1. Robotics Systems Laboratory, Changzhou Institute of Advanced Manufacturing Technology, Changzhou, Jiangsu 213164, China; 2. Robotics Systems Laboratory, Institute of Advanced Manufacturing Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Changzhou, Jiangsu 213164, China
Abstract: To address the problems of long manual positioning time and inaccurate positioning in existing demolition robots, a control system software based on laser positioning for the autonomous movement of the robot arm was developed. Kinematics, closed-loop feedback adjustment, hydraulic device control, hand-eye calibration, and motion control modules were designed. Practical application results show that the software system can achieve the control system design specifications and improve work efficiency.
Keywords: demolition robot; kinematics; autonomous movement
Chinese Library Classification Number: TP273 Document Identifier: A DOI: 10.19358/j.issn.1674-7720.2017.09.030
Citation format: He Feng, Zhang Xiaojian, Zhao Jianghai. Software design of autonomous motion control system for demolition robot robotic arm [J]. Microcomputer Applications, 2017, 36(9): 103-105.
introduction
Currently, demolition robots used in construction sites are operated manually [1]. Due to the large inertial force generated during the movement of the demolition robot's robotic arm, operators need to make repeated adjustments to move it to the approximate position. Therefore, it is essential to develop an autonomous motion control system for the demolition robot's robotic arm and design the upper-level software of this control system to realize the autonomous motion control function of the robotic arm's target guidance [2].
This paper designs the various modules of the control system software, gives the design method of each module, and finally proves through experiments that the control system software can complete the autonomous control function of target guidance of the demolition robot arm [3].
1. Overall Hardware Architecture of the Demolition Robot
This paper's experimental platform is based on the GTRC-15 demolition robot from Jingtian Hydraulic Co., Ltd., and focuses on the functional component research. The actual physical diagram of the robot is shown in Figure 1. As can be seen from Figure 1, the robot consists of a robotic arm, a mobile rotary platform, a walking track, and a working support. The robotic arm mainly consists of a main arm, a second arm, a third arm, and an end effector. The rotary platform is driven by a rotary motor and is equipped with an angle encoder to obtain the current offset angle of the rotary platform. Each joint of the robotic arm is driven by a hydraulic cylinder and is equipped with an angle encoder sensor to detect the current joint angle value. The walking track is driven by a hydraulic motor and can drive the robot to perform forward, backward, and rotational movements. The robot's communication uses the CAN bus mode [4]. The control system mainly includes the electro-hydraulic proportional control system for each joint of the robotic arm, a wireless video acquisition and transmission system, a laser positioning platform control system, and the robot's main control PC. The electro-hydraulic proportional control system mainly controls the flow rate of the liquid in the hydraulic cylinder by controlling the opening of the proportional valve, and the flow rate has an approximately linear relationship with the movement speed of the hydraulic cylinder and the instantaneous speed of the joint angle. The wireless video acquisition and transmission system is used by operators to observe and locate work points remotely via wireless video when they are far from the work site. This allows the laser positioning platform to automatically find the target work point and the hydraulic robotic arm to move automatically to the designated target position.
2. Software Design of Demolition Robot
The control system software for the demolition robot uses the Windows XP system platform, and the program framework is developed using the MFC class library in VC6. The control system includes modules for kinematics calculation, closed-loop feedback regulation, hydraulic device control, motion control, hand-eye calibration, and wireless video acquisition and processing. The control system software interface is shown in Figure 2.
The working principle of the overall control system for the autonomous motion of the robotic arm under laser positioning of the demolition robot is as follows: The operator observes the work site through a remote video monitoring system and determines the target point to be broken according to the site environment. Then, the operator operates the laser positioning gimbal to perform pitch and rotation movements until the laser point falls on the positioned target point. At this time, the control system can easily calculate the spatial coordinates of the target point in the laser positioning platform coordinate system using the laser gimbal coordinate system solution formula given above, based on the rotation angle of the laser positioning platform in two directions and the current reading of the laser rangefinder. This coordinate can be converted into the spatial coordinate value of the target in the robotic arm coordinate system through rigid transpose. The spatial motion quantity is converted into the joint motion output quantity through inverse kinematics solution [5] or online trajectory planning method [6], thereby driving the hydraulic actuator to enable the robotic arm to achieve multi-axis linkage action and quickly and accurately reach the designated demolition target.
The algorithm steps for autonomous demolition are as follows: (1) Based on the distance of the target measured by the calibration device, the three-dimensional coordinates under the visual gimbal coordinate system are obtained after rotation and offset. The laser range sensor is installed at the origin of the Y-axis under the gimbal coordinate system, and the laser emission direction is consistent with the positive direction of the Y-axis. If the current laser range sensor reading is dm, the coordinates of the target under the fifth coordinate system are represented as [0,d,0]. (2) The coordinates of the target object under the laser range system are obtained by calculating the homogeneous transformation matrix between the base coordinate system and the end effector coordinate system. (3) The coordinates of the target point under the first coordinate system and the set end effector pose are converted into the angle values of the five joint angles of the hydraulic manipulator by inverse kinematics solution. (4) The target length of the hydraulic cylinder movement is obtained by the mapping relationship between the joint angle and the length of the hydraulic cylinder. (5) The speed signal of the hydraulic component is controlled by the PID algorithm to make it reach the target point quickly and accurately, and complete the demolition positioning.
3. Kinematic Module Design
The kinematics module mainly includes a joint angle calculation part, a forward kinematics calculation part, and an inverse kinematics calculation part [7]. The joint angle calculation part mainly establishes the linear mapping relationship between the change in hydraulic cylinder and the change in joint angle. The forward kinematics calculation part establishes the correspondence between the current value of each joint angle and the spatial coordinates and attitude of the end effector. The inverse kinematics calculation part calculates the target value of each joint angle under the specified target position and specified end effector attitude, satisfying the energy optimal condition [8]. Through the kinematics module design, the current attitude of the robotic arm and the position of the end effector can be obtained online, and the target value of each joint angle can be obtained according to the specified target point and end effector attitude [9], and the target value is converted into the length of hydraulic cylinder movement.
4. Closed-loop feedback adjustment module design
The closed-loop feedback control module is mainly used to adjust the positional motion error of each hydraulic cylinder. PID control is used for closed-loop adjustment, and the PID control parameters are designed empirically to obtain suitable PID control parameters for each hydraulic cylinder. After PID closed-loop feedback adjustment, the positional motion error of each hydraulic cylinder joint is within 0.01mm. The PID control model diagram is shown in Figure 3.
5. Hydraulic Device Control Module Design
The hydraulic control module communicates with the electro-hydraulic proportional control system of each joint via a CAN bus, controlling the valve opening, direction of movement, and speed of each hydraulic cylinder. Through this hydraulic control module design, the movement of each joint of the robotic arm can be controlled.
6. Hand-eye calibration module design
The control system software uses a laser positioning platform to obtain the coordinate system points of the vision system. In order to integrate the vision coordinate system and the robotic arm coordinate system, a hand-eye calibration module needs to be designed. The hand-eye calibration module uses the least squares algorithm to calculate the homogeneous transformation matrix between the two coordinate systems. The target coordinates and target posture in the vision coordinate system can be transformed into coordinates in the robotic arm coordinate system through the homogeneous transformation matrix. The change of each hydraulic cylinder is further calculated through the kinematics module, and the hydraulic device control module and closed-loop feedback adjustment module move to the specified target position [10].
7. Motion Control Module Design
The motion control module primarily handles the multi-axis motion control of the robotic arm. The software employs cubic spline interpolation, using the target values of each joint angle obtained through inverse kinematics, the current values of each joint angle fed back from the encoders, and the agreed-upon overall motion time as interpolation input parameters to calculate the position, velocity, and acceleration values of each joint angle in each timer cycle. The motion characteristics of the hydraulic cylinders are utilized for multi-axis motion control of the robotic arm.
8. Conclusion
Using the control system software designed in this paper, six target points were acquired through a laser positioning device. The final posture was taken as the end effector hammer being perpendicular to the ground. Experimental data (Table 1) shows that the positioning error was within 5 cm, and the positioning time was 10 seconds, which is adjustable. The experiment demonstrates that the control system software meets the goal of automating the movement of the demolition robot's robotic arm.
References
[1] Liu Qingyun. Key technologies of demolition and rescue robots [J]. Modern Manufacturing Engineering, 2009, 30(7): 149-153.
[2] Yang Zheng, Shang Jianzhong, Wang Biao, et al. Trajectory planning of hydraulically driven robotic arms [J]. Mechanical Research and Application, 2009, 25(4):47-51.
[3] Zhou Mengran, Wu Yongxiang. Research on mine hoist position tracking system based on infrared laser positioning technology [J]. Journal of Coal Science and Technology, 2002, 27(6):658-660.
[4] Hu Min, Liang Conghui, Zheng Qinghua, et al. Design and implementation of real-time control software for multi-joint hydraulic boom [J]. Proceedings of the Chinese Society for Construction Machinery, 2013, 11(3): 243-247.
[5] Cai Zixing. Robotics [M]. Beijing: Tsinghua University Press, 2000.
[6] CRAIGJ. Introduction to robotics: mechanics and control (third edition) [M]. Pearson Education, Inc., 2005.
[7] Lu Zhen. Principles and Applications of Redundant Degrees of Freedom Robots [M]. Beijing: China Machine Press, 2006.
［8］
[9] CARRILERWF, KHOSLAPK, KROGHBH. Path planning for mobile manipulators for multiple task execution [J]. IEEE Transaction son Robotics and Automation, 1991, 7 (3): 403-408.
[10] FINZELR, HELDUSERS, JANGDS. Electrohydraulic control systems for mobile machinery with low energy consumption [C]. Proceedings of the Seventh International Conference on Fluid Power Transmission and Control, 2009: 214-219.
