Yunong Zhang

Bio: Yunong Zhang is an academic researcher from The Chinese University of Hong Kong. The author has contributed to research in topics: Recurrent neural network & Artificial neural network. The author has an hindex of 12, co-authored 14 publications receiving 1357 citations. Previous affiliations of Yunong Zhang include Sun Yat-sen University.

A recurrent neural network for solving Sylvester equation with time-varying coefficients

Abstract: Presents a recurrent neural network for solving the Sylvester equation with time-varying coefficient matrices. The recurrent neural network with implicit dynamics is deliberately developed in the way that its trajectory is guaranteed to converge exponentially to the time-varying solution of a given Sylvester equation. Theoretical results of convergence and sensitivity analysis are presented to show the desirable properties of the recurrent neural network. Simulation results of time-varying matrix inversion and online nonlinear output regulation via pole assignment for the ball and beam system and the inverted pendulum on a cart system are also included to demonstrate the effectiveness and performance of the proposed neural network.

A dual neural network for redundancy resolution of kinematically redundant manipulators subject to joint limits and joint velocity limits

Abstract: In this paper, a recurrent neural network called the dual neural network is proposed for online redundancy resolution of kinematically redundant manipulators. Physical constraints such as joint limits and joint velocity limits, together with the drift-free criterion as a secondary task, are incorporated into the problem formulation of redundancy resolution. Compared to other recurrent neural networks, the dual neural network is piecewise linear and has much simpler architecture with only one layer of neurons. The dual neural network is shown to be globally (exponentially) convergent to optimal solutions. The dual neural network is simulated to control the PA10 robot manipulator with effectiveness demonstrated.

Obstacle avoidance for kinematically redundant manipulators using a dual neural network

Abstract: One important issue in the motion planning and control of kinematically redundant manipulators is the obstacle avoidance. In this paper, a recurrent neural network is developed and applied for kinematic control of redundant manipulators with obstacle avoidance capability. An improved problem formulation is proposed in the sense that the collision-avoidance requirement is represented by dynamically-updated inequality constraints. In addition, physical constraints such as joint physical limits are also incorporated directly into the formulation. Based on the improved problem formulation, a dual neural network is developed for the online solution to collision-free inverse kinematics problem. The neural network is simulated for motion control of the PA10 robot arm in the presence of point and window-shaped obstacle.

Planar inverted F antenna loaded with high permittivity material

Abstract: Enhanced gain of an inverted F antenna loaded with a very high permittivity substrate has been demonstrated. Measurements show that the size of the antenna can be reduced without sacrificing the antenna radiation efficiency. The gain-enhancement mechanism, bandwidth, and antenna radiation gain and patterns are presented and discussed.

Zhang Neural Network Versus Gradient Neural Network for Solving Time-Varying Linear Inequalities

Abstract: By following Zhang design method, a new type of recurrent neural network [i.e., Zhang neural network (ZNN)] is presented, investigated, and analyzed for online solution of time-varying linear inequalities. Theoretical analysis is given on convergence properties of the proposed ZNN model. For comparative purposes, the conventional gradient neural network is developed and exploited for solving online time-varying linear inequalities as well. Computer simulation results further verify and demonstrate the efficacy, novelty, and superiority of such a ZNN model and its method for solving time-varying linear inequalities.

Compact and Broadband Microstrip Antennas

Abstract: Preface. 1. Introduction and Overview. 1.1 Introduction. 1.2 Compact Microstrip Antennas. 1.3 Compact Broadband Microstrip Antennas. 1.4 Compact Dual-Frequency Microstrip Antennas. 1.5 Compact Dual-Polarized Microstrip Antennas. 1.6 Compact Circularly Polarized Microstrip Antennas. 1.7 Compact Microstrip Antennas with Enhanced Gain. 1.8 Broadband Microstrip Antennas. 1.9 Broadband Dual-Frequency and Dual-Polarized Microstrip Antennas. 1.10 Broadband and Dual-Band Circularly Polarized Microstrip Antennas. 2. Compact Microstrip Antennas. 2.1 Introduction. 2.2 Use of a Shorted Patch with a Thin Dielectric Substrate. 2.3 Use of a Meandered Patch. 2.4 Use of a Meandered Ground Plane. 2.5 Use of a Planar Inverted-L Patch. 2.6 Use of an Inverted U-Shaped or Folded Patch. 3. Compact Broadband Microstrip Antennas. 3.1 Introduction. 3.2 Use of a Shorted Patch with a Thick Air Substrate. 3.3 Use of Stacked Shorted Patches. 3.4 Use of Chip-Resistor and Chip-Capacitor Loading Technique. 3.5 Use of a Slot-Loading Technique. 3.6 Use of a Slotted Ground Plane. 4. Compact Dual-Frequency and Dual-Polarized Microstrip Antennas. 4.1 Introduction. 4.2 Some Recent Advances in Regular-Size Dual-Frequency Designs. 4.3 Compact Dual-Frequency Operation with Same Polarization Planes. 4.4 Compact Dual-Frequency Operation. 4.5 Dual-Band or Triple-Band PIFA. 4.6 Compact Dual-Polarized Designs. 5. Compact Circularly Polarized Microstrip Antennas. 5.1 Introduction. 5.2 Designs with a Cross-Slot of Unequal Arm Lengths. 5.3 Designs with a Y-Shaped Slot of Unequal Arm Lengths. 5.4 Designs with Slits. 5.5 Designs with Spur Lines. 5.6 Designs with Truncated Corners. 5.7 Designs with Peripheral Cuts. 5.8 Designs with a Tuning Stub. 5.9 Designs with a Bent Tuning Stub. 5.10 Compact CP Designs with an Inset Microstrip-Line Feed. 6. Compact Microstrip Antennas with Enhanced Gain. 6.1 Introduction. 6.2 Compact Microstrip Antennas with High-Permittivity Superstrate. 6.3 Compact Microstrip Antennas with Active Circuitry. 7. Broadband Microstrip Antennas. 7.1 Introduction. 7.2 Use of Additional Microstrip Resonators. 7.3 Microstrip Antennas with an Air Substrate. 7.4 Broadband Slot-Loaded Microstrip Antennas. 7.5 Broadband Microstrip Antennas with an Integrated Reactive Loading. 7.6 Broadband Microstrip Antennas with Reduced Cross-Polarization Radiation. 8. Broadband Dual-Frequency and Dual-Polarized Microstrip Antennas. 8.1 Introduction. 8.2 Broadband Dual-Frequency Microstrip Antennas. 8.3 Broadband Dual-Polarized Microstrip Antennas. 9. Broadband and Dual-Band Circularly Polarized Microstrip Antennas. 9.1 Introduction. 9.2 Broadband Single-Feed Circularly Polarized Microstrip Antennas. 9.3 Broadband Two-Feed Circularly Polarized Microstrip Antennas. 9.4 Broadband Four-Feed Circularly Polarized Microstrip Antennas. 9.5 Dual-Band Circularly Polarized Microstrip Antennas. Index.

Design and analysis of a general recurrent neural network model for time-varying matrix inversion

Abstract: Following the idea of using first-order time derivatives, this paper presents a general recurrent neural network (RNN) model for online inversion of time-varying matrices. Different kinds of activation functions are investigated to guarantee the global exponential convergence of the neural model to the exact inverse of a given time-varying matrix. The robustness of the proposed neural model is also studied with respect to different activation functions and various implementation errors. Simulation results, including the application to kinematic control of redundant manipulators, substantiate the theoretical analysis and demonstrate the efficacy of the neural model on time-varying matrix inversion, especially when using a power-sigmoid activation function.

A recurrent neural network for solving Sylvester equation with time-varying coefficients

Abstract: Presents a recurrent neural network for solving the Sylvester equation with time-varying coefficient matrices. The recurrent neural network with implicit dynamics is deliberately developed in the way that its trajectory is guaranteed to converge exponentially to the time-varying solution of a given Sylvester equation. Theoretical results of convergence and sensitivity analysis are presented to show the desirable properties of the recurrent neural network. Simulation results of time-varying matrix inversion and online nonlinear output regulation via pole assignment for the ball and beam system and the inverted pendulum on a cart system are also included to demonstrate the effectiveness and performance of the proposed neural network.

Dual-frequency planar inverted-F antenna

Abstract: Cellular telephone handsets are now being designed to have dual-mode capabilities. In particular, there is a requirement for internal antennas for GSM and DCS1800 systems. This paper describes a novel planar dual-band inverted-F antenna for cellular handsets, which operates at the 0.9-GHz and 1.8-GHz bands. The dual-band antenna has almost the same size as a conventional inverted-F antenna operating at 0.9 GHz and has an isolation between bands of better than 17 dB. The bandwidths of the antenna are close to those required for the above systems. Good dual-band action is also obtained for other frequency ratios in the range of 1.3-2.4. Studies also show that the dual-band antenna can operate with one or two feeds. A finite-difference time-domain analysis has been shown to give calculated results close to those measured.

Neural network-based motion control of an underactuated wheeled inverted pendulum model.

Abstract: In this paper, automatic motion control is investigated for one of wheeled inverted pendulum (WIP) models, which have been widely applied for modeling of a large range of two wheeled modern vehicles. First, the underactuated WIP model is decomposed into a fully actuated second order subsystem Σa consisting of planar movement of vehicle forward and yaw angular motions, and a nonactuated first order subsystem Σb of pendulum motion. Due to the unknown dynamics of subsystem Σa and the universal approximation ability of neural network (NN), an adaptive NN scheme has been employed for motion control of subsystem Σa . The model reference approach has been used whereas the reference model is optimized by the finite time linear quadratic regulation technique. The pendulum motion in the passive subsystem Σb is indirectly controlled using the dynamic coupling with planar forward motion of subsystem Σa , such that satisfactory tracking of a set pendulum tilt angle can be guaranteed. Rigours theoretic analysis has been established, and simulation studies have been performed to demonstrate the developed method.