(3.238.130.97) 您好!臺灣時間:2021/05/14 19:39
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:鄭煒翰
研究生(外文):Wei-HanCheng
論文名稱:重度肢體障礙者互動電動輪椅之模糊類神經控制系統
論文名稱(外文):Fuzzy Neural Network Control System of Interactive Electrical Power Wheelchair for People with Severe Physical Disabilities
指導教授:羅錦興羅錦興引用關係陳沛仲陳沛仲引用關係
指導教授(外文):Ching-Hsing LuoPei-Chung Chen
學位類別:碩士
校院名稱:國立成功大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2016
畢業學年度:104
語文別:英文
論文頁數:79
中文關鍵詞:電動輪椅嵌入式系統模糊類神經控制摩斯碼控制器重度肢體障礙STM32F4-Discovery
外文關鍵詞:Electrical Power WheelchairEmbedded SystemFuzzy-Neural Network ControlMorse Code ControllerSevere Physical DisabilitiesSTM32F4-Discovery
相關次數:
  • 被引用被引用:0
  • 點閱點閱:137
  • 評分評分:
  • 下載下載:31
  • 收藏至我的研究室書目清單書目收藏:0
一直以來,社會上有許多重度肢體障礙者缺乏行動自主性而長期無法出門。雖然市面上目前有許多電動輪椅,提供搖桿、腦波以及聲控等不同的控制方式,但始終缺乏適合重度肢體障礙者之電動輪椅系統。障礙者操作輪椅時會面臨多種困境,其一是控制器的輸入方式,再來是電動輪椅本身的安全性與舒適性,以及因為缺乏活動能力所造成的視覺死角。過去本實驗室針對這三點困境提出了解決方案,其中最為重要的即是電動輪椅的安全性。
本研究延續過去之研究提出一個模糊類神經網路控制系統,以改善過去系統控制方面的不足,提升電動輪椅的安全性,本篇論文所述之互動式電動輪椅系統以過去實驗室開發之摩斯碼控制器為基礎,由電動輪椅、嵌入式系統、平板電腦搭配編碼器、電流感測器、藍芽模組等感測模組所構成。使用者透過摩斯碼控制器輸入指令至平板電腦,平板電腦將指令傳輸至電動輪椅系統之的控制單元,以控制電動輪椅之動作,例如:輪椅擺位調整、前後左右之行動。而在系統架構上,延續過去研究提出的模組化設計,以因應不同使用者的使用需求。在系統整合上應用開放原始碼嵌入式作業系統FreeRTOS 作為控制單元STM32F4-Discovery的作業系統,以增進系統運作時的效率,並使指令的處理過程更加順暢,以確保系統之穩定。電動輪椅的各項元件與細節將於第二章介紹。
本研究之控制系統主要以電動輪椅之動態模型為基礎,輸出對應速度之控制訊號,並將回授之誤差作為模糊類神經控制方法之輸入,經過模糊類神經之調整法則後,輸出補償訊號。每20毫秒執行一次模糊類神經控制以期調整電動輪椅之控制訊號輸出至最佳狀態。電動輪椅之動態模型經由實驗數據推算取得亦可進一步以迭代法取得較佳的動態模型。本研究應用的模糊類神經系統立基於Lyapunov穩定性定理,模糊類神經控制之調整法則經由推導證明能使模糊類神經系統穩定。
經由實驗可推算出電動輪椅之動態模型,利用此輪椅動態模型可得知轉速與控制訊號之間之對應關係,以此對應關係為基礎,加入模糊類神經網路控制器修正其中之誤差,並以程式模擬誤差不同時模糊類神經網路控制器的各項控制訊號之輸出。實驗結果顯示,在輪椅控制方面,模糊類神經網路控制器於負載不同的情形下能使輪椅穩定前進。實驗條件為磁磚地板環境,並設定目標速度為3 km/hr左右,當輪椅無載時,移動5秒鐘後,約3公尺後,兩輪之誤差為0.063%。同情況下,當使用者體重為60公斤時,兩輪之誤差為1.34%。此外,執行了數次個案測試,重度障礙使用者以嘴控開關及摩斯碼控制器輸入指令實際操作電動輪椅,包含擺位調整及各方向之行動。
本研究延續過去之研究成果,保留模組化之設計、移除非必要之元件,以電動輪椅之動態模型為基礎,應用模糊類神經控制於電動輪椅控制系統上,以提升電動輪椅移動時的穩定性與安全性。透過本研究所設計的系統,重度肢體障礙者藉由摩斯碼控制器輸入指令來穩定地自行操作電動輪椅,使得重度肢體障礙者更容易也樂於走向戶外迎向陽光。

Lots of people with severe physical disabilities could not go outside for a long time due to the loss of mobility. There are many different kinds of commercial electrical power wheelchair on the market which provides different input methods that the users with disabilities could control the wheelchair with some method like joystick, brain wave control or sound control. But, none of them match the demands of the user with severe physical disabilities. The difficulties would occur when they operate the wheelchair and need to be overcome. One is the input method of the wheelchair, another is the safety and comfort of the wheelchair, and the other is the blind spot caused by the vision constraint due to the lack of body mobility. Several researches for the difficulties have been proposed, the most important is the safety of the wheelchair and the user. The goal of this researches is to increase the safety of the wheelchair by improving the control system with fuzzy neural network control.
The study is based on the researches of seniors in the past and propose a control system based on fuzzy neural network to make the system of wheelchair better. The system of interactive electrical power wheelchair in the study include the electrical power wheelchair, the embedded system, an Android tablet and the components of the wheelchair. The system is based on well-developed Morse code controller, which the users with disabilities could input command to the tablet with. The App on the tablet would transfer the command to the control unit of the wheelchair to operate the wheelchair such as displacement adjustments of the wheelchair and wheelchair moving. On the system architecture, the system is modularized to match the different demands of the user. Moreover, open source real-time operating system for embedded system is applied as the OS of the control unit STM32F4-Discovery, which make the control system work efficiently, smoothly and stably. More details about the electrical power wheelchair system would be introduced in Chapter 2.
The control system of this study outputs the corresponding control signals of the target speed based on the dynamic model of the wheelchair. And, fuzzy neural network control would output the compensating control signals with the algorithms proposed according to the error every 20ms. The dynamic model of the wheelchair could be obtained from the data of the experiments, and the better dynamic model could be obtained by iterative experiments. The fuzzy neural network control system applied to the research is based on Lyapunov stability theorem. The tuning methods of fuzzy neural network is guaranteed to make the system stable.
With the dynamic model of the wheelchair, the relation between the speed and the control signals would be known. The simulations of the outputs of the control signals are performed with the different error and conditions according to the relation mentioned above and fuzzy neural network controller. Practical tests are also performed. The experiments show the wheelchair could move straightly and work stably with different load on the tile ground. The target speed is set to about 3 km/hr, after the wheelchair moving forward for 5 seconds and 3 meters only with the load of the wheelchair, the error is 0.063%. And, when the wheelchair moves forward in the same conditions with the load of the wheelchair and 60kg-weight, the error is 1.34%. The case tests are performed which the users with severe physical disabilities control the wheelchair on their own with their own switch. The tests include displacement adjustments of the wheelchair and the moving test in all directions
The study is based on the researches of seniors. The concept of modularization is retained and the unnecessary components are removed. The study applies the dynamic model of the wheelchair and fuzzy neural network control to the control system to improve the stability and the safety. With the control system proposed, people with severe physical disabilities could input the command with the Morse code controller and their switches to operate the electrical power wheelchair safely and stably. This makes people with severe physical disabilities be able to and be glad to go outdoors.

摘要 I
Abstract III
誌謝 VI
Contents VII
List of Table IX
List of Figures X
Chapter 1 Introduction 1
1.1 Motivation and objective 1
1.2 Background 2
1.3 Related research 3
1.3.1 Morse Code Controller 4
1.3.2 Smart Electrical Power Wheelchair Project 4
1.3.3 Wheelchair Dynamics 7
1.3.4 Fuzzy Neural Controller 7
Chapter 2 System Architecture and Design 9
2.1 System Architecture of Electrical Power Wheelchair 9
2.2 Basic Components of Electrical Power Wheelchair 11
2.2.1 Encoders 11
2.2.2 Current Transducer 15
2.2.3 Actuators 19
2.2.4 Motors and Drivers 23
2.2.5 Integrated Circuit of Electrical Power Wheelchair 27
2.3 Embedded System Platform 30
2.3.1 STM32F4-Discovery 31
2.3.2 FreeRTOS 33
Chapter 3 Methods 38
3.1 Control Flow 38
3.2 Model of Electrical Power Wheelchair 39
3.3 Control Algorithm 42
3.3.1 Fuzzy Neural Network 43
3.3.2 Fuzzy Neural Controller 47
Chapter 4 Results 52
4.1 Simulation 52
4.1.1 Modeling of Wheelchair 52
4.1.2 FNN Controller of Wheelchair 54
4.1.3 Simulation Results 55
4.2 Practical Test 59
4.2.1 Results of Motor Running Test 61
4.2.2 Results of Wheelchair without Load 64
4.2.3 Results of Wheelchair with Load of 60kg-Weight 67
4.2.4 Difference between FNN control and PID control 70
4.3 Case Test 71
Chapter 5 Discussion and Conclusion 74
5.1 Discussion 74
5.2 Conclusion 75
5.3 Future Work 75
References 77

[1]Chung-Min Wu and Ching-Hsing Luo. “Morse Code Recognition System with Fuzzy Algorithm for Disabled Persons. J. Med. Eng. & Tech., Vol. 26, Num. 5, pp. 202-207, 2002.
[2]Chin-Hsien Liang, Chung-Min Wu, Shu-Wen Lin and Ching-Hsing Luo. A Portable and Low-cost Assistive Computer Input Device for Quadriplegics. Technology and Disability, vol. 21, no. 3, pp. 67-78, 2009.
[3] Jau-Yuan Shiao, Yin-Chen Wang, Ching-Hsing Luo, Pei-Chung Chen, Shih-Chung Chen and Meng-Dar Shieh. “Design and Approach of Interactive Assistive Electrical Power Wheelchair for Quadriplegic. Symposium on Engineering, Medicine and Biology Applications, Kaohsiung, Taiwan, Jan. 30 – Feb.1 2015.
[4]Yin-Chen Wang. “Interactive Assistive Electric Wheelchair with Embedded System. National Cheng Kung University, 2014.
[5]Jau-Yuan Shiao. “Neural Network Control and Driver Assistance System of Interactive Electrical Power Wheelchair for People with Quadriplegia, National Cheng Kung University, 2015.
[6]Qiang Zeng, Chee Leong Teo, Brice Rebsamen and Etienne Burdet. “A Collaborative Wheelchair System. IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 16, no. 2, pp 161-170, April 2008.
[7]ST Semiconductor. “RM0090 Reference manual – STM32F405xx, STM32F407xx, STM32F415xx and STM32F417xx advanced ARM-based 32-bit MCUs. September, 2011.
[8]ST Semiconductor. “UM1472 User manual - Discovery kit for STM32F407/417 lines. 2014.
[9]Richard Barry. “Using the FreeRTOS Real Time Kernel - A Practical Guide. FreeRTOS.org, 2010.
[10]Richard Barry. “FreeRTOS Reference Manual - API Functions and Configuration Options. FreeRTOS.org.
[11]Junichi Miyata, Yukiko Kaida, and Toshiyuki Murakami, “ν-ø̇-Coordinate-Based Power-Assist Control of Electric Wheelchair for a Caregiver. IEEE Transactions on Industrial Electronics, vol.55, no.6, pp 2517-2524, June 2008.
[12]Xiaoping Yun and Yoshio Yamamoto. “Internal Dynamics of a wheeled mobile robot. IEEE/RSJ International Conference on Intelligent Robots and Systems, vol.2 pp 1288-1294, July 1993.
[13]Robert Fuller. “Neural Fuzzy Systems. Abo Akademis tryckeri, Abo, ESF Series A:443 [ISBN 951-650-624-0, ISSN 0358-5654], 1995.
[14]Tuan Nghia Nguyen, Steven Su and Hung T Nguyen. “Neural Network Based Diagonal Decoupling Control of Powered Wheelchair Systems IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 22, no.2, pp 371-378, March 2014.
[15]Rong-Jong Wai. “Tracking Control Based on Neural Network Strategy for Robot Manipulator. Neurocomputing, vol. 51, pp 425-445, 2003.
[16]Pei-Chung Chen and Chi-Ruei Chen, “Design of Fuzzy-Neural Controller for BLDC Motor Drives. The 8th International Conference on Fuzzy Systems and Knowledge Discovery, Shanghai, China. pp. 852-856, July 2011.
[17]Faa-Jeng Lin and Po-Hung Shen. “Robust Fuzzy-Neural-Network Control for Two-Axis Motion Control System Based on TMS320C32 Control Computer. IEEE International Conference on Mechatronics, Taipei, Taiwan. July 10-12, 2005.
[18]Tamoghna Das and Indra Narayan Kar. “Design and Implementation of an Adaptive Fuzzy Logic-Based Controller for wheeled mobile robots. IEEE Transactions on control systems technology, vol. 14, no.3, pp 501-510. May 2006.
[19]Nguyen Tan Tien. “Fundamentals of Lyapunov Theory. Applied Nonlinear Control, 2003.
[20]PG Drives Technology. “S-Drive Scooter ControlSystem.
[21]Omron Corp. “Rotary Encoder E6B2-C.
[22]Fairchild Semiconductor. “High Speed Logic Gate Optocouplers. 2011.

連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top