| Whole Machine and System Design |
|
|
|
|
| Design and control research of aeroengine blade detection robot |
| ZHENG Yu-chen, JU Feng, WANG Dan, SUN Jing-bin, WANG Ya-ming, CHEN Bai |
| College of Mechanical and Electrical Engineering， Nanjing University of Aeronautics and Astronautics， Nanjing 210016， China |
|
|
|
Abstract Aiming at the problems that it is difficult to manually repair in the narrow internal space of aeroengine, as well as the weak bearing capacity of traditional continuum robot, a aeroengine blade detection robot was designed. The main body of the robot was composed of multi joint arms, which were connected by universal joints and driven by ropes; a continuum was installed at the end of the robot. In order to improve the end position accuracy of the robot, the closed-loop controller was designed and studied. The kinematics model of the joint arm was established, the calculation formula of the change of the length of the driving rope was deduced, and the method of converting the reading of the inertial measurement unit into the joint angle was introduced. Kalman filter was used to reduce measurement noise, and the best estimated value of joint angle was obtained as feedback. Based on multi-dimensional Taylor network optimal control (MTNOC), a closed-loop controller was designed, the characteristics and advantages of MTNOC controller were analyzed, and the effectiveness of MTNOC controller was verified by simulation and experiment. The results showed that Kalman filter could effectively reduce the measurement noise; MTNOC controller had better adaptability and dynamic characteristics than PID controller, so that the joint arm also had faster response speed at the moment of large state change. Under the control of MTNOC controller, the end position accuracy of blade detection robot was improved, which increased the accuracy and reliability of aeroengine blade detection results.
|
|
Received: 20 July 2020
Published: 28 October 2021
|
|
|
航空发动机叶片检测机器人的设计与控制研究
针对航空发动机内部空间狭窄难以进行人工检修，以及传统连续体机器人承载能力弱等问题，设计了一种用于航空发动机叶片检测的机器人。机器人主体由多节关节臂组成，关节臂通过万向节连接，由绳驱动；在机器人末端安装连续体。为了提高机器人末端位置精度，开展了闭环控制器的设计和研究。建立了关节臂运动学模型，推导了驱动绳绳长变化量的计算公式，介绍了将惯导传感器读数转换为关节角的方法。采用卡尔曼滤波器以减小测量噪声，得到关节角的最佳估计值，并将其作为反馈量。基于多维泰勒网优化控制（multi-dimensional Taylor network optimal control，MTNOC）设计了闭环控制器，分析了MTNOC控制器的特点及优势，并通过仿真和实验来验证MTNOC控制器的有效性。结果表明，卡尔曼滤波器能有效减小测量噪声；MTNOC控制器比PID控制器具有更好的适应性和动态特性，使关节臂在状态变化较大的瞬间也具有较快的响应速度。在MTNOC控制器的控制下，叶片检测机器人的末端位置精度得到提高，从而提高了航空发动机叶片检测结果的准确性和可靠性。
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
