A ground test method for visual servoing verification of space manipulator

Abstract

A ground test method for space mechanical arm visual servo verification, the mechanical arm controller, the mechanical arm product, the camera product and the matched software and the like are all 100% involved in ground test verification, the simulation system is used to project the space mechanical arm plane motion to three-dimensional motion, the ground hardware test equipment is reduced, and the economic cost of test verification is reduced; the invention does not limit the mechanical arm motion speed, can carry out ground test verification according to the on-orbit real working condition input according to the overall task, and greatly improves the space-ground consistency of system test; the invention does not need to model the mechanical arm dynamics characteristics, eliminates the error caused by dynamics modeling or external mechanical arm dynamics coupling, and improves the space-ground consistency of dynamics motion equivalence.

Description

Technical Field

[0001] This invention relates to a ground-based test method for visual servo verification of a space robotic arm, and relates to the field of spacecraft on-orbit servicing technology. Background Technology

[0002] Limited by joint output torque, space robotic arms struggle to achieve free movement in three-dimensional space under ground gravity. Therefore, validating on-orbit vision servoing tasks under ground gravity is crucial. Currently popular methods include:

[0003] Chinese Patent CN201410155425.X: Ground Simulation System and Method for Visual Servo Capturing of Moving Targets by Space Robots. In this invention, a first industrial robot is used to simulate the end-effector motion of a space robotic arm with a floating base, and a second industrial robot is used to simulate the target to be captured, equipped with a cooperative marker. The on-orbit motion characteristics of the floating robotic arm and the floating target are calculated using a dynamics computer, and the kinematic positions are solved to the end-effector and cooperative marker. The movements of the first and second industrial robots are then controlled to simulate the visual servo capture of moving targets by the space robot. This method mainly utilizes a dynamics computer to calculate the dynamic motion characteristics of the robotic arm and map the end-effector position to the end-effector of the industrial robots. However, the dynamics calculation requires dynamic modeling of the spacecraft platform and the robotic arm. The modeling process introduces errors in mass, center of mass, friction, joint flexibility, and other aspects, leading to deviations between the simulation and real-world conditions. Furthermore, this ground verification system only verifies the algorithm logic layer; the correctness of the robotic arm software implementation and the hardware-software compatibility are not verified.

[0004] The existing scientific literature "A Review and Outlook on Space Robotic Arm Technology," in Section 2.7, on ground testing and verification technology, points out that current ground acceptance testing of space robotic arms using visual servoing mostly employs semi-physical verification methods. This involves using computers to verify the three-dimensional kinematics and dynamics of the robotic arm, while physical products are used for the end-effector contact capture stage. The existing scientific literature "Research on Hardware-in-the-Loop Semi-Physical Simulation System for Space Robotic Arm Mission Verification" proposes the MTVF method, which incorporates the robotic arm controller into the ground verification system, but the kinematics and dynamics of the robotic arm are still calculated by computer. These methods also suffer from errors introduced by computer modeling and cannot realistically simulate the dynamic processes of the robotic arm. Although this method incorporates the controller into the ground verification system, the physical product of the robotic arm is not included in the ground verification system, thus failing to verify the correctness of the robotic arm software implementation and the compatibility of the hardware and software.

[0005] The existing scientific literature, "Research on Three-Dimensional Full-Physical Ground Test Method for Space Robotic Arms," ​​points out that a suspension method is used to counteract the gravity of the space robotic arm, enabling three-dimensional spatial movement and thus conducting three-dimensional ground verification of on-orbit tests. While this method incorporates the physical robotic arm into the ground verification system, it has limitations: firstly, the suspension wire restricts the robotic arm's movement space, limiting it to small movements at key locations; secondly, the suspension wire system restricts the robotic arm's movement speed, failing to meet the requirements of real-world on-orbit conditions. Summary of the Invention

[0006] The technical problem to be solved by this invention is to overcome the shortcomings of the prior art and eliminate the errors introduced by dynamic modeling in the current ground test scheme.

[0007] The objective of this invention is achieved through the following technical solutions:

[0008] A ground-based testing method for visual servo verification of a space robotic arm includes:

[0009] (1) The pose information of the cooperative marker and the base coordinate system at the end of the industrial robotic arm relative to the hand-eye camera is calibrated;

[0010] (2) The simulation computer starts the simulation. The floating target in the simulation computer moves relative to the world coordinate system according to the target floating motion law.

[0011] (3) During the target floating process, the simulation computer reads the position and posture of the cooperative marker simulation model relative to the hand-eye camera simulation model in real time, and sends the relative pose relationship to the real-time motion controller of the industrial robot arm; the real-time motion controller controls the industrial robot arm to reach the corresponding pose state;

[0012] (4) After the hand-eye camera is powered on, it acquires the pose data of the cooperative marker in real time and sends it to the space robotic arm controller;

[0013] (5) The space robotic arm controller activates visual servo control based on the measurement data from the hand-eye camera, sends joint commands to the space robotic arm, and controls multiple joints of the space robotic arm to move to the commanded position.

[0014] (6) The space robotic arm controller reads the joint angles of the space robotic arm in real time and sends the joint angles to the simulation computer;

[0015] (7) The simulation computer controls the movement of the robotic arm model in the simulation model according to the joint feedback angle sent by the space robotic arm controller, and the hand-eye camera simulation model moves with the movement of the robotic arm.

[0016] (8) The simulation computer reads the pose data of the cooperative marker model relative to the hand-eye camera model in real time during the movement of the robotic arm model and the floating target model, and then proceeds to step (3) until the capture criterion is met, and the experiment ends.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] (1) All hardware and software components of the present invention, including the robotic arm controller, robotic arm product, camera product and supporting software, are 100% tested and verified on the ground. The system is fully verified in terms of the correctness of the algorithm logic, the correctness of the software implementation, and the compatibility of the hardware and software.

[0019] (2) This invention does not limit the movement speed of the robotic arm, and can conduct ground tests to verify the actual working conditions on the track according to the overall task input, which greatly improves the consistency between the system test and the ground test.

[0020] (3) The present invention uses a simulation system to project the planar motion of the space robotic arm into three-dimensional motion, which reduces the ground hardware test equipment and lowers the economic cost of testing and verification.

[0021] (4) This invention does not require modeling of the dynamic characteristics of the robotic arm, eliminating errors caused by dynamic modeling or external robotic arm dynamic coupling, and improving the consistency of dynamic motion equivalence. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of a ground-based remote operation test and verification system for a space robotic arm.

[0023] Figure 2 This is a simulation scenario for simulating a computer.

[0024] Figure label: Figure 1 Figure 1 shows the motion diagram of the space robotic arm in an air-floating state. The other components are: industrial robotic arm-2, cooperative marker-2-1, hand-eye camera-3, industrial robotic arm control computer-4, simulation computer-5, space robotic arm controller-6, simulation model of the space robotic arm in the simulator-7, simulation model of the hand-eye camera fixed at the end of the robotic arm-7-1, floating target simulation model-8, and cooperative marker simulation model-8-1. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0026] A ground-based testing method for visual servo verification of a space robotic arm, the ground-based testing system including an industrial robotic arm 2, a hand-eye camera 3, a real-time motion controller 4, a simulation computer 5, and a space robotic arm controller 6, such as... Figure 1 As shown; the industrial robotic arm 2 has a cooperation identifier 2-1 at its end, and the hand-eye camera 3 is used to identify the cooperation identifier 2-1; the simulation scene in the simulation computer 5 is shown in [the image / description]. Figure 2 .

[0027] Ground testing methods include:

[0028] (1) Before the system starts working, the pose information of the cooperative marker 2-1 at the end of the industrial robotic arm 2 and the base coordinate system relative to the hand-eye camera 3 is calibrated. Assume the pose matrix of the cooperative marker 2-1 relative to the hand-eye camera 3 is... The pose matrix of hand-eye camera 3 relative to the industrial robotic arm 2 base coordinate system is: This matrix is ​​a fixed value; the pose matrix of the cooperative identifier 2-1 relative to the base coordinate system is: The relationship between the three pose matrices above is as follows: Where inv() is the matrix inversion operation;

[0029] (2) Simulation computer 5 starts simulation. In the simulation computer, the floating target moves relative to the world coordinate system according to the target floating motion law.

[0030] (3) The simulation computer reads the position and attitude of the cooperative marker simulation model 8-1 relative to the hand-eye camera simulation model 7-1 in real time during the target's floating process, assuming it is... The relative pose relationship is then sent to the real-time motion controller 4 of the industrial robotic arm 2. This control... but After this matrix is ​​obtained, the real-time motion controller 4 controls the industrial robotic arm 2 to reach the corresponding pose state.

[0031] (4) After the hand-eye camera 3 is powered on, it acquires the pose data of the cooperative marker 2-1 in real time and sends it to the space robotic arm controller 6.

[0032] (5) The space robotic arm controller 6 activates visual servo control based on the measurement data of the hand-eye camera 3, sends joint commands to the space robotic arm, and controls the 7 joints of the space robotic arm to move to the command position.

[0033] (6) The space robotic arm controller 6 reads the joint angles of the space robotic arm in real time and sends the joint angles to the simulation computer 5;

[0034] (7) The simulation computer 5 controls the movement of the robotic arm model 7 in the simulation model according to the joint feedback angle sent by the space robotic arm controller 6, and the hand-eye camera simulation model 7-1 moves with the movement of the robotic arm.

[0035] (8) The simulation computer 5 reads the pose data of the cooperative marker model 8-1 relative to the hand-eye camera model 7-1 in real time during the movement of the robotic arm model 7 and the floating target model 8, and returns to step (3) until the capture criterion is met, and the experiment ends.

[0036] Among them, the real-time motion controller 4 is particularly important as the information flow interaction center of the entire system. The information flow interaction between robotic arm 1, industrial robotic arm 2, and simulation computer 5 is realized through this controller. The controller adopts the Linux operating system and has undergone real-time kernel modification. It controls the information flow interaction through real-time kernel functions, minimizing the latency caused by communication between systems and thus reducing system inconsistency.

[0037] (1) The real-time motion controller 4 serves as the controller of the industrial robotic arm 2. Its kinematic and dynamic calculation processes are implemented therein. It sends joint data to the joint layer of the industrial robotic arm 2 via the Ethercat bus and adopts DC synchronization mode. The control cycle is 1ms and the cycle jitter is no more than 50us. The interactive data includes the joint position command data and joint position feedback data of the industrial robotic arm 2.

[0038] (2) The real-time motion controller 4 and the space robotic arm controller 6 communicate via a CAN bus, with a data communication cycle of 1ms. The communication data mainly includes the joint position feedback data of the space robotic arm 1.

[0039] (3) The real-time motion controller 4 and the simulation computer 5 communicate using the TCP / IP protocol. The data interaction period is 1ms. The TCP / IP communication data reading and writing is implemented in the simulation computer 5 through shared memory. The interactive data is the position and attitude of the cooperative identifier simulation model 8-1 relative to the hand-eye camera simulation model 7-1, and the joint feedback data of the spatial robotic arm 1.

[0040] The contents not described in detail in this specification are common knowledge to those skilled in the art.

[0041] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A ground test method for vision servo verification of a space manipulator, characterized by, include: (1) The pose information of the cooperative marker and the base coordinate system at the end of the industrial robotic arm relative to the hand-eye camera is calibrated; (2) The simulation computer starts the simulation, and the floating target in the simulation computer moves relative to the world coordinate system according to the target floating motion law; (3) During the target floating process, the simulation computer reads the position and attitude of the cooperative marker simulation model relative to the hand-eye camera simulation model in real time, and sends the relative pose relationship to the real-time motion controller of the industrial robot arm; the real-time motion controller controls the industrial robot arm to reach the corresponding pose state; (4) After the hand-eye camera is powered on, it acquires the pose data of the cooperative marker in real time and sends it to the space robotic arm controller; (5) The space robotic arm controller activates visual servo control based on the measurement data from the hand-eye camera, sends joint commands to the space robotic arm, and controls multiple joints of the space robotic arm to move to the commanded position. (6) The space robotic arm controller reads the joint angles of the space robotic arm in real time and sends the joint angles to the simulation computer; (7) The simulation computer controls the movement of the robotic arm model in the simulation model according to the joint feedback angle sent by the space robotic arm controller, and the hand-eye camera simulation model moves with the movement of the robotic arm; (8) The simulation computer reads the pose data of the cooperative marker model relative to the hand-eye camera model in real time during the movement of the robotic arm model and the floating target model, and then proceeds to step (3) until the capture criterion is met, and the experiment ends.

2. The ground testing method according to claim 1, characterized in that, In step (1), assume the pose matrix of the co-marker with respect to the hand-eye camera is ; The pose matrix of the hand-eye camera relative to the base coordinate system of the industrial robotic arm is: ; The pose matrix of the cooperative identifier relative to the base coordinate system is The relationship between the above three pose matrices is as follows: =inv( , where inv() is the matrix inversion operation.

3. The ground testing method according to claim 2, characterized in that, In step (3), the simulation computer reads the position and orientation of the cooperative marker simulation model relative to the hand-eye camera simulation model in real time during the target's floating process, assuming it is... The relative pose relationship is sent to the real-time motion controller of the industrial robotic arm, and then controlled. = ,according to = = The real-time motion controller controls the industrial robotic arm to reach the corresponding position and posture.

4. The ground testing method according to claim 1, characterized in that, The real-time motion controller sends joint data to the joint layer of the industrial robot via the EtherCAT bus and adopts DC synchronization mode. The interaction data between the real-time motion controller and the industrial robot includes the joint position command data and joint position feedback data of the industrial robot.

5. The ground testing method according to claim 1, characterized in that, The real-time motion controller and the space robotic arm controller communicate via a CAN bus, and the communication data includes the joint position feedback data of the space robotic arm.

6. The ground testing method according to claim 1, characterized in that, The real-time motion controller and the simulation computer interact with each other using the TCP / IP communication protocol. The simulation computer uses shared memory to read and write data for TCP / IP communication. The interactive data includes the position and posture of the cooperative identifier simulation model relative to the hand-eye camera simulation model, and the joint feedback data of the spatial robotic arm.

7. The ground testing method according to claim 1, characterized in that, This ground-based testing method does not limit the movement speed of the space robotic arm.

8. The ground testing method according to claim 1, characterized in that, This ground testing method does not require a model of the robotic arm's dynamic characteristics.

Images