This study addresses the collaborative tracking problem for a set of linear systems with noisy measurements using decentralized iterative learning control. The collaborative tracking problem indicates that the desired...
详细信息
This study addresses the collaborative tracking problem for a set of linear systems with noisy measurements using decentralized iterative learning control. The collaborative tracking problem indicates that the desired reference trajectory over a finite time interval is required to be precisely tracked by a combination of the outputs of several independent subsystems. A description of the achievable tracking reference and its corresponding input is elaborated. Two decentralized learning schemes are then provided to solve the collaborative tracking problem iteratively. Numerical simulations are given to verify the derived results.
In order to solve the problems of slow tracking speed and low accuracy in vehicle mobile satellite communication system, a new step tracking algorithm based on the steepest descent method was proposed. This algorithm ...
详细信息
In order to solve the problems of slow tracking speed and low accuracy in vehicle mobile satellite communication system, a new step tracking algorithm based on the steepest descent method was proposed. This algorithm adjusted the tracking direction and step size by the steepest descent method, and set the proper threshold to stop tracking. The simulation results showed that the numbers of tracking were reduced to17 steps and tracking accuracy is improved.
In this paper, numerical aspects of a sensitivity control for the semi-active suspension system with a magneto-rheological (MR) damper are investigated. A 2-dof quarter-car model together with a 6th order polynomial m...
详细信息
In this paper, numerical aspects of a sensitivity control for the semi-active suspension system with a magneto-rheological (MR) damper are investigated. A 2-dof quarter-car model together with a 6th order polynomial model for the MR damper are considered. For the purpose of suppressing the vertical acceleration of the sprung mass, the square of the vertical acceleration is defined as a cost function and the current input to the MR damper is adjusted in the fashion that the current is updated in the negative gradient of the cost function. Also, for improving the handling performance, a weighted absolute velocity of the sprung mass is added to the control law. The implementation of the proposed algorithm requires only the measurement of the relative displacement of the suspension deflection. The local stability of the equilibrium point of the closed loop nonlinear system is proved by investigating the eigenvalues of the linearized one. Through simulations, the passive suspension, the skyhook control, and the proposed sensitivity control are compared.
暂无评论