成后即建立了硬件设备与虚拟现实中仿真设备的姿态与位置的对应关系。
优选的，所述方法步骤(2)中进行各个自由度运动轴位置标定的步骤包括：
(1)预先对各个自由度运动轴的使能、速度、软件限位、控制器限位进行初始化设置；
(2)标定开始后，运动轴向负机械限位进行运动，时刻检测运动轴是否到达负机械限位；
(3)到达负机械限位后向正方向运动，根据各个自由度运动轴的特点和需要到达的预定位置确定运动量；
(4)标定结束，通过标定确定各个自由度运动轴的参数。
优选的，所述方法步骤(4)中控制无损检测机器人同步运动的具体步骤包括：
1)无损检测机器人各个自由度运动轴运动后按照预定周期向控制卡或 者控制盒发送无损检测机器人各个自由度运动轴的编码器反馈值；
2)控制卡或者控制盒读取无损检测机器人各个自由度运动轴的编码器返回值，换算得到无损检测机器人各个自由度运动轴中直线轴的运动距离及旋转轴的运动角度；
3)无损检测机器人仿真模型和反应堆压力容器仿真模型通过三维坐标的确定及图形变换算法，将编码器返回值所代表的实际物理量进行解算，获得驱动无损检测机器人仿真模型进行变换的解，驱动无损检测机器人仿真模型变换，实现虚拟现实中无损检测机器人仿真模型与实际无损检测机器人的同步运动。
优选的，所述方法中无损检测机器人设置有与核电站反应堆压力容器具体安装工位配合的安装臂，每个安装臂末端设置气动定心装置与核电站反应堆压力容器具体安装工位插接配合；所述安装臂间设置立柱，所述安装臂固定连接在立柱上，所述立柱下端连接若干个用来进行无损检测的机械臂；所述机械臂末端固定探头工具架。
优选的，所述方法中假设将无损检测机器人的机械臂视为2个终端连杆间连接若干个中间连杆的机构，连杆之间通过关节连接；假设从非执行器端到执行器端的连杆依次编号为0，1，2...，n，连接第i-1个连杆和第i个连杆的运动副记作第i各关节，在第i个关节处建立第i个坐标系，由原点Oi和坐标轴Xi，Yi，Zi定义的坐标系Fi被固结到第i-1个连杆上，则坐标系Fi定义如下：
1)Zi定义是第i个关节的轴线；轴线的正方向在一条直线上具有两种可能，移动关节的Zi轴位于任意位置，而关节定义了它的方向；
2)Xi定义为Zi-1和Zi的公垂线，方向是Zi-1指向Zi；如果Zi-1和Zi相交，Xi的正方向任意指定，如果单位矢量ii，ki-1和ki分别固定在Xi，Zi-1和Zi上，则ii被定义为ki-1×ki；如果Zi-1和Zi平行，规定Xi通过第i-1个坐标系的原点；
3)Zi和Zi+1之间的距离定义为ai，是非负值；
4)Zi和Xi+1的交点的Zi坐标定义bi；bi的绝对值是Xi和Xi+1之间的距离，称为连续公垂线之间的偏移；
5)Zi和Zi+1之间的夹角定义为αi，测量的正方向与Xi+1定义相关，称为连续一对轴的扭角。
6)Xi和Xi+1之间的夹角定义为θi，测量的正方向与Zi定义相关；
按照上述定义对无损检测机器人的各个机械臂构建整体的全局坐标系。
优选的，所述方法步骤(4)中进行位置姿态变换时，按照公式(I)：
rot(z，θi)*trans(0，0，di)*trans(ai，0，0)*rot(x，αi)  (I)；
计算无损检测机器人的运动轴末端姿态，得到末端姿态矩阵：
Figure PCTCN2014087770-appb-000001
其中
Figure PCTCN2014087770-appb-000002
Figure PCTCN2014087770-appb-000003
其中用来进行位置姿态变换时采用的图形变换参数表为：
Figure PCTCN2014087770-appb-000004
采用上述的图形变换参数表代入公式(II)即可得到各运动轴的末端姿态。
优选的，所述方法步骤(4)进行位置姿态变换时，如果无损检测机器人仿真模型与反应堆压力容器仿真模型存在干涉，则停止无损检测机器人各个运动轴的运动；否则继续运行检测。
优选的，所述方法中运动轴的运动可以通过预先确定运动轨迹，然后按 照运动轨迹进行运动实现智能检测。
具体的，运动轨迹智能检测方法包括以下步骤：
(1)将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；
(2)构建全局坐标系和轴坐标系，根据无损检测机器人在待检测的反应堆压力容器内的位置确定无损检测机器人各个运动轴上探头进行无损检测时各个运动轴在坐标系中的源坐标；
(3)根据待检测的反应堆压力容器内待检测对象的位置确定无损检测机器人各个运动轴上探头进行无损检测时各个运动轴在坐标系中的目标坐标；
(4)根据源坐标和目标坐标确定无损检测机器人各个运动轴的运动量和运动方式；由各个运动轴的运动量和运动方式确定无损检测机器人进行无损检测时的运行轨迹；
(5)无损检测机器人根据检测指令沿运行轨迹进行无损检测。
构建全局坐标系和轴坐标系前，需要将无损检测机器人各个运动轴恢复到初始状态后，并标定为一个自由度的运动轴位置为工作零点；以工作零点为零点坐标构建全局坐标系和轴坐标系，并确定其他运动轴的末端位置坐标作为源坐标。
本发明的另一目的在于提供一种无损检测机器人智能检测系统，其特征在于所述系统包括用于采用虚拟现实技术构建无损检测机器人仿真模型以及反应堆压力容器仿真模型，并用于将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；将无损检测机器人各个运动轴恢复到初始状态后，并标定一个自由度的运动轴末端位置，确定工作零点；并以工作零点为零点坐标构建全局坐标系和轴坐标系；使无损检测机器人仿真模型与被仿真的标定后无损检测机器人建立对应关系以及使反应堆压力容器仿真模型与被仿真的标定后反应堆压力容器之间建立对应关系的三维虚拟显示模块；和，
用于无损检测机器人仿真模型根据实时获取的无损检测机器人各个自由度运动轴的位置姿态信息反馈值在三维虚拟环境中进行位置姿态变换，虚拟显示并控制无损检测机器人进行无损检测的检测模块。
其中三维虚拟显示模块包括用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型在使用场景内进行标准视图操作和任意视角调整操作的视图操作子模块、用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型所处环境进行显示与隐藏操作的模型消隐模块、用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型的每一个零件的初始位姿数据，遍历模型装配体中所有零件当前位置，并将位置信息存入到数据库进行更新的初始位置设置模块、用于读取数据库中模型装配体中所有零件的初始位置信息，并将相应模型移动到初始位置的复位模块、用于以模型当前位置为基准，计算模型下一工作位置的动作捕捉开始模块、用于暂停模型运动的捕捉结束模块以及用于清除模型运动误差，实现仿真模型与实际物体位置统一的清除误差模块。
模型进行运动捕捉时，一旦模型发生干涉，发出警报并停住模型所有运动。
用户发送检测指令时，无损检测机器人接收用户的指令，并通过将指令和运动轴的运动轨迹解析，传输给无损检测机器人的电机驱动器及电机驱动器控制的电机，电机驱动各运动轴沿运动轨迹运动到目标坐标进行检测。
无损检测机器人欲达到某一特定位置，各个运动轴需要运动的量取决于当前位置与目标位置的关系，同时可能受到其他特定点或特定条件的影响，如果无损检测机器人设备没有控制算法解算各个运动轴的运动过程及运动轨迹规划，无法根据目标位置、当前位置、客观存在的约束条件及特定位置到达需求完成智能化的运动控制。
本发明通过虚拟现实技术实现三维实时显示控制检测的进行，通过全比例全尺寸的三维显示的方式反映反应堆压力容器无损检测机器人是一个自由度运动轴的姿态以及机器人与反应堆压力容器的相对位置关系，实现一种所见即所得的模拟实际现场工作环境及工况的显示控制方式，并可通过视图变换、部件的消隐、干涉检测对整个无损检查活动提供全方位立体化的显示和监控，对机器人所有检查动作及运动控制进行监控，起到极为重要的安全保障和显示控制功能。
本发明技术方案中无损检测机器人各个自由度运动轴上固定的表面超声检测探头或涡流检测探头。优选的技术方案是：所述无损检测机器人设置 有2～20个自由度的运动轴。优选的是所述无损检测机器人至少设置有2个运动轴；每个运动轴至少具有2个自由度。更为优选的是所述无损检测机器人设置有2个运动轴；每个运动轴具有4个自由度。
当所述无损检测机器人的运动轴具有4个自由度时，则4个自由度的运动轴依次为可在预定空间(三维空间坐标系)内进行精确移动的X轴、Y轴、Z轴以及用于在预定空间内进行360°范围内旋转运动的W轴。这些运动轴与伺服电机连接，通过控制卡或控制盒控制运动步进或旋转角度。无损检测机器人每个运动轴的运动可以是每个自由度的运动轴依次运动进行，也可以同时运动进行。本发明所述的依次运动进行是指当一个自由度的运动轴通过控制指令和相应控制参数运动到停止后，另一个自由度的运动轴开始运动。本发明所述的同时运动进行是指多个自由度的运动轴根据各自接收的控制指令和相应控制参数进行同时运动，其可以同时停止，也可以不同时停止；这根据多个自由度的运动轴各自运动的路径决定。
具体的，X轴运动为X轴在核电站反应堆压力容器内进行水平方向上的直线运动；Y轴运动为Y轴在三维空间坐标系(预定核电站反应堆压力容器空间)内进行水平方向上的直线运动，且与X轴运动方向垂直；Z轴运动为Z轴在三维空间坐标系(预定核电站反应堆压力容器空间)内进行竖直方向上的直线运动，其与X轴运动方向、X轴运动方向构成的平面垂直；W轴运动为W轴在三维空间坐标系(预定核电站反应堆压力容器空间)内进行360°范围内旋转运动。
本发明技术方案获得运动轨迹后，由运动轨迹形成各个自由度运动轴的控制指令和相应的控制参数的具体方法不再详述。商业上有现成的可编程的控制卡或控制盒可供参考。将相应的运动轨迹按照核电站相应的操作规程进行编程后，写入控制卡或者控制盒可以形成各个自由度运动轴的控制指令和相应的控制参数。这些控制卡或者控制盒可以控制各个自由度运动轴的伺服电机进行相应的运动处理。
相对于现有技术中的方案，本发明的优点是：
与现有手动扫查模式相比，本发明的无损检测机器人智能检测系统及智能检测方法有效简化了检测流程、提高了检测精度及速度、减少了在役是检查人员辐照剂量水平。本发明技术方案中通过三维虚拟环境的建立、三维机 器人以及被检查部件的建模、机器人全局坐标以及运动自由度坐标系的建立、图形变换算法的计算并驱动图形变换的机制和显示控制方式，将三维虚拟现实技术应用到反应堆压力容器无损检测机器人进行无损检测项目实施中，实现了全方位立体化的所见即所得的显示控制模式，大大提高了现场项目实施的安全性、提高了项目实施效率、提高了机器人控制效率及控制性能、减少了不必要的操作并降低了人因失误发生概率，提高了机器人控制系统整体性能稳定性、安全性和操作性。
附图说明
下面结合附图及实施例对本发明作进一步描述：
图1为无损检测机器人智能检测系统的结构框图；
图2为无损检测机器人智能检测系统进行检测作业时的连接示意图；
图3为无损检测机器人的结构示意图；
图4为无损检测机器人智能检测系统进行检测作业时在三维环境中进行模拟现实的流程图；
图5为无损检测机器人仿真模型与实际的无损检测机器人进行坐标转换的原理图；
图6为无损检测机器人智能检测系统进行检测作业的具体实际操作流程图；
图7为无损检测机器人仿真模型与实际的无损检测机器人进行同步运动的流程图
图8为实际的无损检测机器人的标定流程图；
图9为无损检测机器人仿真模型与实际的无损检测机器人运动时反馈与坐标转换过程示意图。
图10为无损检测机器人智能检测系统进行检测作业的整体流程图。
具体实施方式
以下结合具体实施例对上述方案做进一步说明。应理解，这些实施例是用于说明本发明而不限于限制本发明的范围。实施例中采用的实施条件可以根据具体厂家的条件做进一步调整，未注明的实施条件通常为常规实验中的 条件。
实施例
如图1所示，该无损检测机器人智能检测系统，包括用于采用虚拟现实技术构建无损检测机器人仿真模型以及反应堆压力容器仿真模型，并用于将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；将无损检测机器人各个运动轴恢复到初始状态后，并标定一个自由度的运动轴末端位置，确定工作零点；并以工作零点为零点坐标构建全局坐标系和轴坐标系；使无损检测机器人仿真模型与被仿真的标定后无损检测机器人建立对应关系以及使反应堆压力容器仿真模型与被仿真的标定后反应堆压力容器之间建立对应关系的三维虚拟显示模块；用于无损检测机器人仿真模型根据实时获取的无损检测机器人各个自由度运动轴的位置姿态信息反馈值在三维虚拟环境中进行位置姿态变换，虚拟显示并控制无损检测机器人进行无损检测的检测模块。
其中三维虚拟显示模块包括用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型在使用场景内进行标准视图操作和任意视角调整操作的视图操作子模块、用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型所处环境进行显示与隐藏操作的模型消隐模块、用于对无损检测机器人仿真模型以及反应堆压力容器仿真模型的每一个零件的初始位姿数据，遍历模型装配体中所有零件当前位置，并将位置信息存入到数据库进行更新的初始位置设置模块、用于读取数据库中模型装配体中所有零件的初始位置信息，并将相应模型移动到初始位置的复位模块、用于以模型当前位置为基准，计算模型下一工作位置的动作捕捉开始模块、用于暂停模型运动的捕捉结束模块以及用于清除模型运动误差，实现仿真模型与实际物体位置统一的清除误差模块。
如图2所示，该无损检测机器人智能检测系统安装在上位机上，并与控制卡连接控制无损检测机器人。所述无损检测机器人每个运动轴上固定的表面超声检测探头或涡流检测探头。所述无损检测机器人设置有个11个运动轴，其中至少有2个运动轴设置有4个自由度。4个自由度依次为可在预定空间内进行精确移动的X轴运动、Y轴运动、Z轴运动以及用于在预定空间内进行360°范围内旋转运动的W轴运动。这些运动轴与伺服电机(可以是 单一的伺服电机，也可以是多个伺服电机进行运动轴驱动)连接，通过控制卡或控制盒控制运动步进或旋转角度。
其中X轴运动为X轴在核电站反应堆压力容器内进行水平方向上的直线运动；Y轴运动为Y轴在核电站反应堆压力容器内进行水平方向上的直线运动，且与X轴运动方向垂直；Z轴运动为Z轴在核电站反应堆压力容器内进行竖直方向上的直线运动，其与X轴运动方向、X轴运动方向构成的平面垂直W轴运动为W轴在核电站反应堆压力容器内进行360°范围内旋转运动。运动原理过程将在获得运动轨迹的过程具体描述。
所述无损检测机器人如图3所示，设置有与核电站反应堆压力容器具体安装工位配合的安装臂1(具有三个，用来支撑无损检测机器人并配合安装在安装工位上)，每个安装臂末端设置气动定心装置2与核电站反应堆压力容器具体安装工位插接配合。所述安装臂与无损检测机器人的本体——立柱3固定连接，所述立柱3下端根据核电站反应堆压力容器内待检测部件的情况设置若干个用来进行无损检测的机械臂4。根据核电站反应堆压力容器内待检测部件的情况，可以设置法兰扫查机械臂、管嘴扫查机械臂、下封头扫查机械臂等；每个机械臂上设置若干个自由度的运动轴。运动轴末端位置根据实际情况可以设置前端旋转副5，所述前端旋转副上固定连接检测探头的探头工具架6，通过运动轴输送探头到目标位置。
如图10所示，进行核电站反应堆压力容器内无损检测时，具体按照如下步骤进行：
1、将反应堆压力容器无损检测机器人按照规定安装位置安装于待检反应堆压力容器上。
具体安装过程不再详述。核电站反应堆压力容器内预设很多安装工位，核电站操作人员根据安装导向柱的提示将无损检测机器人安装到预定工位即可。
2、对反应堆压力容器无损检测机器人的是一个自由度运动轴进行位置标定，确定唯一个工作零点及全局坐标写及轴坐标系。
通常先将无损检测机器人各个运动轴恢复到初始状态，并标定为一个自由度的运动轴位置为工作零点。如以立柱的垂直自由度的运动轴的末端位置标定为零点坐标(0，0，0)。将无损检测机器人各个自由度运动轴恢复到初始 状态后，然后对各个自由度运动轴进行位置标定，并确定一个自由度的运动轴末端位置为工作零点，构建全局坐标系和各个自由度运动轴的轴坐标系。
如图8所示，进行各个自由度运动轴位置标定的步骤包括：
1)预先对各个自由度运动轴的使能、速度、软件限位、控制器限位进行初始化设置；
2)标定开始后，运动轴向负机械限位进行运动，时刻检测运动轴是否到达负机械限位；
3)到达负机械限位后向正方向运动，根据各个自由度运动轴的特点和需要到达的预定位置确定运动量；
4)标定结束，通过标定确定各个自由度运动轴的参数。
当将检测系统构建形成软件时，先对运动轴使能、速度、软件限位、控制器限位进行初始化设置；然后采用软件开始标定，标定程序下达控制卡，运动轴向负机械限位进行运动，时刻检测运动轴是否到达负机械限位；到达负机械限位后向正方向运动，运动量取决于各轴特点和所要到达的预定位置，标定动作完成。
3、连接设备，启动三维显示控制界面，经过标定的机器人与三维虚拟现实环境中的无损检测机器人以及反应堆压力容器即建立了对应关系。
如图7所示，所述方法中使仿真模型与被仿真的硬件设备建立对应关系是按照如下步骤进行：
1)将无损检测机器人各部件组装后进行运动控制功能、安全功能测试，确保硬件设备功能正常运行；
2)将无损检测机器人安装至反应堆压力容器规定位置，然后对无损检测机器人各个自由度运动轴进行位置标定；
3)各个自由度运动轴位置标定完成后，并确认标定准确后，启动无损检测机器人仿真模型、反应堆压力容器仿真模型进行同步处理，以反应堆压力容器仿真模型为应用场景，将无损检测机器人仿真模型各个自由度运动轴同步到标定后的位置坐标；
4)同步完成后即建立了硬件设备与虚拟现实中仿真设备的姿态与位置的对应关系。
其中无损检测机器人组装后进行基本运动控制功能测试和安全功能测 试。该系统形成软件后，通过软件控制无损检测机器人十一个自由度进行位置标定，确认位置标定完成且位置标定正确后打开三维虚拟显示界面进行同步，同步完成后即建立了硬件设备与虚拟现实中仿真设备的姿态与位置的对应关系。
如图4和图6所示，此时无损检测机器人仿真模型以及反应堆压力容器仿真模型已经构建。其构建方法可以采用SolidWorks软件作为显示软件，在其基础上进行二次开发，通过SolidWorks软件提供的API调用三维软件本身的功能，直接在已有功能的基础上开发特定功能的软件模块。如采用Visual C#软件进行二次开发，在SolidWorks中编写插件程序，使用插件程序去控制SolidWorks软件中的三维模型。
针对无损检测机器人和反应堆压力容器的具体结构构造连接关系，利用SolidWorks软件提供的API实现各种无损检测机器人和反应堆压力容器的仿真模型构建，并提供三维实时显示界面，用来显示与现场条件相一致的操作环境，实时显示无损检测机器人在反应堆压力容器内运行情况，能够对虚拟模型进行标准视图和任意角度视图的调整，图形不出现失真，变形。并且为方便模型表达，增加对模型所处环境进行显示与隐藏操作。这些构成三维虚拟显示模块的子模块。
视图操作子模块可以实现对虚拟模型进行标准视图和任意角度视图的调整，图形不出现失真，变形。可以对虚拟模型进行任意角度的调整，在捕捉过程中实时的放大缩小模型，调整视图主要分为两种调节模式。第一种是任意视角调整模式可以对任意角度进行调整，输入要旋转的角度，通过点击相应的调整按钮来实现任意角度的模型角度调整；第二种是标准视图操作模式，通过调用SolidWorks API函数，在编程界面上添加按钮，实现虚拟模型正视，下视，右视等这些标准视图操作。
模型消隐模块用于方便模型表达，通过点击相应的显示与隐藏按钮，对核反应堆压力容器仿真模型进行分块的显示与隐藏。
4、控制软件控制机器人运动，并实时获取机器人各轴编码器反馈值，该反馈值代表了机器人各自由度运动轴的位置姿态信息，三维模型经过全局坐标系、运动轴自由度坐标系、机器人运动模型等要素的解算，将实际运动值及机器人现姿态进行计算并驱动三维模型进行矩阵计算以此完成机器人 模型相对反应堆压力容器模型在三维虚拟环境中进行准确变换，以完成反映机器人及被检测设备的虚拟显示控制的功能。
如图9所示，控制无损检测机器人同步运动的具体步骤包括：
1)无损检测机器人各个自由度运动轴运动后按照预定周期向控制卡或者控制盒发送无损检测机器人各个自由度运动轴的编码器反馈值；
2)控制卡或者控制盒读取无损检测机器人各个自由度运动轴的编码器返回值，换算得到无损检测机器人各个自由度运动轴中直线轴的运动距离及旋转轴的运动角度；
3)无损检测机器人仿真模型和反应堆压力容器仿真模型通过三维坐标的确定及图形变换算法，将编码器返回值所代表的实际物理量进行解算，获得驱动无损检测机器人仿真模型进行变换的解，驱动无损检测机器人仿真模型变换，实现虚拟现实中无损检测机器人仿真模型与实际无损检测机器人的同步运动。
其中操作人员可以根据操作习惯、操作机器人所运动的轨迹、监察规程等情况对三维虚拟现实界面进行视图变换、部件消隐、干涉检测等功能。
无损检测机器人仿真模型是用来对无损检测机器人的运动过程进行的三维虚拟显示。反应堆压力容器仿真模型是反应堆压力容器的进行的三维虚拟显示，由于在检测过程中反应堆压力容器为静止状态，因此可以将反应堆压力容器仿真模型作为使用场景。反应堆压力容器仿真模型通过显示与隐藏零件可以实现模型消隐，方便用户使用时观察。而无损检测机器人仿真模型除了这些功能，还需要对运动轴的动作进行捕捉和计算，以方便进行检测。
这些仿真模型进行虚拟显示需要采用显示算法。为了保证仿真和数据传输的实时性，需要整体模型的准确构建和显示算法的合理。其中整体模型可以参照实物模型按照预定比例进行构建；而显示算法则采用计算机图形学进行建模，经过仿真实验验证了算法的合理性。
计算机图形学的本质是坐标变换，按照计算机图形学中图形变换的规则在无损检测机器人的移动及转动关节建立坐标，然后进行相应的坐标变换，得到关节变量参数表，带入方程得到各个关节变化规律，写入程序即可。
根据坐标系在关节建立位置的不同，计算机图形学中坐标系的建立通常可以分为坐标系前置法和坐标系后置法。
本发明技术方案中采用坐标系前置法，在本发明中无损检测机器人可以看作是一组连杆的集合，连杆之间通过运动副相连接。连接两个连杆的关节，约束它们之间的相对运动。无损检测机器人的运动链由两类连杆组成：中间连杆和终端连杆。将一个终端连杆作为基础连杆，然后连接关节，关节再连接中间连杆，然后连接关节，关节连接中间连杆，以此类推，最后关节连接另外一个终端连杆。连杆从基础连杆到末端执行器依次编号为0，1，2...，n，连接第i-1个连杆和第i个连杆的运动副记作第i各关节，这样机械本体可以看作是由n+1个连杆和n个关节组成。在第i个关节处建立第i个坐标系，由原点Oi和坐标轴Xi，Yi，Zi定义的坐标系Fi被固结到第i-1个连杆上，坐标系根据如下规则定义：
1Zi是第i个关节的轴线。轴线的正方向由两种可能，而且，移动关节的Zi轴可以位于任意位置，因为该运动副只定义了它的方向。
2Xi定义为Zi-1和Zi的公垂线，方向是前者指向后者。如果Zi-1和Zi相交，Xi的正方向就不确定，可任意指定。在这种情况下规定：如果单位矢量ii，ki-1和ki分别固定在Xi，Zi-1和Zi上，则ii被定义为ki-1×ki。如果Zi-1和Zi平行，Xi的位置就不确定。为保证定义唯一，规定Xi通过第i-1个坐标系的原点。
3Zi和Zi+1之间的距离定义为ai，是非负值。
4Zi和Xi+1的交点的Zi坐标记作bi。由于这个量是坐标，可正可负。它的绝对值是Xi和Xi+1之间的距离，称为连续公垂线之间的偏移。
5Zi和Zi+1之间的夹角定义为αi，测量的正方向关于Xi+1定义。这个量被称为连续一对轴的扭角。
6Xi和Xi+1之间的夹角定义为θi，测量的正方向关于Zi定义。
按照这种规则，对无损检测机器人进行整体建立全局坐标。如图5所示为建立的无损检测机器人全局坐标示意图。其相应的参数表，如表1所示，可以进而推算出图形变换方程。
表1 图形变换参数表
Figure PCTCN2014087770-appb-000005
Figure PCTCN2014087770-appb-000006
这些参数表经过验证，这种建立坐标方法及根据坐标系确定的参数表是正确的。
为计算机器人末端姿态，将图形变换参数表代入末端姿态计算公式(I)：
rot(z，θi)*trans(0，0，di)*trans(ai，0，0)*rot(x，αi)  (I)；
计算无损检测机器人的运动轴末端姿态，得到末端姿态矩阵：
Figure PCTCN2014087770-appb-000007
其中
Figure PCTCN2014087770-appb-000008
Figure PCTCN2014087770-appb-000009
以表中第二行参数表为例，里面对应参数θi＝90，di＝5.773，ai＝0，αi＝90。带入计算公式中得到末端姿态矩阵
Figure PCTCN2014087770-appb-000010
当无损检测机器人存在十一个自由度运动轴，需要实时获取十一个自由度的编码器返回值。通过工业以太网与无损检测机器人进行连接，通过控制卡指令读取各自由度编码器返回值，通过换算得到直线轴运动距离及旋转轴运动角度，通过三维模块坐标的确定及图形变换算法的实现，将编码器返回值所代表的实际物理量进行解算，通过算法解算后获得驱动三维模型进行变换的解，驱动三维模型变换，实现虚拟现实中机器人与实际无损检测机器人的同步运动。
在三维模型显示过程中，操作人员通过键盘、鼠标等输入设备对三维图形进行旋转、缩放操作，图形不出现失真、变形、消失等错误画面。保证三维模型精度，确保软件仿真中反应堆压力容器结构、机器人运动结果精度。具体指标如下：1.平移误差≤0.5mm，转动角度误差≤0.1°；2.数据采集率：≥95％；3.动态信息及时率：≥95％；4.静态信息全面率：≥95％；5.信息准确率：≥95％。
从时间特性上来讲，为了保证软件仿真和数据传输的实时性，必须保证：1.通讯响应时间≤0.5秒；2.实时信息刷新周期≤1秒；3.软件持续运行时间≥7天。
用户发送检测指令时，无损检测机器人接收用户的指令，并通过将指令和运动轴的运动轨迹解析，传输给无损检测机器人的电机驱动器及电机驱动器控制的电机，电机驱动各运动轴沿运动轨迹运动到目标坐标进行检测。
在实际无损检测过程中，可能待检测的目标位置确定，无损检测机器人的安装位置确定，待检测的目标的空间位置坐标、无损检测机器人各个运动轴的运动轨迹已经确定，此时可以按照预定的轨迹进行运行(可谓正解)。而一般情况下，运动轨迹需要根据待检测的目标的空间位置坐标、无损检测机器人各个运动轴末端位置坐标通过上述的三维空间算法计算获得(谓之逆解)。实际上，正解过程中已经预先确定运动轨迹也是根据待检测的目标的空间位置坐标、无损检测机器人各个运动轴末端位置坐标通过上述的三维空间算法计算获得的。
进行无损检测时，各个运动轴按照运动轨迹进行运动。需要对运动轴的运动情况进行扫查。当且仅当所有运动轴均运动到位后，才进行探头的无损检测。判断各个运动轴是否运动到位，可以通过在运动轨迹上设置若干个路径点进行。当某个运动轴到达运动轨迹的路径点时，通过控制运动轴运动的控制卡向用户发送脉冲提示信号，确认运动轴运动到确定位置。
其中上位机为PC机，配置：戴尔Precision T3500，1个Xeon W3503CPU处理器，16GB内存，2048GB硬盘，运行于Windows XP SP3操作系统。无损检测机器人各个自由度运动轴的行程范围可以预先设置。x轴行程范围0-1200mm，y轴行程范围0-1000mm，z轴行程范围0-600mm，w轴为周向旋转轴，可做360°周向旋转。轴向定位精度±0.05mm，重复定位精度正负 0.05mm。周向定位精度±0.1°，重复定位精度±0.05°。
控制卡采用Galil公司的四轴运动控制卡，可完成4个运动轴的双闭环(速度环和位置环)伺服控制，经过光电隔离保护后可接受实时位置信号及零点信号反馈。通常运动控制卡与PC机构成主从式控制结构：PC机负责人机交互界面的管理和控制系统的实时监控等方面的工作(例如键盘和鼠标的管理、系统状态的显示、运动轨迹规划、控制指令的发送、外部信号的监控等等)；控制卡完成运动控制的所有细节(包括脉冲和方向信号的输出、自动升降速的处理、原点和限位等信号的检测等等)。自带库函数包括S型、T型加速，直线插补和圆弧插补，多轴联动函数等。实现运动控制的底层软件和硬件集成在一起，具有伺服电机控制所需的各种速度、位置控制功能，这些功能通过计算机可以方便地调用，简化了控制软件的编程。
伺服电机通过伺服驱动器进行驱动。伺服驱动器可以采用MAXON公司生产的大功率PWM驱动器，用于DC有刷伺服电机的驱动。输出功率范围10-250W，满足设计需要。由于是PWM型驱动器其高频分量对编码器信号、超声/涡流信号有极大的干扰，为了抑制电磁干扰，在硬件上设计了滤波电路，在软件上增加了部分滤波算法。
伺服电机的电源驱动采用系统可靠性高，技术成熟、性价比高的台湾明纬电源。目前该类产品已广泛应用于各类检测设备中。
经在测试环境下进行扫查，其误差能控制在≤1％以下，检测精度极高；由于可以通过pc机进行控制，实现了远端网络控制检测，避免了人身辐射剂量的累积。与现有手动模式相比，该智能检测系统及智能检测方法有效简化了检测流程、提高了检测精度及速度、减少了在役是检查人员辐照剂量水平。
上述实例只为说明本发明的技术构思及特点，其目的在于让熟悉此项技术的人是能够了解本发明的内容并据以实施，并不能以此限制本发明的保护范围。凡根据本发明精神实质所做的等效变换或修饰，都应涵盖在本发明的保护范围之内。

Claims (9)

  1. 一种基于虚拟现实技术的无损检测机器人智能检测方法，其特征在于在检测前需要预先采用虚拟现实技术构建无损检测机器人仿真模型以及反应堆压力容器仿真模型，所述方法包括以下步骤：
    (1)将进行无损检测的无损检测机器人安装到待检测的反应堆压力容器内的预定位置；
    (2)将无损检测机器人各个自由度运动轴恢复到初始状态后，然后对各个自由度运动轴进行位置标定，构建全局坐标系和各个自由度运动轴的轴坐标系；
    (3)使无损检测机器人仿真模型与被仿真的标定后无损检测机器人建立对应关系以及使反应堆压力容器仿真模型与被仿真的标定后反应堆压力容器之间建立对应关系；
    (4)无损检测机器人仿真模型根据实时获取的无损检测机器人各个自由度运动轴的位置姿态信息反馈值在三维虚拟环境中进行位置姿态变换，虚拟显示并控制无损检测机器人同步运动进行无损检测。
  2. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法中无损检测机器人仿真模型以及反应堆压力容器仿真模型通过三维建模软件构建，启动无损检测机器人仿真模型以及反应堆压力容器仿真模型后，以反应堆压力容器为应用场景开始位置标定无损检测机器人各个自由度运动轴。
  3. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法中使仿真模型与被仿真的硬件设备建立对应关系是按照如下步骤进行：
    1)将无损检测机器人各部件组装后进行运动控制功能、安全功能测试，确保硬件设备功能正常运行；
    2)将无损检测机器人安装至反应堆压力容器规定位置，然后对无损检测机器人各个自由度运动轴进行位置标定；
    3)各个自由度运动轴位置标定完成后，并确认标定准确后，启动无损检测机器人仿真模型、反应堆压力容器仿真模型进行同步处理，以反应堆压力容器仿真模型为应用场景，将无损检测机器人仿真模型各个自由度运动轴同步到标定后的位置坐标；同步完成后即建立了硬件设备与虚拟现实中仿真 设备的姿态与位置的对应关系。
  4. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法步骤(2)中进行各个自由度运动轴位置标定的步骤包括：
    1)预先对各个自由度运动轴的使能、速度、软件限位、控制器限位进行初始化设置；
    2)标定开始后，运动轴向负机械限位进行运动，时刻检测运动轴是否到达负机械限位；
    3)到达负机械限位后向正方向运动，根据各个自由度运动轴的特点和需要到达的预定位置确定运动量；
    4)标定结束，通过标定确定各个自由度运动轴的参数。
  5. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法步骤(4)中进行同步控制无损检测机器人进行无损检测的具体步骤包括：
    1)无损检测机器人各个自由度运动轴运动后按照预定周期向控制卡或者控制盒发送无损检测机器人各个自由度运动轴的编码器反馈值；
    2)控制卡或者控制盒读取无损检测机器人各个自由度运动轴的编码器返回值，换算得到无损检测机器人各个自由度运动轴中直线轴的运动距离及旋转轴的运动角度；
    3)无损检测机器人仿真模型和反应堆压力容器仿真模型通过三维坐标的确定及图形变换算法，将编码器返回值所代表的实际物理量进行解算，获得驱动无损检测机器人仿真模型进行变换的解，驱动无损检测机器人仿真模型变换，实现虚拟现实中无损检测机器人仿真模型与实际无损检测机器人的同步运动。
  6. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法中无损检测机器人设置有与核电站反应堆压力容器具体安装工位配合的安装臂，每个安装臂末端设置气动定心装置与核电站反应堆压力容器具体安装工位插接配合；所述安装臂间设置立柱，所述安装臂固定连接在立柱上，所述立柱下端连接若干个用来进行无损检测的机械臂；所述机械臂末端固定探头工具架。
  7. 根据权利要求6所述的无损检测机器人智能检测方法，其特征在于 所述方法中假设将无损检测机器人的机械臂视为2个终端连杆间连接若干个中间连杆的机构，连杆之间通过关节连接；假设从非执行器端到执行器端的连杆依次编号为0，1，2...，n，连接第i-1个连杆和第i个连杆的运动副记作第i各关节，在第i个关节处建立第i个坐标系，由原点Oi和坐标轴Xi，Yi，Zi定义的坐标系Fi被固结到第i-1个连杆上，则坐标系Fi定义如下：
    1)Zi定义是第i个关节的轴线；轴线的正方向在一条直线上具有两种可能，移动关节的Zi轴位于任意位置，而关节定义了它的方向；
    2)Xi定义为Zi-1和Zi的公垂线，方向是Zi-1指向Zi；如果Zi-1和Zi相交，Xi的正方向任意指定，如果单位矢量ii，ki-1和ki分别固定在Xi，Zi-1和Zi上，则ii被定义为ki-1×ki；如果Zi-1和Zi平行，规定Xi通过第i-1个坐标系的原点；
    3)Zi和Zi+1之间的距离定义为ai，是非负值；
    4)Zi和Xi+1的交点的Zi坐标定义bi；bi的绝对值是Xi和Xi+1之间的距离，
    称为连续公垂线之间的偏移；
    5)Zi和Zi+1之间的夹角定义为αi，测量的正方向与Xi+1定义相关，称为连续一对轴的扭角。
    6)Xi和Xi+1之间的夹角定义为θi，测量的正方向与Zi定义相关；
    按照上述定义对无损检测机器人的各个机械臂构建整体的全局坐标系。
  8. 根据权利要求7所述的无损检测机器人智能检测方法，其特征在于所述方法步骤(4)中进行位置姿态变换时，按照公式(I)：
    rot(z，θi)*trans(0，0，di)*trans(ai，0，0)*rot(x，αi)  (I)；
    计算无损检测机器人的运动轴末端姿态，得到末端姿态矩阵：
    Figure PCTCN2014087770-appb-100001
    其中
    Figure PCTCN2014087770-appb-100002
    Figure PCTCN2014087770-appb-100003
    其中用来进行位置姿态变换时采用的图形变换参数表为：
    Figure PCTCN2014087770-appb-100004
    采用上述的图形变换参数表代入公式(II)即可得到各运动轴的末端姿态。
  9. 根据权利要求1所述的无损检测机器人智能检测方法，其特征在于所述方法步骤(4)进行位置姿态变换时，如果无损检测机器人仿真模型与反应堆压力容器仿真模型存在干涉，则停止无损检测机器人各个运动轴的运动；否则继续运行检测。
PCT/CN2014/087770 2013-10-18 2014-09-29 基于虚拟现实技术的无损检测机器人智能检测方法 WO2015055085A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
BR112016008313A BR112016008313B8 (pt) 2013-10-18 2014-09-29 Método de teste inteligente de um robô de teste não destrutivo em um recipiente de pressão de reator nuclear com base em tecnologia de realidade virtual
EP14854066.9A EP3059738B1 (en) 2013-10-18 2014-09-29 Intelligent testing method of nondestructive robot testing based on virtual reality technology

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310493138.5 2013-10-18
CN201310493138.5A CN103761996B (zh) 2013-10-18 2013-10-18 基于虚拟现实技术的无损检测机器人智能检测方法

Publications (1)

Publication Number Publication Date
WO2015055085A1 true WO2015055085A1 (zh) 2015-04-23

Family

ID=50529220

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/087770 WO2015055085A1 (zh) 2013-10-18 2014-09-29 基于虚拟现实技术的无损检测机器人智能检测方法

Country Status (4)

Country Link
EP (1) EP3059738B1 (zh)
CN (1) CN103761996B (zh)
BR (1) BR112016008313B8 (zh)
WO (1) WO2015055085A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107756376A (zh) * 2017-10-18 2018-03-06 江西制造职业技术学院 一种应用人形机器人的空间站维修与维护方法
CN109483601A (zh) * 2018-12-24 2019-03-19 合肥欣奕华智能机器有限公司 一种工业机器人功能测试系统及测试方法
CN109531566A (zh) * 2018-11-16 2019-03-29 国网江苏省电力有限公司盐城供电分公司 一种基于虚拟现实系统的机器人带电作业控制方法
CN110704983A (zh) * 2019-10-12 2020-01-17 中国铁路设计集团有限公司 一种参数驱动下关节联动的吊车动态作业仿真方法
CN111883271A (zh) * 2020-06-03 2020-11-03 湖北工业大学 核反应堆压力容器自动检测平台精确定位方法及系统
CN113359987A (zh) * 2021-06-03 2021-09-07 煤炭科学技术研究院有限公司 基于vr虚拟现实的半实物综采实操平台
CN114367975A (zh) * 2021-11-15 2022-04-19 上海应用技术大学 串联工业机器人控制算法的验证系统
CN114519759A (zh) * 2022-02-14 2022-05-20 同恩(上海)工程技术有限公司 一种视图变换方法、装置及介质
CN114619436A (zh) * 2020-12-08 2022-06-14 山东新松工业软件研究院股份有限公司 一种基于EtherCAT的六轴机器人控制系统测试设备及其方法
WO2024060141A1 (zh) * 2022-09-22 2024-03-28 宁德时代新能源科技股份有限公司 一种修改运动副参数的方法、装置及产线系统

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103985424B (zh) 2014-05-29 2017-02-15 中广核检测技术有限公司 核反应堆压力容器无损检测机器人及其检测方法
CN103995478B (zh) * 2014-05-30 2016-05-18 山东建筑大学 基于现实虚拟互动的模块化液压机械臂实验平台及方法
CN105810078B (zh) * 2014-12-31 2018-08-24 国核电站运行服务技术有限公司 一种用于核反应堆压力容器的检测仿真系统
CN106997175A (zh) * 2016-10-21 2017-08-01 遨博（北京）智能科技有限公司 一种机器人仿真控制方法及装置
US11436811B2 (en) 2017-04-25 2022-09-06 Microsoft Technology Licensing, Llc Container-based virtual camera rotation
JP6538760B2 (ja) * 2017-06-22 2019-07-03 ファナック株式会社 複合現実シミュレーション装置及び複合現実シミュレーションプログラム
CN107817695B (zh) * 2017-09-28 2022-02-18 广州明珞汽车装备有限公司 一种批量检测与设置机器人仿真软限位的方法及系统
CN109958837A (zh) * 2017-12-14 2019-07-02 湘潭宏远电子科技有限公司 一种管道机器人控制装置
CN111351849A (zh) * 2018-12-20 2020-06-30 核动力运行研究所 一种超声波信号自动采集与协同控制系统及方法
CN109702746A (zh) * 2019-01-18 2019-05-03 弗徕威智能机器人科技（上海）有限公司 一种运动和位置双向同步方法
CN110559083B (zh) * 2019-09-10 2020-08-25 深圳市精锋医疗科技有限公司 手术机器人及其末端器械的控制方法、控制装置
CN111300412A (zh) * 2020-02-28 2020-06-19 华南理工大学 一种基于虚幻引擎的控制机器人的方法
CN111251305B (zh) * 2020-03-13 2023-02-07 南方科技大学 机器人力控制方法、装置、系统、机器人及存储介质
CN112066905B (zh) * 2020-07-03 2021-09-03 河南省锅炉压力容器安全检测研究院 承压类特种设备外表面形变量检测工具和建模方法
CN112432999B (zh) * 2020-10-30 2024-03-29 中广核检测技术有限公司 一种超声检查机器人扫查定位方法
CN113126568B (zh) * 2021-03-10 2022-08-09 上海乾庾智能科技有限公司 一种基于增强现实技术的工业机器人操作和演示系统
CN113001142B (zh) * 2021-03-18 2023-02-03 北京空间机电研究所 一种用于大型分块光学组件的双机械臂自动装配系统
CN113295732B (zh) * 2021-04-22 2022-09-30 杭州申昊科技股份有限公司 可供检测管道缺陷的管道机器人及其控制方法与控制系统
CN115206079A (zh) * 2022-05-16 2022-10-18 中国第一汽车股份有限公司 一种无损探伤检测的高效率调整系统及方法
CN114974506B (zh) * 2022-05-17 2024-05-03 重庆大学 人体姿态数据处理方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04269185A (ja) * 1991-02-20 1992-09-25 Fujitsu Ltd ロボットの遠隔制御装置
US5809099A (en) * 1997-05-05 1998-09-15 Korea Atomic Energy Research Institute Laser-guided underwater wall climbing robot for reactor pressure vessel inspection
CN101434067A (zh) * 2008-12-12 2009-05-20 四川成焊宝玛焊接装备工程有限公司 机器人离线编程与现场调试无缝衔接方法
JP2012042379A (ja) * 2010-08-20 2012-03-01 Kayaba Ind Co Ltd 原子炉検査用ロボット
CN102629110A (zh) * 2012-04-12 2012-08-08 余大利 一种基于实物模型的压水堆半实物仿真系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE2154015C3 (de) * 1971-10-29 1974-05-09 Maschinenfabrik Augsburg-Nuernberg Ag, 8900 Augsburg Einrichtung zum Durchführen von Untersuchungen und Wiederholungsprüfungen an den Innenflächen von oben offenen Druckbehältern
CN100398083C (zh) * 2002-08-30 2008-07-02 延自强 虚拟现实针灸穴位定位方法及系统
DE102008027475A1 (de) * 2008-06-09 2009-12-10 Kuka Roboter Gmbh Vorrichtung und Verfahren zur rechnergestützten Generierung einer Manipulatorbahn
JP5916320B2 (ja) * 2011-08-31 2016-05-11 株式会社ハイボット 遠隔操縦装置
CN102848389B (zh) * 2012-08-22 2015-06-17 浙江大学 基于视觉运动捕捉的机械臂标定及跟踪系统实现方法
CN103268381B (zh) * 2013-05-28 2016-02-10 哈尔滨工业大学 一种基于虚拟现实技术的双工件台半物理仿真方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH04269185A (ja) * 1991-02-20 1992-09-25 Fujitsu Ltd ロボットの遠隔制御装置
US5809099A (en) * 1997-05-05 1998-09-15 Korea Atomic Energy Research Institute Laser-guided underwater wall climbing robot for reactor pressure vessel inspection
CN101434067A (zh) * 2008-12-12 2009-05-20 四川成焊宝玛焊接装备工程有限公司 机器人离线编程与现场调试无缝衔接方法
JP2012042379A (ja) * 2010-08-20 2012-03-01 Kayaba Ind Co Ltd 原子炉検査用ロボット
CN102629110A (zh) * 2012-04-12 2012-08-08 余大利 一种基于实物模型的压水堆半实物仿真系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LIU, CHANG'AN ET AL.: "Research on kinematics simulation of maintenance and inspection robot in nuclear power plants.", JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY., vol. 36, 31 October 2008 (2008-10-31), pages 23 - 26, XP008183623 *
See also references of EP3059738A4 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107756376A (zh) * 2017-10-18 2018-03-06 江西制造职业技术学院 一种应用人形机器人的空间站维修与维护方法
CN109531566A (zh) * 2018-11-16 2019-03-29 国网江苏省电力有限公司盐城供电分公司 一种基于虚拟现实系统的机器人带电作业控制方法
CN109531566B (zh) * 2018-11-16 2022-08-19 国网江苏省电力有限公司盐城供电分公司 一种基于虚拟现实系统的机器人带电作业控制方法
CN109483601A (zh) * 2018-12-24 2019-03-19 合肥欣奕华智能机器有限公司 一种工业机器人功能测试系统及测试方法
CN109483601B (zh) * 2018-12-24 2023-11-28 合肥欣奕华智能机器股份有限公司 一种工业机器人功能测试系统及测试方法
CN110704983A (zh) * 2019-10-12 2020-01-17 中国铁路设计集团有限公司 一种参数驱动下关节联动的吊车动态作业仿真方法
CN110704983B (zh) * 2019-10-12 2023-01-20 中国铁路设计集团有限公司 一种参数驱动下关节联动的吊车动态作业仿真方法
CN111883271B (zh) * 2020-06-03 2022-08-16 湖北工业大学 核反应堆压力容器自动检测平台精确定位方法及系统
CN111883271A (zh) * 2020-06-03 2020-11-03 湖北工业大学 核反应堆压力容器自动检测平台精确定位方法及系统
CN114619436A (zh) * 2020-12-08 2022-06-14 山东新松工业软件研究院股份有限公司 一种基于EtherCAT的六轴机器人控制系统测试设备及其方法
CN113359987A (zh) * 2021-06-03 2021-09-07 煤炭科学技术研究院有限公司 基于vr虚拟现实的半实物综采实操平台
CN113359987B (zh) * 2021-06-03 2023-12-26 煤炭科学技术研究院有限公司 基于vr虚拟现实的半实物综采实操平台
CN114367975A (zh) * 2021-11-15 2022-04-19 上海应用技术大学 串联工业机器人控制算法的验证系统
CN114519759A (zh) * 2022-02-14 2022-05-20 同恩(上海)工程技术有限公司 一种视图变换方法、装置及介质
CN114519759B (zh) * 2022-02-14 2023-11-14 同恩(上海)工程技术有限公司 一种视图变换方法、装置及介质
WO2024060141A1 (zh) * 2022-09-22 2024-03-28 宁德时代新能源科技股份有限公司 一种修改运动副参数的方法、装置及产线系统

Also Published As

Publication number Publication date
BR112016008313A2 (zh) 2017-08-01
BR112016008313B8 (pt) 2022-05-31
CN103761996A (zh) 2014-04-30
CN103761996B (zh) 2016-03-02
EP3059738A1 (en) 2016-08-24
BR112016008313B1 (pt) 2022-01-25
EP3059738B1 (en) 2019-11-20
EP3059738A4 (en) 2017-06-28

Similar Documents

Publication Publication Date Title
WO2015055085A1 (zh) 基于虚拟现实技术的无损检测机器人智能检测方法
AU2020201554B2 (en) System and method for robot teaching based on RGB-D images and teach pendant
CN110370314A (zh) 基于多激光跟踪仪的双臂机器人性能测量方法
US8406923B2 (en) Apparatus for determining pickup pose of robot arm with camera
US20130278725A1 (en) Integrated Structured Light 3D Scanner
CN109859275A (zh) 一种基于s-r-s结构的康复机械臂的单目视觉手眼标定方法
CN109483516A (zh) 一种基于空间距离和极线约束的机械臂手眼标定方法
CN109580775A (zh) 利用双多轴机器人装置检查零件的系统和方法
CN103817692B (zh) 无损检测机器人进行智能检测的方法
CN110142770B (zh) 一种基于头戴显示装置的机器人示教系统及方法
WO2023061113A1 (zh) 便携式远程超声扫查系统与安全超声扫查柔顺控制方法
JP2020128000A (ja) 非破壊検査装置及び方法
CN102917844A (zh) 用于检查传感器放置的方法和系统
CN114289934B (zh) 一种基于三维视觉的大型结构件自动化焊接系统及方法
CN109059768B (zh) 一种容器内置零件检测系统的位姿标定方法
Pedersen et al. Using human gestures and generic skills to instruct a mobile robot arm in a feeder filling scenario
JPH0428512B2 (zh)
JPH07237158A (ja) 位置・姿勢検出方法及びその装置並びにフレキシブル生産システム
CN104760044B (zh) 一种无动力关节臂式示教器及其在工业机器人中的应用
CN204525481U (zh) 一种无动力关节臂式示教器
CN109916392A (zh) 一种基于编码器的机械装备位置检测装置及其系统、方法
Thoo et al. Online and offline robot programming via augmented reality workspaces
JP2003271993A (ja) モニタ画像処理方法及び画像モニタシステム並びに保全作業システム
CN202079595U (zh) 新型遥操作远端机器人控制平台
Brown et al. Automated full matrix capture for industrial processes

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14854066

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112016008313

Country of ref document: BR

REEP Request for entry into the european phase

Ref document number: 2014854066

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2014854066

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 112016008313

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20160414