Browsing by Author "Hwang, MC"

Now showing 1 - 8 of 8

A finite difference method with adaptive time mesh for hyperbolic traffic flow
(2002)
Cho, HJ
;
Hwang, MC
;
Jou, TJ
;
運輸與物流管理系註：原交通所+運管所
;
Department of Transportation and Logistics Management
This paper presents a finite difference method with variable time mesh for the hyperbolic traffic flow. The width of each time step is determined by the ratio of uniform space mesh size and maximal characteristic velocity. The proposed adaptive time mesh scheme Is compatible with most of explicit finite difference methods. Numerical examples with different initial and boundary conditions of the Lighthill-Whitham-Richards (LWR) model and Payne-Whitham model are provided to contrast the effects of adaptive time mesh on the Lax-Friedrichs scheme. Simulation results are generally satisfied. The number of time steps of adaptive-Lax method is much less than that of Lax method with no significant difference in solutions of LWR model. Convergence of solution is readily claimed by the Courant-Friedrichs-Lewy (CFL) condition.
A stimulus-response model of day-to-day network dynamics
(2005-03-01)
Cho, HJ
;
Hwang, MC
;
運輸與物流管理系註：原交通所+運管所
;
運輸研究中心
;
Department of Transportation and Logistics Management
A general structure of stimulus-response formula is presented to specify the interacted network dynamics under the assumption of a daily learning and adaptive travel behavior. By taking the time derivative of system variable as a response term, the evolution is formulated as a dynamic system. Issues of existence, uniqueness, and stability for the proposed differential equations are briefly discussed. Approximation of a time-varying route-choice model is derived from the addressed path-flow dynamics. Threshold effects on path-flow dynamics are encapsulated into the proposed general structure by incorporating a discontinuous stimulus term. Then, the quasi user equilibrium is achieved when all users feel indifferent between the experienced and predicted travel time provided by intelligent transportation systems, i.e., the whole system dynamics stay within a bounded range. The derived quasi user equilibrium is reduced to Wardrop's user equilibrium as the threshold effects of path-flow dynamics vanish.
A stimulus-response model of day-to-day network dynamics
(2003)
Cho, HJ
;
Hwang, MC
;
運輸與物流管理系註：原交通所+運管所
;
Department of Transportation and Logistics Management
A general structure of stimulus-response formula is presented to specify the interacted network dynamics in a day-to-day time scale under the assumption of a learning and adaptive behavioral process. By taking the time derivative of system variable as a reaction in stimulus-response equation the evolution of whole system is formulated as a dynamical system. Issue of stability for the proposed differential equations are briefly discussed. Approximation of time dependent route-choice model is derived from the addressed path flow dynamics, The threshold effect on path flow dynamics is developed in term of defining a discontinuous stimulus. Quasi user equilibrium researched to be a steady state when all users feel indifferent between experienced travel time and predicted travel time provided by Advanced Traveler Information Systems. The proposed quasi user equilibrium is reduced to Wardrop's user equilibrium if threshold effect is vanishing.
Day-to-day vehicular flow dynamics in intelligent transportation network
(2005-02-01)
Cho, HJ
;
Hwang, MC
;
運輸與物流管理系註：原交通所+運管所
;
Department of Transportation and Logistics Management
In this paper, a flow evolution model is developed by using the dynamical system approach for a vehicular network equipped with predicted travel information. The concerned system variables, path flow, and predicted minimal travel time of an origin-destination (OD) pair, are measured on the peak-hour-average base for each day. The time change rates of these two variables are formulated as a system of ordinary differential equations under the assumption of daily learning and adaptive processes for commuters. By incorporating the total perceived travel time loss (or saving) into the proposed models, time change rates of path flows are generated with a flow-related manner to prevent path flow dynamics from being insensible to traffic congestion which had been formulated in the existing studies. Heterogeneous models with various user adjusting sensitivities and predicted travel information are also presented. Equilibrium solutions of the proposed network dynamics satisfy the Wardrop user equilibria and are proved to be asymptotically stable by using the stability theorem of Lyapunov. The issue of existence and uniqueness of solutions is proved both on the lemma of Lipschitz condition and the fundamental theorem of ordinary differential equations. In addition, some simple examples are demonstrated to show the asymptotic behaviors of the proposed models numerically. (c) 2005 Elsevier Ltd. All rights reserved.
Estimation of dynamic origin-destination by Gaussian state space model with unknown transition matrix
(2003)
Jou, YJ
;
Hwang, MC
;
Wang, YH
;
Chang, CH
;
統計學研究所
Dynamic origin-destination (O-D) pattern representing time-dependent trip demands from one place (origin) to another (destination) is one of the most essential input data for most traffic operational analyses. Historical studies assumed that the transition matrix is known or at least approximately known, which is unrealistic for a real world network. And due to the fact that the number of trips to a specific destination, y, is easy to obtain and the O-D variable, x (path flow based in this research), is not directly observable, a Gaussian state space model is formulated to describe the relationships of x and y, observation equations, and the dynamics of x, state equations with unknown transition matrix. Under the assumption of Gaussian noise terms in state space model, the distribution of random transition matrix F is derived. A solution algorithm combining Gibbs sampler and Kalman filter to tackle the problem of simultaneous estimation of F and x, based on the latest available information is proposed. Real O-D data from Taipei Rapid Transit is used to verify the presented model and solution method. Preliminary results are generally satisfactory, showing that also in the unknown transition matrix case, significant estimates are achieved.
Quasi user equilibrium and its stability
(2004)
Cho, HJ
;
Hwang, MC
;
Chiu, YT
;
科技管理研究所
;
Institute of Management of Technology
In the historical studies of dynamic traffic assignment (or the so-called dynamic route choice modeling), user equilibrium is a fundamental paradigm to guide the final state of flow diversion for a transportation network. However, the user behavior of path selection had been observed and indicated that there exists an adaptive and learning process to adjust the daily travel decision as travel information is provided by Intelligent Transportation Systems in particular. In this paper, a general structure of stimulus-response mechanism to formulate daily path adjustment is reviewed first. Quasi user equilibrium is then introduced in the same modeling framework but replacing a threshold-effect response term. A proof of the stability of quasi user equilibrium in the sense of Lyapunov is presented. Finally, a simple network is illustrated to show the verification of the proposed approach numerically.
Robust attitude control of spacecraft using sliding mode control and productive networks
(1997-05-01)
Chiou, JC
;
Hwang, MC
;
Wu, SD
;
Yang, JY
;
交大名義發表
A new robust attitude control design of spacecraft is proposed by combining sliding mode control (SMC) and productive networks (PN). Essentially, the sliding mode control uses discontinuous control action to drive state trajectories toward a specific hyperplane in the state space, and to maintain the state trajectories sliding on the specific hyperplane. This principle provides a guideline to design a robust controller. Productive networks, which are a special type of artificial neural network, are then used to implement reaching and sliding conditions, and tackle the drawbacks of SMC such as chattering and high control gains. Attractive features of the proposed method include a systematic procedure of controller design, a reduction in chattering, robustness against model uncertainties and external disturbances. An inverted pendulum and a spacecraft attitude control problem are given to deomonstrate the effectiveness of the proposed method.
Robust attitude control of spacecraft using sliding mode control and productive networks
(1997-05-01)
Chiou, JC
;
Hwang, MC
;
Wu, SD
;
Yang, JY
;
電控工程研究所
A new robust attitude control design of spacecraft is proposed by combining sliding mode control (SMC) and productive networks (PN). Essentially, the sliding mode control uses discontinuous control action to drive state trajectories toward a specific hyperplane in the state space, and to maintain the state trajectories sliding on the specific hyperplane. This principle provides a guideline to design a robust controller. Productive networks, which are a special type of artificial neural network, are then used to implement reaching and sliding conditions, and tackle the drawbacks of SMC such as chattering and high control gains. Attractive features of the proposed method include a systematic procedure of controller design, a reduction in chattering, robustness against model uncertainties and external disturbances. An inverted pendulum and a spacecraft attitude control problem are given to deomonstrate the effectiveness of the proposed method.