Jianping Zeng

Robust consensus for linear multi-agent systems with mixed uncertainties

Abstract This paper studies the robust consensus problem for a class of linear multi-agent systems (MASs) with polytopic uncertainties and external disturbances. The uncertainties appear in both the dynamics of each agent and the topology of the overall network. Based on Lyapunov stability theory, the robust consensus problem is converted to the robust H ∞ stabilization of the error system. With the aid of the homogeneous parameter-dependent (HPD) Lyapunov functions and sum of squares (SOS) technique, the robust consensus problem is further formulated as an SOS programming. Sufficient conditions to achieve robust consensus with disturbance rejection capability are presented in terms of SOS constraints. As an extension, the robust consensus problem for discrete-time MASs is studied in a similar way. Numerical examples are provided to illustrate the effectiveness of the approach.

Robust Control Synthesis of Polynomial Nonlinear Systems Using Sum of Squares Technique

Abstract In this paper, sum of squares (SOS) technique is used to analyze the robust state feedback synthesis problem for a class of uncertain affine nonlinear systems with polynomial vector fields. Sufficient conditions are given to obtain the solutions to the above control problem either without or with guaranteed cost or H∞performance objectives. Moreover, such solvable conditions can be formulated as SOS programming problems in terms of state dependent linear matrix inequalities (LMIs) which can be dealt with by the SOS technique directly. Besides, an idea is provided to describe the inverse of polynomial or even rational matrices by introducing some extra polynomials. A numerical example is presented to illustrate the effectiveness of the approach.

Design of intelligent elevator remote monitoring system based on ethernet

In order to improve the real-time, reliability and commonality of elevator remote monitoring, this paper adopts LPC2294 to construct embedded hardware platform for elevator remote monitoring system. On the basis of the platform, rtl8019 device driver is developed. In addition, TCP/IP protocol, socket and SQLite3 database technology are applied to construct C/S model, which is able to make seamless connection and communication with back-end processor. Thus the functions of real-time remote monitoring, fault alarm, parameter setting and simulation for elevator are realized. This not only greatly improves elevators security and reliability, but also decreases elevator human resource costs as well as fault maintenance costs.

Multivariable decoupling control of boiler steam temperature system based on diagonal dominance

A design method for a constant pre-compensator in frequency domain of boiler steam temperature multivariable control system based on diagonal dominance and the method of linear matrix inequalities(LMI) is presented. Compared with previous methods, the algorithm presented de-emphasizes the importance of choosing a suitable frequency in designing process. This method is aimed at reducing the effects of interactions in multivariable control system. The results show that this approach can effectively reduce interactions and the order of the controller.

Robust attitude control of flexible spacecraft with quality characteristic parameter uncertainty

In this paper, the robust attitude controller designed for flexible spacecraft with quality characteristic parameter uncertainty is studied. The problem about satellite attitude control is transformed into a standard H∞ control problem through approximate linearization method, finally the dynamic output feedback controller can be designed by using variable substitution method, which can transform nonlinear matrix inequalities into linear ones that are easy to be solved by LMI toolbox. The simulation results show that the controller has strong robustness and can adapt to biggish quality characteristic parameter uncertainty and overcome both internal and external disturbances well. This paper has some referential value in achieving very accurate and stable satellite attitude control.

Uncertainty modeling for flight control system with wide operation range

A method of uncertainty modeling is proposed for the flight control system with wide operation range. The model built according to this method contains two parts: one is a polynomial matrix obtained via least square fitting, and the other is the fitting error. The polynomial matrix is related to the parameters and reflects the time-varying characteristic. The fitting error is considered as norm-bounded uncertainties and reflects the uncertain and nonlinear characteristics of the system. And as an example, a model is built for a certain flight control system.

Stability of sampled-data linear active disturbance rejection control: A classical control approach

Stability of continuous-time and sampled-data LADRC is investigated with classical control approach. For the second-order uncertain LTI plant, the LADRC is described as an interconnection structure. Then sufficient conditions to guarantee continuous-time and sampled-data LADRC stabilizing the plant are proved respectively, while in proving the main theoretical tool is Nyquist diagram. With these results, a continuous-time controller designed with bandwidth tuning approach can be used as sampled-data controller after Euler approximation with a small sampling period.

Sampled-data feedback control of Han canonical form

The problem of how to design sampled-data controller for Han canonical form is investigated. With the stability results of approximated discrete-time model, sufficient conditions are proved to guarantee a controller semi-globally exponentially stabilizes the Han form while the controller can be designed by directly canceling the total disturbance and stabilizing a cascade-integrator. Simulations show that the sampled-data controller may have comparable performance with the controller designed with backstepping while the design procedure is much easier.

An interactive simulation platform design for analyzing ship steer system

This paper describes an Interactive Platform of Ship Steer System Design based on Matlab. The platform is consist of five function modules: i) modeling module, ii) config module, iii) controller design module, iv) graphic display module, v) data management module. All of the five modules are connected by a main unit which is the center of star network of the platform. It is useful for decreasing the coupling of five sub-modules and exchanging data with each other, and also makes the platform more extensible. The interactive platform is showed to be an efficient tool for enhancing the design process of ship steer system. Finally, a typical example is given to illustrate the utility of the platform.

Robust stablization for a class of large envelope flight control systems

This paper deals with robust control problems for large envelope flight control systems. Based on data fitting, a class of uncertainty model is improved to describe the flight control systems with wide operation range. The uncertainty model can reflect both the time-varying characteristic and the fitting error. Robust stability criterion and stabilizers design method via state feedback are proposed for flight control systems with the class of uncertainty. Moreover, a convex optimization method is provided to estimate the parameter perturbation range. Finally, a simulation example is given to illustrate that the method is feasible.