此篇論文貢獻為整合Robot Operating System (ROS)及深度學習實作一個具有同步定位及地圖構建、自主導航、安全護具檢測的機器人系統，目的為降低工安意外發生的機率。利用ROS系統可將機體與程式軟體分離設計的優點，本研究開發出的監視系統並不僅侷限於在一種機體上使用，對於任何能安裝ROS系統之都能使用，除了導航功能外，機器人透過雙目攝影機及雷射測距儀等感測器得以判斷自己現在身處的環境而並非單純繪製地圖，因此可及時避開周邊的障礙物，且運用ROS系統之使用者操作介面，能讓人更容易地對機器人下達指令；更因其高準確率的監視機制及機器人的移動靈活性，本研究的系統很適合在施工環境中做為工人防護具檢測。以實際環境探討各種地圖構建演算法及深度學習辨識模型的準確性與模型檢測速度，並呈現各種演算法與機器人系統結合的成果。

For the construction industry, the fatal injury rate is higher than the average for all industries. In the construction industry, it is a fact that researchers have an increased interest in occupational safety recently. However, all the current methods using machine learning with stationary cameras suffer from some severe limitations. These limitations include perceptual aliasing, occlusion, and significant viewpoint changes.
This paper proposes a perception module. Based on a differential-drive mobile robot, this module used deep-learning models and laser SLAM (Simultaneous Localization and Mapping) for object recognition and navigation. This paper validated the navigation strategies and various deep-learning frameworks by particular evaluation metrics to select the best model. For object detection models, these metrics evaluated the deep learning model’s predictions, then generated each model’s accuracy and test time. These criteria include Recall, P-R curve, F1 score, mean average precision, and Frame per second (FPS).
The experimental results demonstrate that the YOLOv3 shows the best trade-off among all algorithms. During real-time detection, this model can process 45 frames per second (FPS), and 57.9% mean average precision (mAP) on the NVIDIA Jetson TX2 device. Thus, the YOLOv3 is suitable for real-time detection and the right candidate for deploying the neural network on a mobile robot. For the comparison of laser-based SLAM, the evaluation metrics used Root Mean Square Error (RMSE) to test the accuracy of maps. Then according to the result, the Google Cartographer SLAM shows the lowest RMSE and adequate processing time. In conclusion, this module can effectively detect construction worker’s non-hardhat-use in different construction site conditions and facilitate improved safety inspection and supervision.

[1] M. Park, S. Park, M. Song, and S. Park, "IoT-based Safety Recognition Service for Construction Site," in 2019 Eleventh International Conference on Ubiquitous and Future Networks (ICUFN), 2019, pp. 738-741.
[2] A. H. M. Rubaiyat et al., "Automatic Detection of Helmet Uses for Construction Safety," in 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops (WIW), 2016, pp. 135-142.
[3] M. Paul, S. M. E. Haque, and S. Chakraborty, "Human detection in surveillance videos and its applications - a review," EURASIP Journal on Advances in Signal Processing, vol. 2013, no. 1, p. 176, 2013/11/22 2013.
[4] Y. Tian, R. S. Feris, H. Liu, A. Hampapur, and M. Sun, "Robust Detection of Abandoned and Removed Objects in Complex Surveillance Videos," IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 41, no. 5, pp. 565-576, 2011.
[5] X. Huang and J. Hinze, "Analysis of Construction Worker Fall Accidents," Journal of Construction Engineering and Management-asce - J CONSTR ENG MANAGE-ASCE, vol. 129, 06/01 2003.
[6] X. Huang and J. Hinze, "Analysis of Construction Worker Fall Accidents," Journal of Construction Engineering and Management, vol. 129, no. 3, pp. 262-271, 2003/06/01 2003.
[7] S. Schneider and P. Susi, "Ergonomics and Construction: A Review of Potential Hazards in New Construction," American Industrial Hygiene Association Journal, vol. 55, no. 7, pp. 635-649, 1994/07/01 1994.
[8] A. Claudi, F. D. Benedetto, G. Dolcini, L. Palazzo, and A. F. Dragoni, "MARVIN: Mobile Autonomous Robot for Video Surveillance Networks," in 2012 Sixth UKSim/AMSS European Symposium on Computer Modeling and Simulation, 2012, pp. 21-26.
[9] K. S. Lee, M. Ovinis, T. Nagarajan, R. Seulin, and O. Morel, "Autonomous patrol and surveillance system using unmanned aerial vehicles," in 2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC), 2015, pp. 1291-1297.
[10] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), 2005, vol. 1, pp. 886-893 vol. 1.
[11] P. Dollar, C. Wojek, B. Schiele, and P. Perona, "Pedestrian detection: A benchmark," in 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 304-311.
[12] R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation," in 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014, pp. 580-587.
[13] R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Region-Based Convolutional Networks for Accurate Object Detection and Segmentation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 1, pp. 142-158, 2016.
[14] R. Girshick, "Fast R-CNN," in 2015 IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1440-1448.
[15] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137-1149, 2017.
[16] N. Chengcheng, Z. Huajun, S. Yan, and T. Jinhui, "Inception Single Shot MultiBox Detector for object detection," in 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 2017, pp. 549-554.
[17] T. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, "Focal Loss for Dense Object Detection," in 2017 IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2999-3007.
[18] J. Redmon and A. Farhadi, "YOLOv3: An Incremental Improvement," arXiv e-prints, p. arXiv:1804.02767, 2018.
[19] G. Grisetti, C. Stachniss, and W. Burgard, "Improved Techniques for Grid Mapping With Rao-Blackwellized Particle Filters," IEEE Transactions on Robotics, vol. 23, no. 1, pp. 34-46, 2007.
[20] S. Kohlbrecher, O. v. Stryk, J. Meyer, and U. Klingauf, "A flexible and scalable SLAM system with full 3D motion estimation," in 2011 IEEE International Symposium on Safety, Security, and Rescue Robotics, 2011, pp. 155-160.
[21] W. Hess, D. Kohler, H. Rapp, and D. Andor, "Real-time loop closure in 2D LIDAR SLAM," in 2016 IEEE International Conference on Robotics and Automation (ICRA), 2016, pp. 1271-1278.
[22] K. Krinkin, A. Filatov, A. y. Filatov, A. Huletski, and D. Kartashov, "Evaluation of Modern Laser Based Indoor SLAM Algorithms," in 2018 22nd Conference of Open Innovations Association (FRUCT), 2018, pp. 101-106.
[23] K. Konolige, G. Grisetti, R. Kümmerle, W. Burgard, B. Limketkai, and R. Vincent, "Efficient Sparse Pose Adjustment for 2D mapping," in 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010, pp. 22-29.
[24] J. M. Santos, D. Portugal, and R. P. Rocha, "An evaluation of 2D SLAM techniques available in Robot Operating System," in 2013 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2013, pp. 1-6.
[25] S. Thrun, D. Fox, W. Burgard, and F. Dellaert, "Robust Monte Carlo localization for mobile robots," Artificial Intelligence, vol. 128, no. 1, pp. 99-141, 2001/05/01/ 2001.
[26] C. Baby, F. Khan, and S. J.N, Home automation using IoT and a chatbot using natural language processing. 2017, pp. 1-6.
[27] N. J. Uke and P. R. Futane, "Efficient method for detecting and tracking moving objects in video," in 2016 IEEE International Conference on Advances in Electronics, Communication and Computer Technology (ICAECCT), 2016, pp. 343-348.
[28] R. Yagfarov, M. Ivanou, and I. Afanasyev, "Map Comparison of Lidar-based 2D SLAM Algorithms Using Precise Ground Truth," in 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV), 2018, pp. 1979-1983.
[29] P. François, C. Francis, and S. Roland, A Review of Point Cloud Registration Algorithms for Mobile Robotics (A Review of Point Cloud Registration Algorithms for Mobile Robotics). now, 2015, p. 1.
[30] K. E. A. v. d. Sande, J. R. R. Uijlings, T. Gevers, and A. W. M. Smeulders, "Segmentation as selective search for object recognition," in 2011 International Conference on Computer Vision, 2011, pp. 1879-1886.
[31] J. Redmon and A. Farhadi, "YOLO9000: Better, Faster, Stronger," in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 6517-6525.