Browsing by Author "Tu, KY"

Now showing 1 - 3 of 3

Design of a multi-layer fuzzy logic controller using pole assignment for bipedal walking at varying speeds
(2004-01-01)
Tu, KY
;
Lee, TT
;
電控工程研究所
;
Institute of Electrical and Control Engineering
When a biped robot detects obstacles along its path, it will reduce its walking speed to change direction for obstacle avoidance. It is essential for a biped robot to be able to vary its speed during walking. To control biped robots with variable walking speed capability, a Multi-Layer Fuzzy Logic Controller (MLFLC) is developed using the pole assignment technique in this paper. This method extends the previous result, that a switching surface, whose parameters are scaling factors, exists in a Multi-Layer Fuzzy Logic Controller (Tu et al., 2000). In this study, the poles of a closed-loop control system are derived relative to the switching surface parameters, i.e. the normalized scaling factors. Therefore, after choosing the appropriate poles, scaling factors are determined and then a Multi-Layer Fuzzy Logic Controller is designed. This design has two important features: (1) simplifying Fuzzy Logic Controller (FLC) design to select only the scaling factors; (2) pursuing the system performance requirement, based on the pole assignment. A three-link biped robot is used as the illustrative example. Simulation results are included.
Design of a new fuzzy suction controller using fuzzy modeling for nonlinear boundary layer
(2005-10-01)
Tu, KY
;
Lee, TT
;
Wang, CH
;
電控工程研究所
;
Institute of Electrical and Control Engineering
There are two types of fuzzy modeling: 1) imitating an expert experiment or fulfilling an engineering knowledge, and 2) modeling a complex or unknown system. In this paper, based on the first type of fuzzy modeling, a new fuzzy suction controller (NFSC) is proposed using its linguistic rules to design nonlinear boundary layer. Two kinds of nonlinear boundary layers are discussed. The first kind is designed by three rules derived according to a new interpretation of the switching conditions for a suction controller such that the new controller reduces chattering and spends less energy than a suction controller does. A design procedure summarizes the NFSC design. The second kind of nonlinear boundary layer is the linguistic rules designed to have sliding sectors to control a mobile robot for trajectory tracking. The discussion emphasizes the advantage of nonlinear boundary layers, compared with traditional suction controllers usually using linear boundary. In addition, the proposed NFSC provides a flexible way to adjust the controller functions using linguistic rules based on the first type of fuzzy modeling.
Random-coincidence corrections using iterative reconstruction for PET images
(2000)
Chen, JC
;
Liu, RS
;
Tu, KY
;
Lu, HHS
;
Chen, TB
Iterative reconstruction (IR) algorithms can reduce artifacts caused by filtered backprojection (FBP) or convolution backprojection (CBP). Recently, the computational effects required for IR of positron emission tomography (PET) studies have been reduced to make it practically appealing. We have made an implementation of the improved Maximum Likelihood-Expectation Maximization (ML-EM) algorithm. The transition matrix was generated based on the geometry of the instrument. Phantoms of 6 line sources and 19 line sources were used to test our accelerated ML-EM algorithms against the FBP method. The singles were used to calculate the random coincidence rates by a well known formula and were compared to the randoms obtained by another geometric method. We also designed a new model using two line sources to determine the ratio of random events to true events. The artifacts near those line sources were eliminated with the ML-EM method. With decay correction, the RC events were uniformly distributed in whole field after 10 iterations. The ML-EM reconstructed images are superior to those obtained with FBP. The patterns of randoms provide insightful information for random correction, which the hardware correction by the delay window can not provide. This information is particular valuable when the delay window correction is not available in the old fashion PET scanner.