未來智慧城市中，在非直視信號（NLOS）情況下，透過無線感測網路（WSNs）實現基於隨機 H_∞ 非直視線容忍無線定位濾波器的移動機器人參考路徑追蹤控制，對於智慧建築非常有用。本研究採用平滑信號模型，將NLOS 所產生的偏差信號嵌入移動機器人的隨機動態中，以防止NLOS對移動機器人的定位和姿態（位置和方向）控制造成損壞，以實現透過WSN進行移動機器人期望參考路徑追蹤。此外，本研究提出了一種基於隨機 H_∞ 機器學習的非直視線容忍無線定位濾波器參考路徑追蹤控制方案，以有效估計和控制移動機器人的姿態，減弱非直視線情況、外部干擾和測量噪聲的影響。為了避免在設計過程中解決高度複雜的濾波器/控制器耦合的Hamilton-Jacobi-Isaacs方程式（HJIEs），本研究提出了一種基於強化學習的深度神經網路（DNN）濾波器/控制器耦合的HJIE方法，用於移動機器人，以直接解決濾波器/控制器耦合的HJIE，實現基於隨機 H_∞ 非直視信號容忍定位濾波器在WSN對於移動機器人參考路徑追蹤控制。在訓練DNN時，我們利用移動機器人的系統模型以及濾波器/控制器耦合的HJIE，並使用濾波器/控制器耦合的HJIE強化Adam學習算法能減少訓練數據量。當Adam學習算法收斂時，所提出的濾波器/控制器耦合的HJIE強化DNN基於 H_∞ 非直視信號容忍無線定位濾波器的輸出反饋參考路徑追蹤控制方案能夠接近在智慧城市中透過WSN對移動機器人進行隨機 H_∞ 非直視信號容忍無線定位濾波器的參考路徑追蹤控制策略。最後，我們提供了兩個例子來說明設計過程並驗證所提方法的有效性：（i）在智慧建築中，移動機器人追蹤所需路徑；（ii）在WSN中，移動機器人根據快速探索隨機星樹（〖RRT〗^*）算法生成的可行路徑來避開障礙物。這兩個例子展示了所提方法的設計過程並驗證了其有效性。

關鍵詞：移動機器人、非直視信號、平滑信號模型、HJIE、非直視線容忍定位濾波器、深度神經網路、快速探索隨機星樹、無線感測網路。

The stochastic H∞ NLOS-tolerant/mitigation wireless localization filter-based reference path tracking control of mobile robot/Autonomous Guide Robot (AGR) via wireless sensor networks (WSNs) under the non-line-of-sight (NLOS) situation is useful in the intelligent buildings in future smart cities. In this study, a smooth signal model is employed to insert the fault signals due to NLOS in the stochastic dynamic of AGR to prevent the corruption of NLOS on the localization and control of the pose (position and orientation) of AGR for the desired reference path tracking of the AGR. Moreover, the stochastic H∞ Machine Learning-based NLOS-mitigation wireless localization filter-based reference path tracking control scheme is proposed to effectively estimate and control the pose of AGR and to significantly attenuate the effect from NLOS situation, the external disturbance and measurement noise. To avoid solving highly complex filter/controller-coupled Hamilton Jacobi Isaacs equation (HJIEs) for the nonlinear wireless localization filter and controller in the design procedure, a filter/controller-coupled HJIE-reinforcement learning-based deep neural network (DNN) is proposed for AGR to directly solve filter/controller-coupled HJIE for the stochastic H∞ NLOS-mitigation wireless localization filter-based reference path tracking control of the AGR. The system model of the AGR and filter/controller-coupled HJIE have been employed for training DNN via the filter/controller-coupled HJIE-reinforcement Adam learning algorithm to significantly reduce the training data. When the Adam learning algorithm converges, the proposed filter/controller-coupled HJIE-reinforcement DNN-based H∞ NLOS-mitigation wireless localization filter-based output feedback reference path tracking control scheme can approach the theoretical stochastic H∞ NLOS-mitigation wireless localization filter-based reference path tracking control strategy of AGR in a smart city. Finally, two examples, i.e., (i) an AGR to track a desired path in the smart building and (ii) an AGR to track a feasible path generated by rapidly exploring random star tree (RRT^*) algorithm to avoid obstacles in WSNs, are given to illustrate the design procedure and validate the effectiveness of the proposed methods.

Keywords: Autonomous Guide Robot (AGR), non-line-of-sight (NLOS), smoothing signal model, Hamilton-Jacobi Isaacs equation (HJIE), NLOS-tolerant localization filter, deep neural network (DNN), rapidly exploring random star tree (RRT^*), wireless sensor networks (WSNs)

摘要 I
Abstract II
致謝 III
Content IV
I. INTRODUCTION 1
II. WSN-BASED LOCALIZATION SYSTEM AND REFERENCE TRACKING SYSTEM OF VEHICLE UNDER NLOS SITUATIONS 4
III. HJIE-REINFORCEMENT LEARNING DNN-BASED H_∞ NLOS- MITIGATION WIRELESS LOCALIZATION FILTER-BASED REFERENCE TRACKING CONTROL DESIGN OF AGR IN WIRELESS SENSOR NETWORKS 15
IV. SIMULATION RESULTS 22
A. EXAMPLE 1 24
B. EXAMPLE 2 29
V. CONCLUSION 30
APPENDIX A 32
APPENDIX B 35
REFERENCES 37

[1] E. Xu, Z. Ding and S. Dasgupta, “Target tracking and mobile sensor navigation in wireless sensor networks,” IEEE Trans. Mobile
Comput., vol. 12, no. 1, pp. 177-186, Jan. 2013.
[2] W.-Y. Chiu and B.-S .Chen, “Mobile location estimation in urban areas using mixed Manhattan/Euclidean norm and convex
optimization,” IEEE Trans. Wireless Commun., vol. 8, no. 1, pp. 414-423, Jan. 2009.
[3] A. C. Paredes, M. Malfaz, and M. A. Salichs, “Signage system for the navigation of autonomous robots in indoor environments,” IEEE
Trans. Ind. Inform., vol. 10, no. 1, pp. 680-688, Feb. 2014.
[4] H. Liu, H. Darabi, P. Banerjee, and J. Liu, “Survey of wireless indoor positioning technique and systems,” IEEE Trans. Syst. Man
Cybern., vol. 37, no. 6, pp. 1067-1080, Nov. 2007.
[5] K. Derr and M. Manic, “Wireless sensor networks-node location for various industry problems,” IEEE Trans. Ind. Inform., vol. 11, no.
3, pp. 752-762, Jun. 2015.
[6] B.-F. Wu, C.-L. Jen, “Particle-filter-based radio localization for mobile robot in the environments with low-density wlan aps,” IEEE
Trans. Ind. Electron., vol. 64, no. 3, pp. 1099-1104, Aug. 2011.
[7] P. Yang and W. Wu, “Efficient particle filter location algorithm in dense passive RFID tag environment,” IEEE Trans. Ind. Electron.,
vol. 61, no. 10, pp. 5641-5651, Nov. 2014.
[8] J. M. Pak, C. K. Ahn, Y. S. Shmaliy and M. T. Lim, “Improving reliability of particle filter-based localization in wireless sensor
networks via hybrid particle/FIR filtering,” IEEE Trans. Ind. Inform., vol. 11, no. 5, pp. 1089-1098, Oct. 2015.
[9] I. Ullah, Y. Shen, X. Su, C. Esposito and C. Choi, “A localization based on unscented Kalman filter and particle filter localization
algorithms,” IEEE Access, vol. 8, pp. 2233-2246, Dec. 2020.
[10] S. Marano W. M. Gifford, H. Wymeersch and M. Z. Win, “NLOS identification and mitigation for localization based on UWB `
experimental data,” IEEE J. Sel. Areas Commun., vol. 28, no. 7, pp. 1026-1035, Sept. 2010.
[11] ˙I. Guvenc¸ and C. Chong, “A survey on TOA based wireless localization and NLOS mitigation techniques,” ¨ IEEE Commun. Surv. Tutor.,
vol. 11, no. 3, pp. 107-124, Aug. 2009.
[12] J.-F. Liao and B.-S. Chen, “ Robust mobile location estimator with NLOS mitigation using interactive multiple model algorithm,” IEEE
Tran. Wireless Commun., vol. 5, pp. 3002-3006, Nov. 2006.
[13] B.-S. Chen, K.-C. Yang and M.-Y. Lee, “Robust H∞ NLOS- tolerant localization filter and NLOS-tolerant remote reference tracking
control of mobile robot in wireless sensor networks,” IEEE Access, vol. 9, pp. 164801-164819, Dec. 2021.
[14] S. Sakagami, S. Aoyama, K. Kuboi, S. Shirota and A. Akeyama, “Vehicle position estimates by multibeam antennas in multipath
environments,” IEEE Trans. Veh. Technol., vol. 41, no. 1, pp. 63-68, Feb. 1992.
[15] R. Sumathi and R. Srinivasan, “RSS-based location estimation in mobility assisted wireless sensor networks,” Proceedings of the 6th
IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems, Sept. 2011.
[16] W. Cui, B. Li, L. Zhang and W. Meng, “Robust mobile location estimation in NLOS environment using GMM, IMM, and EKF,” IEEE
Syst. J., vol. 13, no. 3, pp. 3490-3500, Sept. 2019.
[17] F. Gustafsson et al., “Particle filters for positioning, navigation, and tracking,” IEEE Trans. Signal Process., vol. 50, no. 2, pp. 425-437,
Feb. 2002.
[18] S. Yin and X. Zhu, “Intelligent particle filter and its application on fault detection of nonlinear system,” IEEE Trans. Ind. Electron.,
vol. 62, no. 6, pp. 3852-3861, Jun. 2015.
[19] B.-S. Chen, M.-Y. Lee and X.-H. Chen, “Security-enhanced filter design for stochastic systems under malicious attack via smoothed
signal model and multiobjective estimation method, ” IEEE Trans. Signal Process., vol. 68, pp. 4971-4986, Aug. 2020.
[20] C.-S. Tseng, B.-S. Chen and H.-J. Uang, “Fuzzy tracking control design for nonlinear dynamic system via T-S fuzzy model,” IEEE
Trans. Fuzzy Syst., vol. 9, no. 3, pp. 381-392, Jun. 2001.
[21] B.-S. Chen, B.-K. Lee and L.-B. Guo, “Optimal tracking design for stochastic fuzzy system,” IEEE Trans. Fuzzy Syst., vol. 11, no. 6,
pp. 796-813, Dec. 2003.
[22] G. Feng, “A survey on analysis and design of model-based fuzzy control system, ” IEEE Trans. Fuzzy Syst., vol. 14, no. 5, pp. 676-697,
Oct. 2006.
27
[23] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” in Proc. ICLR, San Diego, CA, USA, pp. 1-15, Dec. 2014.
[24] A. L. Maas, P. Qi, Z. Xie, A. Y. Hannun, C. T. Lengerich, D. Jurafsky, and A. Y. Ng, “Buliding DNN acoustic models for large
vocabulary speech recognition,” Comput. Speech Lang., vol. 41, pp.195-213, Jan. 2017.
[25] A. Amjad, L. Khan, N. Ashraf, M. B. Mahmood and H.-T. Chang, “Recognizing semi-natural and spontaneous speech emotions using
deep neural networks,” IEEE Access, vol. 10, pp. 37149-37163, Mar. 2022.
[26] D. Cires¸an, U. Meier, and J. Schmidhuber, “Multi-column deep neural networks for image classification ,” in Proc. IEEE Conf. Comput.
Vis. Pattern Recognit., Jun. 2012, pp. 3642-3649.
[27] Y. Dong and G. Lian, “Research and discussion on image recognition and classification algorithm based on deep learning,” in Proc.
Int. Conf. Mach. Learn., Big Data Bus. Intell. (MLBDBI), Nov. 2019, pp. 274-278.
[28] X. Lu, H. Wang, and M. Li, “Kalman fixed-interval and fixed-lag smoothing for wireless sensor systems with multiplicative noises,”
in Proc. 012 24th Control and Decision Conference, Chinese, pp. 3023-3026, May 2012.
[29] K. A. Hoo and J. C. Kantor, “Global linearization and control of a mixed-culture bioreactor with competition and external inhibition,”
Math. Bioscience, vol, 82, no. 1, pp. 43-62, Nov. 1986.
[30] B.-S. Chen, W.-H. Chen and H.-L. Wu, “Robust H2/H∞ global linearization filter design for nonlinear stochastic system,” IEEE Trans.
Circuits Syst. I: Regul. Pap., vol. 56, no. 7, pp. 1441-1454, Jul. 2009.
[31] B.-S. Chen and C.-F. Wu, “Robust scheduling filter design for a class of nonlinear stochastic Poisson signal system,” IEEE Trans.
Signal Process., vol. 63, no. 23, pp. 6245-6257, Dec. 2015.
[32] K. Jouili, H. Jerbi, “A computationally Lyapunov nonlinear gain scheduling control of nonlinear systems with stability guarantees,”
ICCMS, vol. 12, pp. 1131-1137, Oct. 2014.
[33] Y. N. Dauphin, H. D. Vries, J. Chung, and Y. Bengio, “RMSProp and equilibrated adaptive learning rates for non-convex optimization,”
Mach. Learn., vol. 1, pp. 1504-1512, Feb. 2015.
[34] H. Zhang, Y. Wang, J. Zheng and J. Yu, “Path planning of industrial robot based on improved RRT algorithm in complex environments,”
IEEE Access, vol. 6, pp. 53296-53306, Sept. 2018.
[35] J. Qi, H. Yang and H. Sun, “MOD-RRT∗
: A sampling-based algorithm for robot path planning in dynamic environment,” IEEE Trans.
Ind. Elect., vol. 68, no. 8, pp. 7244-7254, Jun. 2020.
[36] Y. Li, R. Jin, X. Xu, Y. Qian, H. Wang, S. Xu and Z. Wang, “A mobile robot path planning algorithm based on improved A∗
algorithm and dynamic window approach,” IEEE Access, vol. 10, pp. 57736 - 57747, May. 2022