本論文旨在開發一套旋翼機虛實整合系統之雛型，該系統包含一實體旋翼機，一電腦虛擬環境，以及協助整合兩者之萬向平台。實體旋翼機在做飛行控制時需要透過適應性權重調變擴展式卡爾曼濾波器(AWA-based EKF)估測出精準姿態方能透過控制器進行穩定控制，而所使用控制器為雙環PID結構，該結構分成內環和外環，外環使用P控制器，內環可選擇P、PI、PD或PID控制器，結構組合4種。本論文採用DNA演算法進行內環結構最佳化，並同時進行所有控制器參數最佳化。DNA演算法主要是針對實體旋翼機之重量、力臂、轉動慣量、升力參數及扭力參數所建構之動態物理模型進行控制器優化，得出最好的控制器結構與該結構之控制器參數。在虛實整合系統部分，關鍵是將實體飛行器的飛行數據回傳至虛擬環境進行擬真飛行，以姿態數據回傳而言，本論文提出兩種方案，一：直接使用旋翼機之姿態、二：使用萬向平台配備之無線磁角感測器模組。實驗結果顯示無線磁角感測器在120秒測試平均角度誤差為7.95度，旋翼機使用適應性權重調變擴展式卡爾曼濾波器做姿態估測平均角度誤差為9.02度。研究結果驗證了此萬向平台提供之無線磁角感測器在估測姿態上具有可行性且可適用各種旋翼機。最後，本論文提出的系統雛型為旋翼機之虛實整合系統設計提供良好之基礎。

This thesis develops the prototype of a quadcopter cyber-physical system (QCPS). A QCPS consists of a physical quadcopter, a computer virtual environment, and a universal platform that integrates the physical quadcopter and the virtual environment. An adaptive weighting adjustment-based extended Kalman filter (AWA-based EKF) was applied to estimate the precise attitude of the physical quadcopter for stable control through a controller with a dual-ring PID control scheme. The double-ring control scheme is composed of an inner ring and an outer ring, the outer ring uses P controller, and the inner ring can be selected from P, PI, PD or PID controllers, and thus there are 4 combinations to be chosen for the control scheme. In this thesis, the DNA algorithm was applied to optimize the selection of the control scheme for the inner and outer ring, and the parameters of each ring at the same time. In addition, the DNA algorithm is mainly used to optimize the controller for the dynamic physical model constructed by the weight, force arm, moment of inertia, lift parameters and torque parameters of the physical quadcopter. For the cyber-physical quadcopter system, the key is to return the flight data of the physical quadcopter to the virtual environment for immersive flight. Two approaches were used for collecting the attitude data: 1. the inertial sensor module of the quadcopter, and 2. the wireless magnetic angle sensor module equipped with the universal platform. The experimental results show that the average angular error of the wireless magnetic angle sensor is 7.95 degrees in 120 seconds, and the average angular error of the quadcopter attitude estimation generated by the adaptive weighting adjustment-based extended Kalman filter is 9.02 degrees. The results verify the feasibility of the attitude estimation for using the wireless magnetic angle sensor embedded with the universal platform, which can be applied to various quadcopters. Finally, the proposed system provides a good foundation for the design of quadcopter cyber-physical system.

[1] Y. Li and S. Shuxi, “A survey of control algorithms for quadrotor unmanned helicopter,” in Advanced Computational Intelligence(ICACI), 2012 IEEE Fifth International Conference, 2012, pp. 365-369.
[2] J. E. Bortz, “A new mathematical formulation for strapdown inertial navigation,” IEEE Transactions on Aerospace and Electronic Systems, vol. 1, no. 1, pp. 61-66, 1971.
[3] A. M. Sabatini, “Quaternion-based extended Kalman filter for determining orientation by inertial and magnetic sensing,” IEEE Transactions on Biomedical Engineering, vol. 53, no. 7, pp. 1346-1356, 2006.
[4] R. Mahony, T. Hamel, and J.-M. Pflimlin, “Nonlinear complementary filters on the special orthogonal group,” IEEE Transactions on Automatic Control, vol. 53, no. 5, pp. 1203-1218, 2008.
[5] S. O. H. Madgwick, A. J. L. Harrison, and R. Vaidyanathan, “Estimation of IMU and MARG orientation using a gradient descent algorithm,” in IEEE International Conference on Rehabilitation Robotics (ICORR), 2011, pp. 1-7.
[6] K. James, “Particle swarm optimization,” Encyclopedia of Machine Learning, pp.760-766, 2011.
[7] C. T. Wu, J. P. Tien, and T.-H. S. Li, “Integration of DNA and real coded GA for the design of PID-like fuzzy controllers,” in Proc. IEEE International Conference on Systems, Man, and Cybernetics, COEX, Seoul, Korea, Oct. Oct. 2012, pp. 2809-2814.
[8] E. Rashedi, H. Nezamabadi-pour, and S. Saryazdi, “GSA: A gravitational search algorithm,” Information Sciences, vol. 179, no. 13, pp. 2232-2248, 2009.
[9] G. Zhu and S. Kwong, “Gbest-guided artificial bee colony algorithm for numerical function optimization,” Applied Mathematics and Computation, vol. 217, pp. 3166-3173, 2010.
[10] J. G. B. Farias Filho, C. E. T. Do ́rea, W. M. Bessa, and J. L. C. B. Farias, “Modeling, test benches and identification of a quadcopter,” in Robotics Symposium and IV Brazilian Robotics Symposium(LARS/SBR), XIII Latin American, 2016, pp. 49-54.
[11] S. Alvarado, N. Certad, G. F. Lo ́pez, and M. Gonza ́lez, “Construction, identification and instrumentation of a low cost quadcopter research platform,” in Robotics Symposium (LARS) and 2017 Brazilian Symposium on Robotics (SBR), Latin American, 2017, pp. 1-6.
[12] D.Gheorghita ̌, I.Vintu, L. Mirea, and C. Bra ̌escu, “Quadcopter Control System,” in International Conference on System Theory on Control and Computing (ICSTCC), Romania, 2015, pp. 421-426.
[13] 莊維德，自主飛行四旋翼的控制與設計，國立成功大學航空太空工程學系碩士論文，2013。
[14] 鄭秉恩，混合慣性與光學感測技術之互動式手寫筆系統之研製，國立成功大學電機工程學系碩士論文，2012。
[15] M. D. Shuster, “A survey of attitude representations,” Navigation, vol. 8, no. 9, pp. 439-517, 1993.
[16] R. Pio, “Euler angle transformations,” IEEE Transactions on Automatic Control, vol. 11, no. 4, pp. 707-715, 1966.
[17] https://www.amebaiot.com/
[18] LSM9DS1 data sheet, ST Microelectronics.
[19] LPC11U35 data sheet, NXP Semiconductor.
[20] BQ24040 data sheet, Texas Instruments.
[21] PCA9685 data sheet, NXP Semiconductor.
[22] MS5611-01BA03 data sheet, TE Connectivity
[23] R. E. Kalman, “A new approach to linear filtering and prediction problems,” Journal of Basic Engineering, vol. 82, no. 1, pp. 35-45, 1960.
[24] MLX90316 data sheet, Triaxis.
[25] 陳冠翔，基於無線身體動作捕捉之籃球投籃訓練系統，國立成功大學電機工程學系碩士論文，2017。