Permanent URI for this collection
Browse
532 results
Search results
Publication Improving joint PD control of single-link flexible robots by strain/tip feedback(1996) Ge, S.S.; Lee, T.H.; Zhu, G.; ELECTRICAL ENGINEERINGThis paper considers improving the tip motion performance of a joint-PD controlled single-link flexible robot by respectively introducing nonlinear base-strain & tip-acceleration feedback. The stability is theoretically proven directly based on the Partial Differential Equations (PDEs) of the system. The controller is independent of system parameters and thus possesses the stability robustness to parameter uncertainties. Using base-strain/tip-acceleration feedback is motivated by the fact that strain gauge and accelerometer have been widely used in control of robots with flexible links, and are comparatively easy for engineering implementation. Simulations are provided to show that the tip motion performance are greatly improved by the two kinds of nonlinear feedback.Publication Real-time neural-network-based control of a gyro-mirror line-of-sight stabilization platform(1998) Lee, T.H.; Ge, S.S.; Hang, C.C.; Zhao, Q.; Au, A.S.; ELECTRICAL ENGINEERINGIn this paper, we consider the real-time neural-network-based control of a gyro-mirror line-of-sight stabilization platform. This is an example of a multivariable nonlinear servomechanism encountered in certain practical applications, and the appropriate neural-network-based modelling and controller design are discussed. In the paper, real-time experimental results in applying the proposed controller to a pilot-scale gyro-mirror platform are presented to demonstrate its effectiveness. These experiments also serve to verify the analytical results in a proptotype real-time application.Publication Modeling and control of a flexible riser with application to marine installation(2012) Ge, S.S.; He, W.; Zhang, S.; ELECTRICAL & COMPUTER ENGINEERINGThis paper investigates the control problem of a marine riser installation system. The riser installation system consisting of a vessel, a flexible riser, a subsea payload is modeled as a distributed parameter system with one partial differential equation (PDE) and four ordinary differential equations (ODEs). Based on the Lyapunov's direct method, robust adaptive boundary control is proposed at the top and bottom boundary of the riser to position the subsea payload to the desired set-point and suppress the riser's vibration. With the proposed control, uniform boundedness of the steady state error between the boundary payload and the desired position is achieved by suitably choosing the design parameters. Numerical simulations are presented for demonstrating the effectiveness of the proposed control. © 2012 AACC American Automatic Control Council).Publication Mutual-reinforcement document summarization using embedded graph based sentence clustering for storytelling(2012-07) Zhang, Z.; Ge, S.S.; He, H.; ELECTRICAL & COMPUTER ENGINEERINGIn this paper, a document summarization framework for storytelling is proposed to extract essential sentences from a document by exploiting the mutual effects between terms, sentences and clusters. There are three phrases in the framework: document modeling, sentence clustering and sentence ranking. The story document is modeled by a weighted graph with vertexes that represent sentences of the document. The sentences are clustered into different groups to find the latent topics in the story. To alleviate the influence of unrelated sentences in clustering, an embedding process is employed to optimize the document model. The sentences are then ranked according to the mutual effect between terms, sentence as well as clusters, and high-ranked sentences are selected to comprise the summarization of the document. The experimental results on the Document Understanding Conference (DUC) data sets demonstrate the effectiveness of the proposed method in document summarization. The results also show that the embedding process for sentence clustering render the system more robust with respect to different cluster numbers. © 2011 Elsevier Ltd. All rights reserved.Publication Adaptive control of a class of uncertain electrostatic microactuators(2007) Tee, K.P.; Ge, S.S.; Tay, E.H.; ELECTRICAL & COMPUTER ENGINEERING; MECHANICAL ENGINEERINGIn this paper, adaptive state feedback control is presented for a class of single-degree-of-freedom (1DOF) electrostatic microactuator systems which can be actively driven in both directions. The control objective is to track a reference trajectory within the air gap without knowledge of the plant parameters. By a suitable change of coordinates, the system is represented in parametric strict feedback form, for which adaptive backstepping is performed to achieve asymptotic output tracking. To deal with the constrained position space, which precludes contact of the movable plate and the electrodes, special barrier functions are employed in Lyapunov synthesis. All closed loop signals are ensured to be bounded. Simulation studies demonstrate the effectiveness of the proposed control. ©2007 IEEE.Publication Approximate dissipative Hamiltonian realization and construction of local Lyapunov functions(2007-02) Wang, Y.; Cheng, D.; Ge, S.S.; ELECTRICAL & COMPUTER ENGINEERINGThe key in applying energy-based control approach is to be able to express the system under consideration as a dissipative Hamiltonian system, i.e., to obtain Dissipative Hamiltonian Realization (DHR) for the system. In general, the precise DHR form is hard to obtain for nonlinear dynamic systems. When a precise DHR does not exist for a dynamic system or such a precise realization is difficulty to obtain, it is necessary to consider its approximate realization. This paper investigates approximate DHR and construction of local Lyapunov functions for time-invariant nonlinear systems. It is shown that every nonlinear affine system has an approximate DHR if linearization of the system is controllable. Based on the diagonal normal form of nonlinear dynamic systems, a new algorithm is established for the approximate DHR. Finally, we present the concept of kth degree approximate Lyapunov function, and provide a method to construct such a Lyapunov function. Example studies show that the methodology presented in this paper is very effective. © 2006 Elsevier B.V. All rights reserved.Publication Queues and artificial potential trenches for multirobot formations(2005-08) Ge, S.S.; Fua, C.-H.; ELECTRICAL & COMPUTER ENGINEERINGIn this paper, we present a novel approach for representing formation structures in terms of queues and formation vertices, rather than with nodes, as well as the introduction of the new concept of artificial potential trenches, for effectively controlling the formation of a group of robots. The scheme improves the scalability and flexibility of robot formations when the team size changes, and at the same time, allows formations to adapt to obstacles. Furthermore, for multirobot teams to operate successfully in real and unstructured environments, the instant goal method is used to effectively solve the local minima problem. © 2005 IEEE.Publication System design and hardware integration of social robot Nancy(2011) Ge, S.S.; Liew, C.F.; Li, Y.; Yang, J.; ELECTRICAL & COMPUTER ENGINEERINGThis paper gives insight into the system design and hardware architecture of Nancy, which is a social robot in humanoid form that is currently being developed in Social Robotics Lab (SRL), Interactive and Digital Media Institute, National University of Singapore. In contrast to industrial robots, the design of Nancy focuses on safety issue, social interaction and attractive appearance. In order to achieve these objectives, Nancy is designed to be user-friendly and approachable. The mechanical structure of Nancy is presented in the beginning. After that, the hardware architecture which includes processor system, vision system, audio system, communication system, power system, and motion system, is described systematically. Finally, the roles of integrated sensors are described and implemented works are summarized. © 2011 IEEE.Publication Adaptive neural-fuzzy control of uncertain nonholonomic systems(2005) Wang, Z.P.; Ge, S.S.; Lee, T.H.; ELECTRICAL & COMPUTER ENGINEERINGAn adaptive neural-fuzzy controller is presented in this paper for mechanical systems with nonholonomic constraints in the presence of uncertainties about plant parameters. The controller is designed based on a reduced model. The neural-fuzzy (NF) controller is constructed in order to eliminate the need for dynamic modeling and error prone process in obtaining the regressor matrix. The proposed controller guarantees that the system motion asymptotically con- verges to the desired manifold. Numerical simulation are conducted to verify the effectiveness of the proposed method. Copyright © 2005 IFAC.Publication Energy-based robust controller design for multi-link flexible robots(1996-10) Ge, S.S.; Lee, T.H.; Zhu, G.; ELECTRICAL ENGINEERINGA class of robust stable controllers to control the tip position of a multi-link flexible robot is presented. In contrast to traditional model-based methods, the controllers are derived by using a basic relationship of system energy and are independent of the system dynamics. The approach allows controller design in the absence of a system model (which is very complicated in the case of the multi-link flexible robot) and provides great freedom in feedback control design. Further, the controllers are easy to implement in practice because, depending on actual instrumentation, all the signals can be chosen to be measurable. Simulation results of a two-link flexible robot are provided to show the effectiveness of the presented approach. Copyright © 1996 Elsevier Science Ltd.