透過您的圖書館登入
IP:18.223.108.186
  • 學位論文

基於Kinect影像辨別之機械手臂輔具研製

Development of Manipulator Arm Assistant Tool Base on Kinect Image Recognition

指導教授 : 邱謙松

摘要


市面上廣泛所見的輔具大多以解決使用者行動上的不便為目的,本研究針對行動不便者往往是以坐著的姿態在「拿東西、撿東西」,使得簡易的日常生活上肢行動變得有些困難的問題進行解決,因此本論文整合關節型機械手臂於電動輪椅上二次開發行動輔具,研製特殊型關節機械手臂,由解析正逆運動學,利用區段性立方多項式差補法(Piecewise Cubic Hermite Interpolating Polynomial,PCHIP)規劃機械手臂運動路徑軌跡,夾取目標物件,接著結合Kinect影像輸入識別物件及針對使用者用戶提出雷射光筆物件選擇判定技術,轉換成機械手臂控制命令,省去使用者學習操作機械設備的時間。最後針對影像判定物件規劃模擬運動軌跡路徑,及即時機械手臂姿態判定測量;實驗結果針對末端點誤差進行探討,判定軌跡規劃運動與循跡控制應用上,可以有效降低偏移以及抑制誤差,強健系統的穩定性。且同時驅動機械手臂輔助完成使用者在取物姿態及餵食姿態上的路徑規劃,幫助使用者完成「取物」及「餵食」的日常工作,進而輔助使用者及達成輔具的智能化,大幅提升醫療服務價值。

並列摘要


In recent years, the market is flooded with Mobility Assistive Technology (MAT), where most of them solved the inconvenient problem on people walking. It's difficult for picking something up or taking something when the disable people are sitting. To this end, this research opens up the new MAT and integrates anthropomorphic manipulator in an electric wheelchair. First, the kinematics of the robotic arm is developed, while the path planning of manipulator is achieved by using Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) for picking target object. On the other hand, the image recognition of the Kinect is used to determine the target object chosen by a laser pointer. Then, the human-machine interface is developed to save the time to learn and operate the machine. Finally, this research is verified by the simulation of the path planning with kinematics and decided robot attitude in real time. The experimental results show the effectiveness of the proposed method and great accurate position in operation. Finally, drive the operator to choose something, or feed the disabled. And reached the AI of the MAT, greatly enhance the value of medical services.

參考文獻


[49] 王文宏,“基於超寬頻測距與慣性感測器之室內定位系統”,中原大學電機工程研究所碩士論文,2016。
[48] 邱謙松,“約束系統之強健非線性控制理論及其在機器人之應用”,中原大學電子工程研究所博士論文,2000。
[5] J. Viola, L. Angel and J. M. Sebastian, ” Design and robust performance evaluation of a fractional order PID controller applied to a DC motor,” IEEE/CAA Journal of Automatica Sinica, vol. 4, np. 2, pp. 304 - 314, 2017.
[6] A. Bhambhani and P. Shah, “PID parameter optimization using Cohort intelligence technique for D.C motor control system,” International Conference on Automatic Control and Dynamic Optimization Techniques (ICACDOT), pp. 465 – 468, 2016.
[7] H. Yang and G. Hu, ” Comparative study on PID Control and Fuzzy Control of motor speed of hydraulic transmission type wind turbine,” IEEE International Conference on Power and Renewable Energy (ICPRE), pp. 399 - 403, 2016.

延伸閱讀