保全機器人的自主監視時常會因為各種不穩定的因素而受到嚴重限制，如光線變化、視角變化、遮擋，或目標的外觀變化。本論文提出了一種基於深度學習的自主機器人系統，以執行視覺感知和控制任務。視覺感知旨在識別場景中移動的所有對象，透過人臉辨識演算法，來判斷目標是否為特定對象，若為允許進入場景的目標則追蹤人臉，讓對象保持在中心；而若為陌生的目標，則發送警訊到手機。控制系統包括運動控制和導航，其中機器人使用了ROS環境開發及運作，透過RP-LiDAR及Hector SLAM演算法來達到路徑規劃及導航避障，並同時達到機器人的定位與建圖。經驗驗證包括各種感知算法的評估指標，如模型的準確性、召回率、PR曲線、ROC曲線等，而建圖準確性的評估指標為誤差的均方根值。實驗結果表明，VggNet在4種不同光照變化下的平均精度為0.95，而對於真實場景中建圖的準確性，建圖誤差的均方根值為0.222公尺。

Conventional surveillance for a security robot suffers from severe limitations, perceptual aliasing (e.g., different places/objects can appear identical), occlusion (e.g., place/object appearance changes between visits), illumination changes, significant viewpoint changes, etc. This thesis proposes an autonomous robotic system based on CNN (Convolutional Neural Network) to perform visual perception and control tasks. The visual perception aims to identify all objects moving in the scene and to verify whether the target is an authorized person. The visual perception system includes a motion detection module, a tracking module, face detection and recognition module. The control system includes motion control and navigation (path-planning & obstacle avoidance). The empirical validation includes the evaluation metrics, such as model speed, accuracy, precision, recall, ROC (Receiver Operating Characteristic) curve, P-R (Precision-Recall) curve, F1-score for AlexNet, VggNet, and GoogLeNet, and RMSE (Root Mean Square Error) value of mapping error. The experimental results showed that the average accuracy rate of VggNet under 4 different illumination changes is 0.95, and it has the best performance under all unstable factors among three CNN architectures. For the accuracy of building map in real scene, the mapping error is 0.222 meters.

[1] D. Tharindu, A. Pradeep, “Design and Implementation of an Autonomous Indoor Surveillance Robot based on Raspberry Pi,” 2019 International Conference on Advancements in Computing, Malabe, Sri Lanka, Dec. 2019.
[2] A. Carlos, C. Jonathan, B. Ramon, “Detecting Objects for Indoor Monitoring and Surveillance for Mobile Robots,” 2014 Fifth International Conference on Emerging Security Technologies, Alcala de Henares, Spain, Sep. 2014.
[3] V. Ratna Kumari, P. Siva Sanjay, “Smart Surveillance Robot using Object Detection,” 2020 International Conference on Communication and Signal Processing, Chennai, India, Jul. 2020.
[4] M. Shubham, R. Jaynendra Kumar, “Wadoro: An autonomous mobile robot for surveillance,” 2016 IEEE 1st International Conference on Power Electronics, Intelligent Control and Energy Systems, Delhi, India, Jul. 2016.
[5] P. Roman, P. R. Rui, S. C. Micael, “People Detection by Mobile Robots Doing Automatic Guard Patrols,” 2020 IEEE International Conference on Autonomous Robot Systems and Competitions, Ponta Delgada, Portugal, Apr. 2020.
[6] K. Jun, C. Min, J. Min, S. Jinhua, and H. Jian, “Face Recognition Based on CSGF (2D) 2 PCANet, ” IEEE Access, vol. 6, no. 1, pp. 45153-45165, Aug. 2018.
[7] Y. Xi, L. Xiaoming, “Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition,” IEEE Transactions on Image Processing, vol. 27, no. 2, pp. 964-975, Oct. 2017.
[8] D. Jia, P. Gaoyang, Z. Zhiyu, P. Zhibo, Y. Huayong, "cGAN Based Facial Expression Recognition for Human-Robot Interaction,” IEEE Access, vol. 7, no. 1, pp. 9848-9859, Jan. 2019.
[9] X. Zheng, T. Hengliang, Y. Wenling, “The Excellent Properties of a Dense Grid-Based HOG Feature on Face Recognition Compared to Gabor and LBP,” IEEE Access, vol. 6, no. 1, pp. 29306-29319, Mar. 2018.
[10] L. Fan, D. Yuhua, X. Feng, Y. Qiaolin, “Learning Low-Rank Regularized Generic Representation With Block-Sparse Structure for Single Sample Face Recognition,” IEEE Access, vol. 7, no. 1, pp. 30573-30587, Mar. 2019.
[11] W. Chenglin, L. Tianhong, Z. Lijun, T. Yunchao, Z. Xiangjun, “Window Zooming–Based Localization Algorithm of Fruit and Vegetable for Harvesting Robot,” IEEE Access, vol. 7, no. 1, pp. 103639-103649, Jul. 2019.
[12] H. Changhui, W. Fei, Y. Jian, J. Xiaoyuan, L. Xiaobo, L. Pan, “Diagonal Symmetric Pattern-Based Illumination Invariant Measure for Severe Illumination Variation Face Recognition,” IEEE Access, vol. 8, no. 1, pp. 63202-63213, Mar. 2020.
[13] Z. Liyun, G. Yepeng, “Improvement of illumination-insensitive features for face recognition under complex illumination conditions,” The Journal of Engineering, vol. 18, no. 12, pp. 1947-1953, Dec. 2018.
[14] H. Chang-Hui, L. Xiao-Bo, L. Pan, J. Xiao-Yuan, Y. Dong, “Single Sample Face Recognition Under Varying Illumination via QRCP Decomposition,” IEEE Transactions on Image Processing, vol. 28, no. 5, pp. 2624-2638, Dec. 2018.
[15] D. Changxing, T. Dacheng, “Trunk-Branch Ensemble Convolutional Neural Networks for Video-Based Face Recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 4, pp. 1002-1014, May 2017.
[16] Z. Ting, W. Hong, D. Qiulei, “Deep Disentangling Siamese Network for Frontal Face Synthesis Under Neutral Illumination,” IEEE Signal Processing Letters, vol. 25, no. 9, pp. 1344-1348, Jul. 2018.
[17] Y. Freund, R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting,” Journal of Computer and System Sciences, vol. 55, no. 1, pp. 119-139, Aug. 1997.
[18] P. Viola, M. Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, HI, USA, Dec. 2001.
[19] R. Lienhart, J. Maydt, “An Extended Set of Haar-like Features for Rapid Object Detection,” International Conference on Image Processing, Rochester, NY, USA, Sep. 2002.
[20] R. Cristin, J. P. Ananth, V. C. Raj, “Illumination-based texture descriptor and fruitfly support vector neural network for image forgery detection in face images,” IET Image Processing, vol. 12, no. 1, pp. 1439-1449, Aug. 2018.
[21] F. N. Ibrahim, Z. M. Zin, N. Ibrahim, “Eye Center Detection Using Combined Viola-Jones and Neural Network Algorithms,” 2018 International Symposium on Agent, Multi-Agent Systems and Robotics, Putrajaya, Malaysia, Aug. 2018.
[22] M. A. Uaday, M. N. Shuzan, S. Shanewaze, R. Rakib, H. U. Zaman, “The Design of a Novel Multi-Purpose Fire Fighting Robot with Video Streaming Capability,” 2019 IEEE 5th International Conference for Convergence in Technology, Bombay, India, Mar. 2019.
[23] A. Talipu, A. Generosi, M. Mengoni, L. Giraldi, “Evaluation of Deep Convolutional Neural Network architectures for Emotion Recognition in the Wild,” 2019 IEEE 23rd International Symposium on Consumer Technologies, Ancona, Italy, Jun. 2019.
[24] K. Rasanayagam, S. D. D. C Kumarasiri, W. A. D. D. Tharuka, N. T. Samaranayake, P. Samarasinghe, S. E. R. Siriwardana, “CIS: An Automated Criminal Identification System,” 2018 IEEE International Conference on Information and Automation for Sustainability, Colombo, Sri Lanka, Dec. 2018.
[25] R. Dhaouadi, A. A. Hatab, “Dynamic Modelling of Differential-Drive Mobile Robots using Lagrange and Newton-Euler Methodologies: A Unified Framework,” Advances in Robotics and Automation, vol. 2, no. 2, pp. 107-113, Jan. 2013.
[26] C. Jakub, J. Anna, K. Tomas, “Modeling and Control of Mobile Robot with Differential Chassis,” Proceedings of the Faculty of Electrical Engineering and Informatics of the Technical University of Košice, Košice, Slovak Republic, Sep. 2015.
[27] J. Binqian, Z. Yilong, L. Ming, “A Triangle Feature Based Map-to-map Matching and Loop Closure for 2D Graph SLAM,” 2019 IEEE International Conference on Robotics and Biomimetics, Dali, China, Dec. 2019.
[28] C. Chang, Z. Hua, W. Lei, L. Yu, “A Stereo Visual-Inertial SLAM Approach for Indoor Mobile Robots in Unknown Environments Without Occlusions,” IEEE Access, vol. 7, no. 1, pp. 185408-185421, Dec. 2019.
[29] S. Weizhao, V. Srinivasan, L. Cong, K. George, “Stereo Visual Inertial LiDAR Simultaneous Localization and Mapping,” 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, Macau, China, Nov. 2019.
[30] S. Kohlbrecher, O. V. Stryk, J. Meyer, U. Klingauf, “A flexible and scalable SLAM system with full 3D motion estimation,” 2011 IEEE International Symposium on Safety, Security, and Rescue Robotics, Kyoto, Japan, Nov. 2011.
[31] O. F. G. Palacios, S. H. Salah, “Mapping marsian caves in 2D with a small exploratory robot,” 2017 IEEE XXIV International Conference on Electronics, Electrical Engineering and Computing, Cusco, Peru, Aug. 2017.
[32] C. Xieyuanli, Z. Hui, L. Huimin, X. Junhao, Q. Qihang, L. Yi, “Robust SLAM system based on monocular vision and LiDAR for robotic urban search and rescue,” 2017 IEEE International Symposium on Safety, Security and Rescue Robotics, Shanghai, China, Oct. 2017.
[33] W. Di, M. Yu, Z. Kai, M. Fei, “A LiDAR SLAM Based on Point-Line Features for Underground Mining Vehicle,” 2018 Chinese Automation Congress, Xi'an, China, Nov. 2018.
[34] Z. Xuexi, L. Guokun, F. Genping, X. Dongliang, L. Shiliu, “SLAM Algorithm Analysis of Mobile Robot Based on LiDAR,” 2019 Chinese Control Conference, Guangzhou, China, Jul. 2019.
[35] W. Weichen, B. Shirinzadeh, S. Esakkiappan, M. Ghafarian, A. Al-Jodah, “Orientation Correction for Hector SLAM at Starting Stage,” 2019 7th International Conference on Robot Intelligence Technology and Applications, Daejeon, South Korea, Nov. 2019.
[36] D. Fox, W. Burgard, S. Thrun, “The dynamic window approach to collision avoidance,” IEEE Robotics & Automation Magazine, vol. 4, no. 1, pp. 23-33, Mar. 1997.
[37] L. Wei, X. Rong, “Dynamical Obstacle Avoidance of Task- Constrained Mobile Manipulation Using Model Predictive Control,” IEEE Access, vol. 7, no. 1, pp. 88301-88311, Jul. 2019.
[38] Z. Yi, X. ZhiCheng, Y. XueXi, L. ShuXiu, L. ShenTao, “Obstacle Avoidance of Two-Wheeled Mobile Robot based on DWA Algorithm,” 2019 Chinese Automation Congress, Hangzhou, China, Nov. 2019.
[39] D. Jeong Chan, O. Ji Man, L. Wang Heon, “A License Plate Recognition using Neural Network and Autonomous Mobile Robot in Intelligent Parking Lot Management System,” 2018 18th International Conference on Control, Automation and Systems, Pyeongchang, South Korea, Oct. 2018.
[40] H. Hong, W. Siyang, M. Shichao, “Research on Object Location Detection based on ROS Machine Vision,” 2019 Chinese Control And Decision Conference, Nanchang, China, Jun. 2019.