本篇論文主旨在於提出新飛球系統的設計發明與模擬驗證，此飛球系統之球型的設計可提高操控靈活度，且驅動能量損耗可能低於傳統的輪形系統。本發明相較於先前研究所提出的球型機器人最大的優勢在於：三個軸向的致動器直接驅動三層球殼，使飛球的運動較能精準地控制。其機構主要包含內、中與外三層球殼以及連接球殼的框體結構，利用三個軸向互相垂直的致動器驅動各個球殼產生不同軸向的轉動，促使飛球能夠在二維平面自由地移動；此外，也利用重力產生的偏心力矩轉向，使飛球的轉向運動更加順暢。
在動態推導方面，本論文以陀螺與陀螺儀感測器的動態作為基礎，利用拉格朗日方程式推導飛球的二維運動運動方程式；由於陀螺儀效應會產生一迴轉力偶，需要利用第三個軸向的致動器抵消此迴轉力偶，使飛球系統能依據設定順暢地完成轉向運動。
在系統模擬方面，本論文建立兩種模型：模型一利用運動方程式於MATLAB/Simulink建立動態模型；模型二利用機構設計圖於ADAMS建立機構模型，模擬飛球一維及二維運動，並分析與比較兩種模型模擬結果的差異。在一維運動的模擬中，透過致動器輸入不同的扭矩，比較飛球軌跡、速度與加速度後，發現兩模型運動表現相似；在二維運動的模擬中，先利用ADAMS確立飛球的轉向機制，接著將飛球的二維運動分為四個型態：左轉、右轉、左迴轉與右迴轉，利用兩種軟體的模型模擬運動軌跡，發現兩者在左轉與右轉的型態時，運動表現相似，但是在左迴轉與右迴轉時，兩者會有明顯的差異，因此本論文提出了幾個可行的解決辦法，以提高飛球系統動態模型與機構模型模擬結果的等效度。未來將以動態模型為基礎，設計飛球系統的控制器，並架設機台及晶片程式的編寫。

The design and simulation study of Flyball System are proposed. The Flyball System which is designed in spherical shape might not only reduce the energy loss but also improve the sensitivity of operation. The most important advantage is that the Flyball System could be controlled accurately because three actuators actuate three shells directly. The Flyball System is able to operate on the planar via three different axes actuators which actuate different shells individually. Besides, the eccentric moment enables the Flyball system to steer smoothly.
In the aspect of dynamics, the equations of planar motion derivative is based on dynamics of gyro and gyroscope sensor via Lagrange equation. Owing to gyroscope effect, there is a couple which deserves to be eliminated by the third axis actuators. As a result, the Flyball system could achieve accurate steering motion.
In the view of simulations, there are two models built. The first model built in MATLAB/Simulink is based on the equations of motion. The others built in ADAMS is based on the mechanism design. The simulations of x-axis motion compare the trajectory, velocity and acceleration according to different torque inputs. In simulations of planar motion, the study confirms the steering mechanism. Also, the planar motion is divided into four types which are left turn, right turn, left U-turn and right U-turn. The simulation results of left turn and right turn are similar in ADAMS and MATLAB/Simulink. However, the simulation results of left U-turn and right U-turn are different. The study proposes feasible solutions to the problem in order to improve the equivalence between two models. In the future, the study will design controllers according to the equations of motion and implement the prototype.

1. B. Cazzolato, J. Harvey, C. Dyer, K. Fulton, E. Schumann, T. Zhu, Z. Prime, B. Davis, S. Hart, E. Pearce, and J. Atterton. Modeling, simulation and control of an electric diwheel. in Australasian Conference on Robotics and Automation. 2011. Melbourne, VIC, Australia: Australian Robotics and Automation Association Inc.
2. H.B. Brown, Jr. and X. Yangsheng. A single-wheel, gyroscopically stabilized robot. in IEEE International Conference on Robotics and Automation. 1996. New York, NY, USA: IEEE.
3. G.C. Nandy and X. Yangsheng. Dynamic model of a gyroscopic wheel. in IEEE International Conference on Robotics and Automation. 1998. New York, NY, USA: IEEE.
4. S.J. Tsai, E.D. Ferreira, and C.J.J. Paredis. Control of the gyrover. in IEEE/RSJ International Conference on Intelligent Robots and Systems: Human and Environment Friendly Robots whith High Intelligence and Emotional Quotients. 1999. Kyongju, South Korea: IEEE.
5. A. Halme, T. Schonberg, and W. Yan. Motion control of a spherical mobile robot. in IEEE International Workshop on Advanced Motion Control. 1996. New York, NY, USA: IEEE.
6. A. Bicchi, A. Balluchi, D. Prattichizzo, and A. Gorelli. Introducing the `SPHERICLE': An experimental testbed for research and teaching in nonholonomy. in IEEE International Conference on Robotics and Automation. 1997. Albuquerque, NM, USA: IEEE.
7. S. Bhattacharya and S.K. Agrawal, Spherical rolling robot: A design and motion planning studies. IEEE Transactions on Robotics and Automation, 2000. 16(6): p. 835-839.
8. A. Amir Homayoun Javadi and P. Mojabi. Introducing August: a novel strategy for an omnidirectional spherical rolling robot. 2002. Piscataway, NJ, USA: IEEE.
9. G. Shu, Q. Zhan, and Y. Cai. Motion control of spherical robot based on conservation of angular momentum. in IEEE International Conference on Mechatronics and Automation. 2009. Changchun, China: IEEE Computer Society.
10. T. Urakubo, M. Osawa, H. Tamaki, Y. Tada, and S. Maekawa. Development of a spherical rolling robot equipped with a gyro. in IEEE International Conference on Mechatronics and Automation. 2012. Chengdu, China: IEEE Computer Society.
11. C.E. Thorne and M. Yim, Design and analysis of a gyroscopically controlled micro air vehicle. Journal of Intelligent & Robotic Systems, 2012. 65(1-4): p. 417-35.
12. P.F. Beer, E.R. Johnston, Jr., and W.E. Clausen, Vector Mechanics for Engineers: Dynamics. Eighth ed. 2007: McGraw-Hill Companies.
13. J.L. Meriam and L.G. Kraige, Engineering Mechanics - Dynamics. Sixth ed. 2008: John Wiley & Sons, Inc.