|
Model-inverse Time Delay Control (MiTDC) for Force Control |
|
Suin Kim, Joonbum Bae(UNIST, Korea) |
| Even though it has been an important issue to control the output force of actuators accurately;especially physical human-robot interaction (pHRI) fields, it is very difficult due to interaction with the complex and time-varying system dynamics and surrounding conditions. In this study, a robust control strategy named model-inverse time delay control (MiTDC) was proposed to realize an ideal force-mode control of the series elastic actuators (SEA) without an accurate model of the control systems and its surroundings. |
|
|
Arbitrarily large gain/phase margin can be achieved by DOB-based controller design |
|
Hyuntae Kim, Gyunghoon Park, Hyungbo Shim(Seoul National University, Korea), Nam Hoon Jo(Soongsil University, Korea) |
| In this paper, we prove that the DOB-based feedback controller can guarantee arbitrarily large but bounded
gain margin and phase margin. This fact reminds that the DOB-based controller plays the role of robust controller even
in the classical sense. We present a design procedure that ensures desired gain/phase margin even for the plants that have
parametric uncertainty. |
|
|
Hâ‚‚ Control for Discrete-time Markovian Jump Fuzzy Systems with Partly Known Transition Probabilities |
|
In Seok Park, Nam Kyu Kwon, PooGyeon Park(POSTECH, Korea) |
| In this paper, the H2 control for discrete-time Markovian jump fuzzy system (MJFS) with partly known transition probabilities is considered. To deal with unknown transition probabilities, convex property of the normalized unknown transition probabilities and lower bound lemma for the inversion of the matrix summation are used. Then, using the properties of fuzzy weights, quadratic condition of stochastic stability and guaranteed cost is obtained. Finally, linear theorem, the practical problem is given as a numerical example. |
|
|
Novel Distributed Hybrid Nonlinear Filtering Method for Robust Localization UsingWireless Sensor Networks |
|
Jung Min Pak, Choon Ki Ahn, Myo Taeg Lim(Korea University, Korea) |
| This paper propose a novel distributed hybrid nonlinear filtering method that is robust against non-line-of-sight (NLOS) situations in indoor localization using wireless sensor networks (WSN). Instead of processing all of the measurements via a single filter, the proposed method distributes the measurements to several local filters. By using distributed filtering and a data association (DA) technique, abnormal measurements due to the NLOS are identified. |
|
|
A Robust Tracking Controller for Robot Manipulators using Disturbance Observer |
|
Wonseok Ha, Juhoon Back(Kwangwoon University, Korea) |
| This paper presents a robust controller for robot manipulators. The controller is based on the disturbance observer which can efficiently compensate the external disturbances and plant uncertainties. Under the assumption that angular displacements as well as angular velocities are available for feedback, we propose a disturbance observer-based robust controller which ensures robust tracking of desired trajectory with prescribed error bound. The proposed controller enjoys a simple structure and can be implemented with n integrators when the manipulator has n joints. |
|
|
Dynamic Analysis of Series Elastic Actuator in Terms of Controller Design |
|
Sehoon Oh(DGIST, Korea), Kyoungchul Kong(Sogang University, Korea) |
| Series elastic actuator (SEA) has been developed and are now widely utilized. In spite of the characteristics as an ideal force source for safe human-robot interactions, SEA is considered difficult to design and control as its mechanical structure is complicated, namely it consists of a motor, a reduction gear.
It is required to investigate how the dynamic characteristics are influenced by these issues, and this paper tackles this problem by deriving and analyzing the dynamics of SEA. The dynamics of SEA with parameter variations is evaluated using the criteria utilized for controller design |
|