目前市面上已有幾種自動餵食系統，針對不同的食物種類及任務提出相對應的機構設計，但大多數餵食系統的機構運動自由度過低導致食物的種類被限制，也常造成機器運行不穩定狀況。本研究提出一套基於自製六軸手臂的餵食系統，為了解決餵食動作的姿態及多方向撈取運動路徑，本研究將設計一機械手臂，手臂的接頭組成相異於一般工業用機械手臂；另外搭配容器形狀及韌體的設計達成穩定取食之效果。
機械手臂的開發考量了精度、負重以及其工作範圍，在餵食系統有限的空間及重量限制下，機械手臂的設計將受到限制，例如:連桿長度、馬達空間配置、元件的選用等，為了解決以上問題，第一，將進行手臂的自由度分析，去提出一套符合餵食系統配置之手臂設計，並依餵食系統中各任務所需之姿態去進行多次模擬及測試，得出具效率又能讓結構穩定之自由度設計。有了桿長度限制後，藉由CAD姿態模擬手臂極限與任務所需之為至合理性，並界定一滿足於系統配置所需的工作範圍；第二，決定餵食機器人之目標性能後，計算手臂關節負載分析以及挑選適合的致動元件；第三，最後推導手臂運動學並反向驗證其運動，控制手臂空間位置之轉移。
容器韌體開發包含了因應食物的多樣化，試圖得出容器之最佳形狀設計，並搭配勺具的幾何形狀，達到最穩定之取食成效，本研究將進行餵食之撈取動作測試，將容器之幾何形狀最佳化，並探討適合餵食系統之容器容量及數量。
為了配合智慧餵食所需的眼球辨識模組，有了各容器大小及數量，去進行容器的底盤設計，底盤的機構將執行非均一行程之直線運動，本研究提出一機構設計理論能搭載一致動器達成以上需求，除了結構上的穩定，亦可利用最小空間去達成機構配置。

There are already a few self-feeding machines on the market and varied mechanism designs are proposed for different food types or tasks. The degree of freedom (DOF) of the mechanism in self-feeding machines is usually too low, which causes the variety of foods to be restricted, or the instability when conducting some feeding motions. This study proposes a six-axis robot arm based on feeding machine and system control strategies. In order to solve the feeding action posture and multi-direction scooping path, the joint composition of the mechanical arm is different from that of the ordinary industrial robot arm; in addition, it is matched with the shape of the container and the firmware. Designed to achieve a stable feeding effect.
The importance of developing a robot arm includes accuracy, speed and working range. Under the limited space and weight limitation of the feeding system, the design of the robot arm will be restricted, for example: the length of the connecting linkages, the space configuration of the motor, the selection of composition, etc. To solve these problems, first of all, will carry out the joint load analysis of the arm to propose a set of mechanism design, in addition to selecting suitable actuator components, and performing multiple tests according to the speed required for each task in the feeding system, to achieve both efficiency and stable control methods. Secondly, under the condition of the link length limitation, the path generated by each task is analyzed by CAD dynamic simulation, and a working space that satisfies the system configuration is defined, and finally deduced. To derive arm inverse kinematics and take verification of its movement, then to control the movement of the arm's position and gesture.
The development of container firmware takes the diversity of foods in to account, trying to get the best shape design of the container, and matching the geometry of the spoon to achieve the most stable feeding effect; also, in order to match the eyeball identification module required for intelligent feeding, the chassis design of each container will perform a linear motion of non-uniform stroke. This study proposes that a mechanism design theory can be equipped with an actuator to achieve the above requirements. In addition to structural stability, the minimum space can be used to achieve the mechanism configuration.

參考文獻
1. 陳幃盛, 六軸機械手臂控制器之設計與研製, in 光機電暨材料研究所. 2011, 吳鳳科技大學: 嘉義縣. p. 81.
2. Naotunna, I., et al. Meal assistance robots: A review on current status, challenges and future directions. in 2015 IEEE/SICE International Symposium on System Integration (SII). 2015.
3. Co., S. Snom. 2015; Available from: http://www.secom.co.jp/.
4. Perera, C.J., T.D. Lalitharatne, and K. Kiguchi. EEG-controlled meal assistance robot with camera-based automatic mouth position tracking and mouth open detection. in 2017 IEEE International Conference on Robotics and Automation (ICRA). 2017.
5. Song, W.-K. and J. Kim, Novel Assistive Robot for Self-Feeding. 2012.
6. 林晉茂, 機器人學. 民國89年, 國立編譯館.
7. 和田週平、木下元一郎著, 機器人工程學. 民國72年, 臺隆書店.
8. Mealtime Partners. 2014; Available from: http://www.mealtimepartners.com/.
9. 林士傑, 以視覺為基礎之餵食機器人, in 資訊工程研究所. 2007, 國立中央大學: 桃園縣. p. 70.