許多場景中，人們需要處理各種重物、長物或不方便搬運的物品，在這方面使用移動式機械臂作為輔助，可以使得處理這些問題變得更加容易，尤其在只有一個人的時候。隨着機器人和機械臂的日益普及，它們正逐漸成為人們日常生活中不可或缺的一部分，為了支持人與機器人之間的互動，認知型人機互動成為一個研究領域。
本研究以協同搬運物件為主題，研究中採用了模仿學習的技術，通過人與人示範性的合作搬運，從過程中獲取力/力矩感測器的量測資料以及透過基於影像上姿態辨識和九軸感測器融合的位置、速度、加速度資料來達成順應性控制。
此外，本研究還將桌子升降、人體遠近和頭部姿態等信息納入考慮，以預測人類的意圖並提供相應的搬運輔助。通過分析這些信息，機器人可以根據人體的位置、動作和頭部的姿態變化來預測搬運的目標和需求，從而更好地協同工作。

In many scenarios,people often encounter the need to handle heavy objects, long items, or items that are inconvenient to move. In such cases, with the help of Autonomous Mobile Manipulators(AMM) as assistance can greatly simplify the task, especially when they are alone. With the increasing popularity of robots and AMM, they will gradually become an integral part of people's daily lives. In order to support interaction between humans and robots, the field of cognitive human-robot interaction has emerged.
In this thesis, we focuses on collaborative object handling. It employs a technique called imitation learning, where humans demonstrate cooperative object handling to obtain force/torque measurements from force sensors. Additionally, a fusion of position, velocity, and acceleration data from image-based pose estimation and nineaxis sensors is used for compliance control
Additionally, this study takes into consideration factors such as the edge of the table, the distance of the human body, and head pose to predict human intentions and provide corresponding assistance during object handling. By analyzing this information, the AMM can estimate the operator’s intention based on the changes in position, movements, and head posture of the human, enabling better collaboration.

[1] J. K. Salisbury, "Active Stiffness Control of a Manipulator in Cartesian Coordinates,” Proceedings of the IEEE Conference on Decision and Control, Stanford, California, pp. 95-100, December.1980
[2] N. Hogan, “Impedance Control︰An Approach to Manipulation︰ Part1- Theory, Part2-Implementation, Part3-An Approach to Manipulation,”ASMEJournal of Dynamic System, Measurement, and Control, Vol. 107, pp. 1-24,1985.
[3] W-S. Lu, and Q-H. Meng, "Impedance control with adaptation for roboticmanipulations," IEEE Transactions on Robotics and Automation, vol. 7, no.3, pp. 408–415, Jun.1991.
[4] V. Duchaine and C. Gosselin, “General model of human–robotcooperation using a novel velocity based variable impedance control,” in Proc.2nd Joint EuroHaptics Conf. Symp. Haptic Interfaces Virtual Environ.Teleoperator Syst, pp. 446–451, 2007.
[5] W. Sheng, A. Thobbi and Y. Gu, "An integrated framework for human–robot collaborative manipulation," IEEE Trans. Cybern, vol. 45, no. 10, pp.2030-2041, Oct. 2015.
[6] D. F. Paez Granados, B. A. Yamamoto, H. Kamide, J. Kinugawa and K. Kosuge, "Dance teaching by a robot: Combining cognitive and physical human–robot interaction for supporting the skill learning process," IEEE Robot. Autom. Lett., vol. 2, no. 3, pp. 1452-1459, Jul.2017.
[7] A. Al-Yacoub, Y. C. Zhao, W. Eaton, Y. M. Goh and N. Lohse,"Improving human robot collaboration through force/torque based learning forobject manipulation," Robot. Comput.-Integr. Manuf., vol. 69, Jun. 2021.

[8] Valentin Bazarevsky, Ivan Grishchenko, Karthik Raveendran, Tyler Zhu,Fan Zhang, Matthias Grundmann. "Blazepose: On-device real-time body pose tracking." arXiv preprint arXiv:2006.10204, 2020.
[9] Terven Juan, Cordova-Esparza Diana, "A Comprehensive Review of YOLO:From YOLOv1 to YOLOv8 and Beyond." arXiv preprint arXiv:2304.00501,2023.

[10] Wang, C. Y., Bochkovskiy, A., & Liao, H. Y. M. "YOLOv7:Trainable bag-of-freebies sets new state-of-the-art for real-time objectdetectors." In Proceedings of the IEEE/CVF Conference on Computer Visionand Pattern Recognition , pp. 7464-7475, 2023.

[11] Maji Debapriya,Nagori Soyeb,Mathew Manu,Poddar Deepak, "Yolo-pose:Enhancing yolo for multi person pose estimation using object keypointsimilarity loss." In: Proceedings of the IEEE/CVF Conference on ComputerVision and Pattern Recognition, pp. 2637-2646, 2022.
[12] Welch, Greg, and Gary Bishop, "An introduction to the Kalmanfilter." 1995.

[13] Roetenberg, Daniel, Per J. Slycke, and Peter H. Veltink. "Ambulatoryposition and orientation tracking fusing magnetic and inertial sensing." IEEETransactions on Biomedical Engineering 54.5, pp.883-890, 2007

[14] Mutlu, Bilge, Nicholas Roy, and Selma Šabanović. "Cognitivehuman–robot interaction." Springer handbook of robotics, pp.1907-1934,2016.

[15] V. Bazarevsky, Y. Kartynnik, A. Vakunov, K. Raveendran, M.Grundmann, "Blazeface: Sub-millisecond neural face detection on mobilegpus." arXiv preprint arXiv:1907.05047, 2019.