Entity

Time filter

Source Type


Chan T.C.,National Tsing Hua University | Chan T.C.,Precision Machinery Research and Development Center | Sung C.K.,National Tsing Hua University | Chao P.C.P.,National Chiao Tung University
Microsystem Technologies | Year: 2012

This study was aimed at evaluating the perfect balancing position of an automatic ball balancer installed in optical disk drives taking into consideration the effects of the rolling friction, speed ratio, and scaling parameter on ball positioning. A mathematical model that is employed to derive the dynamic equations of the ABB system was constructed. Stability of the steady-state solutions was then analyzed. A numerical simulation and an experimental study were conducted to verify the mathematical model. The simulation and experimental results were in good agreement. © Springer-Verlag 2012.


Hwang K.-S.,National Sun Yat - sen University | Jiang W.-C.,National Chung Cheng University | Chen Y.-J.,National Chung Cheng University | Wang W.-H.,Precision Machinery Research and Development Center
Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013 | Year: 2013

In this paper, a model learning method based on tree structures is present to achieve the sample efficiency in stochastic environment. The proposed method is composed of Q-Learning algorithm to form a Dyna agent that can used to speed up learning. The Q-Learning is used to learn the policy, and the proposed method is for model learning. The model builds the environment model and simulates the virtual experience. The virtual experience can decrease the interaction between the agent and the environment and make the agent perform value iterations quickly. Thus, the proposed agent has additional experience for updating the policy. The simulation task, a mobile robot in a maze, is introduced to compare the methods, QLearning, Dyna-Q and the proposed method. The result of simulation confirms the proposed method that can achieve the goal of sample efficiency. © 2013 IEEE.


Chou H.-C.,National Taiwan University of Science and Technology | Chung J.-C.,National Taiwan University of Science and Technology | Kuo C.-H.,National Taiwan University of Science and Technology | Chou B.-Y.,Precision Machinery Research and Development Center
Proceedings of the World Congress on Intelligent Control and Automation (WCICA) | Year: 2011

In this paper, a kid size humanoid robot, HuroEvolution, is presented. The HuroEvolution is developed for two autonomous robot competitions of RoboCup 2010 and FIRA 2010. In order to perform various competition functions, the robot structure, sensor modules, control systems and software programs are designed to meet the competition requirements of participated games. In addition to the brief introductions of the mechanical structure, sensor modules and control system, this paper focuses on the illustrations of autonomy for various skill-oriented functions. Our solutions are developed based on autonomous sensing and decision approaches. The autonomous sensing approach is realized in terms of typical sensor fusion techniques, which combine the information collected from a web camera and a 3-axis magneto-resistive sensor. The fused sensor information is further used for skill-oriented autonomous decision systems which are responsible of navigations and task executions. Currently, the HuroEvolution is capable of playing the games of soccer, weightlifting, obstacle run, sprint, marathon and basketball. Finally, the proposed skill-oriented functions were verified in the RoboCup 2010 and FIRA 2010 competitions. © 2011 IEEE.


Sung C.K.,National Tsing Hua University | Chan T.C.,National Tsing Hua University | Chan T.C.,Precision Machinery Research and Development Center | Chao C.P.,National Chiao Tung University | Lu C.H.,National Tsing Hua University
Mechanism and Machine Theory | Year: 2013

This study examined the influence of external excitations on ball positioning in an automatic ball balancer (ABB) installed in a rotor system. The authors' previous studies adopted a model that considered the ABB as an autonomous system by neglecting external excitations. We examined how the magnitude, the frequency and even the phase of an external excitation affected ball positioning. Simulations were performed to predict the ball positions under various external forces. Then, we constructed an experimental rig by employing a shaker to apply excitations to the rotor system and the associated ABB to verify the theoretical development. Simulation results indicated that the balancing balls of the ABB could counterbalance the external force by the change of the ball positions. However, it was observed from the experiment that the ball would not be displaced if the external force was applied after the ball had been positioned because the excessive rolling resistance between the ball and the runway prevented the ball from moving to desired positions. © 2013 Elsevier Ltd.


Hwang K.S.,National Sun Yat - sen University | Ling J.L.,Shih Hsin University | Chen Y.-Y.,National Chung Cheng University | Wang W.-H.,Precision Machinery Research and Development Center
Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics | Year: 2014

In this paper, a non-expert learning system was proposed to guide the robots learn their behaviors by humans' emotional expressions. The proposed system used interval fuzzy type-2 algorithm to recognize the human's facial expressions, which were captured by a web camera. Furthermore, emotion value (E-value), generated based on non-expert human's facial expressions, was applied to the reinforcement learning to train robots. Two kinds of problems were experimented. One was the human being know the exact solution to train robots and could clearly observe good or bad choice robots had been made. The other one was human being did not know the exact solution but robots could still learn from human's experience. The experiment results show that no matter the learning environment could be clearly observed by human being or not, robots could learn from human's facial expressions by the proposed learning system. © 2014 IEEE.

Discover hidden collaborations