面向欠驱动攀爬机器人的自适应算法研究及控制系统设计
文献类型：学位论文
| 作者 | 王福华 |
| 学位类别 | 硕士 |
| 答辩日期 | 2017-05-24 |
| 授予单位 | 中国科学院沈阳自动化研究所 |
| 授予地点 | 沈阳 |
| 导师 | 刘玉旺 |
| 关键词 | 欠驱动 攀爬机器人 弹簧耦合 模型参考自适应控制 李雅普诺夫稳定性理论 |
| 其他题名 | Research on Adaptive Algorithm and Control System Design for Underactuated Climbing Robot |
| 学位专业 | 机械电子工程 |
| 中文摘要 | 攀爬机器人作为一种新型机器人，在工业、航天等领域的应用潜力大，不仅可以用于深水区、石油化工、核工业等设备管道的建设和检修，还可从事空间站维修等工作，具有极大应用前景。将欠驱动理念引入到攀爬机器人的研究中，研究团队提出了一种新型欠驱动轮手式攀爬机器人构型，解决了较少驱动源约束下攀爬对象的自适应性问题。但新构型对控制系统提出了新要求和挑战。为了合理地利用欠驱动机构的特点，增强攀爬机器人的自适应性，本文以国家自然科学基金(No:51605474)为依托，展开对该机器人的动力学模型和控制系统的研究，旨在提出一种可调增益模型参考自适应控制算法，使欠驱动机构性能得到更好的发挥，并设计一种实时嵌入式操作系统，提高攀爬机器人的响应速度。首先，在对欠驱动控制算法进行广泛调研的基础上，建立了新型欠驱动攀爬机器人系统的状态空间方程并将其近似线性化，并对该非线性系统以及线性化后的系统进行了稳定性分析。针对欠驱动攀爬机器人数学模型，根据李雅普诺夫理论设计了可调增益模型参考自适应控制器，并进行了仿真分析。然后，进行了系统集成与设计，包括硬件系统和软件系统设计。硬件系统上主要设计了中央控制电路、数据采集电路、电机驱动电路；软件系统上主要将改进的模型参考自适应控制器移植到嵌入式实时操作系统中，保证了控制系统的稳定运行。 最后，为了验证所设计控制算法的有效性，对样机进行组装与调试，安置了相应的传感器以便进行转角信号和压力信号采集，搭建和研究了欠驱动攀爬机器人实验平台，进行了不同直径圆柱体的攀爬实验，并采集实验数据经串口通信设备发送至上位机，对实验结果进行了关节角与接触力响应情况分析。 仿真与实验结果表明，即使存在系统参数不确定性及噪声等因素，本文所提出的自适应控制律也能使系统保持良好的渐进稳定性，得到了稳态的抱持与攀爬过程，从而保证了欠驱动攀爬机器人运行的稳定性与可靠性。 |
| 英文摘要 | Climbing robot is widely applied as a novel robot in the industrial, aerospace and other fields. It has great application prospects, not only equipped in pipeline for construction and maintenance of deep water, petrochemical, nuclear industry, but also engaged in space station maintenance work. The idea of underactuated mechanism is introduced into study of climbing robot. The research team proposed a novel underactuated wheel-manipulator climbing robot, which solves the problem of adaptability to climbing objects in the constraints of minimal driving source. But the new configuration puts forward new requirements and challenges for control system. This paper is based on the National Natural Science Foundation of China (No: 51605474), and studies dynamic model and control system in order to rationally utilize characteristics of underactuated mechanism and enhance adaptability of climbing robot. It aims to design a real-time embedded operating system to improve response speed, and proposed an adjustable gain model reference adaptive control algorithm to play better performance of underactuated mechanism. Firstly, based on widely investigation for underactuated control algorithm, state space equations of underactuated system are established and approximately linearized. Besides, the nonlinear system and the system after linearization are analyzed on stability. Model reference adaptive control with adjustable gain based on Lyapunov theoretical is designed and simulated for mathematical model of underactuated climbing robot. Then, system integration and design are researched, including hardware systems and software system. The central control circuit, data acquisition circuit and motor drive circuit are mainly designed in hardware system. The improved model reference adaptive controller is transformed into embedded real-time operating system in software system, which ensures stable operation of control system. Finally, prototype is assembled and revised to verify effectiveness of the designed control algorithm. Corresponding sensors are placed for angle and pressure signal acquisition, so underactuated climbing robot experiment platform is constructed and researched. Climbing on different diameter cylinders and then experimental data were sent to host computer via serial communication device. And response of joint angle and contact force was analyzed. The simulation and experimental results show that the proposed adaptive control law can maintain good gradual stability in this paper, even if the presence of uncertain system parameters and noise. Steady holding process and stability and reliability of underactuated climbing robot are obtained, which ensures stability and reliability of the underactuated climbing robot. |
| 语种 | 中文 |
| 产权排序 | 1 |
| 源URL | [http://ir.sia.cn/handle/173321/20532]  |
| 专题 | 沈阳自动化研究所_空间自动化技术研究室 |
| 推荐引用方式 GB/T 7714 | 王福华. 面向欠驱动攀爬机器人的自适应算法研究及控制系统设计[D]. 沈阳. 中国科学院沈阳自动化研究所. 2017. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。