期刊文献+

基于神经网络PID控制器的外骨骼系统设计 被引量:2

Design of Exoskeleton System Design Based on NN-PID Controller
在线阅读 下载PDF
导出
摘要 外骨骼关节驱动电机使用区别于常规需要对应情况长时间调参的普通PID控制器,以可自适应的神经网络PID控制器为核心控制方法。使用STM32F103RCT6作为外骨骼机器人的局部控制器,OrangePi PC为高性能核心控制器,使用nRF24L01、WH148 B10K、HX711、肌电传感器、AB相编码器、MPU9250、ACS712等模块分别实现各区间2.4 GHz无线/CAN(controller area network,独立控制器局域网络)有线通信、关节角度采集、足底压力采集、EMG采集、整体姿态采集和电源信息采集。该外骨骼机器人将使用高性能的锂电池组作为能源,整体机械结构使用SolidWorks对各部分关节等结构进行设计和建模,并考虑加工成本和难度问题,使用3D打印和钻铣等加工方式完成各零件加工。在算法中使用基于肌电信号(electromyogrsphy, EMG)、零力矩点(zero moment point, ZMP)步态稳定性判据和惯性测量单元(inertial measurement unit, IMU)作为主动控制器的核心,并在单片机程序底层中实现类分时操作系统,各分区使用带校验数据帧的通信方式。经开发验证,该设计可行并方便模块化升级改进。 The exoskeleton joint drive motors use the PID controller which is different from the common PID controller which needs to adjust parameters for a long time,and takes the adaptive NN-PID(neural network PID)as the core control method.STM32F103RCT6 is used as the local controller of exoskeleton robot,OrangePi PC is the high-performance core controller,nRF24L01,WH148 B10K,HX711,EMG sensor,AB phase encoder,MPU9250,ACS712 and other modules are used to realize the 2.4 GHz wireless/CAN(controller area network)cable communication,joint angle collection,plantar pressure collection,EMG collection,AHRS and power information collection.The exoskeleton robot uses the high-performance lithium battery as the energy source,and its whole mechanical structure uses SolidWorks to design and model the joints and other parts of the exoskeleton.Considering the problem of processing cost and difficulty,the 3D printing,drilling and milling and other processing methods are used to complete the processing of each part.In the algorithm,electromyography(EMG),zero moment point(ZMP)stability criterion and inertial measurement unit(IMU)are used as the core of active controller,and the class time-sharing operating system is implemented in the underlying MCU program.The communication mode with check data frame is used in each partition.It is proved that the design is feasible and convenient for modular upgrading.
作者 游博洋 王险峰 赵玲 李龙 逄万山 党志强 YOU Bo-yang;WANG Xian-feng;ZHAO Ling;LI Long;JIANG Wan-shan;DANG Zhi-qiang(School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318,China)
出处 《计算机技术与发展》 2021年第5期1-6,共6页 Computer Technology and Development
基金 2019年黑龙江省大学生创新创业训练计划项目(201910220040) 大庆市科学计划项目(zd-2019-21)。
关键词 外骨骼 STM32 SOLIDWORKS 肌电信号 稳定性判据 exoskeleton STM32 SolidWorks EMG stability criterion
  • 相关文献

参考文献8

二级参考文献89

  • 1归丽华,杨智勇,顾文锦,张远山,杨秀霞.能量辅助骨骼服NAEIES的开发[J].海军航空工程学院学报,2007,22(4):467-470. 被引量:29
  • 2赵晓军,黄强,彭朝琴,张利格,李科杰.基于人体运动的仿人型机器人动作的运动学匹配[J].机器人,2005,27(4):358-361. 被引量:34
  • 3马培荪,曹曦,赵群飞.两足机器人步态综合研究进展[J].西南交通大学学报,2006,41(4):407-414. 被引量:11
  • 4Nelson Costa,Milan Bezdicek,Michael Brown,John O.Gray,Darwin G.Caldwell,Stephen Hutchins.Joint Motion Control of a Powered Lower Limb Orthosis for Rehabilitation[J].International Journal of Automation and computing,2006,3(3):271-281. 被引量:7
  • 5M.伍科布拉托维奇 马培荪 等.步行机器人和动力型假肢[M].科学出版社,1983..
  • 6M. Vukobratovi? and J. Stempanenko. On the Stable of Anthropomorphic Systems[J]. Mathematical Bioscience, 1972. Vol.15:1-37P.
  • 7M. Vukobratovi?, B. Borobvac, D. Surla, and D. Stoki?. Biped Locomotion- Dynamics, Stability, Control and Application[J]. Springer- Verlag, 1990.
  • 8Shuuji Kajita, Hiwhisa Hirukawa, Kazuhito Yokoi and Kensuke Harada:Humanoid Robotics[M]. Ohm-sha, Ltd. 2005:63P.
  • 9俞志伟.双足机器人仿生机构设计与运动仿真[D].哈尔滨:哈尔滨工程大学,2005.
  • 10CHESTNUTT J, LAU M, CHEUNG G, KUFFNER J, HODGINS J. Footstep planning for the Honda ASIMO humanoid [ C ]//Proceedings of the 2005 IEEE International Conference on Robotics and Automation. Barcelona, Spain, 2005.

共引文献81

同被引文献16

引证文献2

二级引证文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部