Transactions of the Institute of Measurement and Control

Liu, Ruochen; Li, Bingjie; Zhang, Lang; Jiao, Licheng

doi: 10.1177/0142331213520178pmid: N/A

The learning vector quantization (LVQ) algorithm is widely used in image compression because of its intuitively clear learning process and simple implementation. However, LVQ strongly depends on the initialization of the codebook and often converges to local optimal results. To address the issues, a new two-step LVQ (TsLVQ) algorithm is proposed in the paper. TsLVQ uses a correcting learning stage after LVQ to move the synaptic weight vector away from the incorrectly clustered training vector and towards the correctly clustered training vector. Experimental results show that TsLVQ outperforms kernel-based LVQ (KLVQ) and LVQ in terms of peak signal-to-noise ratio.

Jia, Bin; Xin, Ming

doi: 10.1177/0142331214523032pmid: N/A

In this paper, a new class of cubature information filters (CIF) is proposed for multiple sensor estimation. This new CIF generalizes the conventional third-degree CIF to attain higher estimation accuracy using a class of higher-degree cubature integration rules including the fifth-degree Mysovskikh’s spherical rule and the arbitrary degree radial rule. The statistical linear error propagation method is utilized to incorporate the high-degree cubature rule into the extended information filtering framework such that more accurate estimation can be achieved than the extended information filter. It also outperforms the unscented information filter as well as the particle filter. In addition, the high-degree CIF maintains close performance to the Gauss–Hermite information filter but uses significantly fewer quadrature points. As a result, the curse of the dimensionality problem existing in the tensor product-based Gauss–Hermite information filter can be greatly alleviated. Besides the improved estimation accuracy and computation efficiency, the high-degree CIF also exhibits the desirable robustness under unknown noise statistics. The proposed CIF is compared with other information filters via the state estimation of a two-phase permanent magnet synchronous motor and a target tracking problem, and demonstrates the best performance.

Wu, Jun; Zhu, Jigui; Yang, Linghui; Shen, Mengting; Xue, Bin; Liu, Zhexu

doi: 10.1177/0142331214524223pmid: N/A

Indoor GPS is one of the most popular positioning systems own to its high accuracy, real-time characteristics and multi-task management. In order to simplify the calibration process and extend its application, a single station model was presented recently. This paper proposes a novel ultrasonic ranging method used for the single station model. The traditional high-accuracy ultrasonic ranging method mainly uses a phase detection method by transmitting a multiple-frequency continuous wave. However, this method requires high accuracy of the phase detector and is still limited to small-scale application as a result of applying a continuous wave. Based on the constant time difference between the corresponding zero-cross points, this paper proposes a novel two-frequency pulse wave method, which can estimate the time of flight using the time differences between two received waves. Then a least squares estimation is used to eliminate random errors. Finally, an ultrasonic ranging experiment was conducted to validate its feasibility and stability.

Liu, Changqing; Chen, Wu-Hua

doi: 10.1177/0142331214528969pmid: N/A

The guaranteed cost control problem for a class of linear uncertain discrete-time impulsive systems is considered. The parametric uncertainties are assumed to be time-varying and norm-bounded. The problem is to design a robust state feedback controller such that the resulting closed-loop system is robustly exponentially stable, and the closed-loop value of a specified quadratic cost function is not more than a certain upper bound for all admissible uncertainties and for all admissible impulse time sequences. A sufficient condition for the existence of guaranteed cost state feedback controllers is derived via a time-varying Lyapunov function approach. This condition is expressed in terms of linear matrix inequalities. Furthermore, the problem of selecting a suboptimal guaranteed cost controller is formulated as a convex optimization problem. An example is provided to demonstrate the effectiveness of the proposed results.

Titi, Sufian Ishaq; Taylor, C James

doi: 10.1177/0142331214529151pmid: N/A

This article considers a combined pole assignment and multivariable decoupling control algorithm using discrete-time, non-minimum state space (NMSS) methods. In contrast to earlier research based on low-order linear models, the approach is applied to a nonlinear mean value internal combustion engine model with three control inputs, namely the throttle plate angle, injected fuel mass flow and spark advance angle. The controlled outputs are the air mass flow pressure, crank shaft speed and air–fuel ratio (AFR). It is well known that, for example, regulating the AFR to the stoichiometric value (i.e. 14.7) leads to a desirable balance between power output and fuel consumption, while reducing pollutant emissions. In this regard, the linear NMSS approach is straightforward to design for a range of performance requirements and yields comparable results to a more complex benchmark sliding mode control system. Furthermore, it retains a similar implementation structure to current production units, which are typically based on conventional proportional-integral compensation. The robustness to changing operating levels and disturbances, including an air leakage signal, are evaluated in simulation.

Yin, Jun; Yang, Yuwang; Cao, Hongxin; Zhang, Zhiyou

doi: 10.1177/0142331214531006pmid: N/A

Based on wireless sensor network (WSN) technology and crop growth simulation techniques, this paper shows the design and realization of an automatic monitoring and closed-loop control system in greenhouses. Firstly, a multi-hop network communication method based on clustering and simple medium access control that is suitable for the monitoring of a large-scale greenhouse environment is designed and analysed, and the simulation results show that its lifetime is 10% longer than LEACH (low energy adaptive clustering hierarchy) when 1% and 20% nodes die. Secondly, a physiological development day-based crop growth simulation model will be built to predict the tomato growth and make further decisions in adjusting the greenhouse climate. In order to obtain the model indicators, early experiments were carried out on four kinds of tomato variety, and the experiment results show that the proposed model has a higher accuracy than the effective temperature model on the root mean square error within 1–4 days, and on the mean absolute error within 2–4 days. Finally, according to the proposed methods, a comprehensive greenhouse dynamic monitoring and closed-loop control system with a 60 MC13213 nodes WSN was implemented. The implementation results show that with three AAA Ni–MH (nominal capacity 750 mAh) batteries, 80% nodes maintained a survival time of 45–60 days, and the model prediction compared with the observed value is at a high correlation efficient of 95%.

Meng, Wenchao; Yang, Qinmin; Sun, Youxian

doi: 10.1177/0142331214531008pmid: N/A

This paper deals with the power acquisition control of variable-speed wind energy conversion systems under inaccurate wind speed measurements. The control goal is to optimize the power capture from wind by tracking the maximum power curve. Firstly, the controller is designed for the case with known aerodynamic torque, which is a common assumption in many literatures. In this controller, the need for the exact knowledge of the system model is waived by using adaptive technologies. The chattering phenomenon in the generator torque, which can result in high mechanical stress, is avoided by adopting a modified robust term. Then, by utilizing an online approximator to learn an auxiliary term induced by the uncertain aerodynamics, the need for the exact knowledge of the aerodynamic torque is waived. Both of the proposed controllers are capable of providing good performance under inaccurate wind speed measurements. The control objective is obtained in the sense that the tracking error is guaranteed to converge to an arbitrarily small set. It is theoretically proved that all the signals in the closed-loop system are bounded via Lyapunov synthesis. Finally, the performance of our proposed controller is shown by simulating on a 1.5 MW three-blade wind turbine using the FAST (Fatigue, Aerodynamics, Structures, and Turbulence) code developed by the National Renewable Energy Laboratory.

Byrne, Steven; Naeem, Wasif; Ferguson, Stuart

doi: 10.1177/0142331214532002pmid: N/A

Manipulator motion planning is a classic problem in robotics, with a number of complete solutions available for their motion in controlled (industrial) environments. Owing to recent technological advances in the field of robotics, there has been a significant development of more complex robots with high-fidelity sensors and more computational power. One such example has been a rise in the production of humanoid robots equipped with dual-arm manipulators which require complex motion planning algorithms. Also, the technological advances have resulted in a shift from using manipulators in strictly controlled environments, to investigating the deployment of manipulators in dynamic or unknown environments. As a result, a greater emphasis has been put on the development of local motion planners, which can provide real-time solutions to these problems. Artificial Potential Fields (APFs) is one such popular local motion planning technique, which can be applied to manipulator motion planning, however, the basic algorithm is severely prone to local minima problems. Here, two modified APF-based strategies for solving the dual-arm motion planning task in unknown environments are proposed. Both techniques make use of configuration sampling and subgoal selection to assist the APFs in avoiding these local minima scenarios. Extensive simulation results are presented to validate the efficacy of the proposed methodology.

Yue, Ming; Wei, Xing; Li, Zhijun

doi: 10.1177/0142331214532999pmid: N/A

In this paper, we propose a novel control methodology based on zero-dynamics theory for a class of wheeled inverted pendulum (WIP) vehicles, which is efficient even in the presence of uncertain system frictions and dynamics parameters. The control schemes are elegantly constructed so that the WIP vehicle can successfully implement stabilizing of the posture (longitudinal and rotational movements), as well as hold the upright position of the vehicle body (tilt angle stability), only by the two control inputs with the aid of the design approach of zero-dynamics. In particular, the dynamics uncertainties, especially the friction effects, would deteriorate the control performance severely in practice. Therefore, we employ adaptive laws for the design parameters of zero-dynamics subsystem and uncertain coefficients of parametric frictions and dynamics. Consequently, the estimated frictions and dynamics are compensated through feedforward to obtain better control performance. To enhance the robustness of the system against parameter variations and external disturbances, sliding mode control techniques are applied to derive the specific algorithms, and then the closed-loop systems are proven to be globally asymptotically stable by Lyapunov techniques and LaSalle’s invariance theorem. In addition, simulation studies have been performed to demonstrate the feasibility and effectiveness of the proposed strategies, which illuminate the promising practical application potentiality of the designed WIP vehicle control system.

Gaied, Khalaf Salloum

doi: 10.1177/0142331214533121pmid: N/A

This paper presents a wavelet-based prognosis technique for fault-tolerant control (FTC) of an induction motor (IM) drive. The wavelet is used for analysis in fault detection due to its good filtering characteristics, localized capability and superior de-noising. The fault-tolerant algorithm is applied to a 1-kW IM to maintain the performance of the machine at an acceptable level. The FTC algorithm has been implemented in a new topology to ensure the best performance recovery in case of open and short stator winding faults, as well as speed sensor faults. In the worst-case scenario of operation, a protection stage has been implemented to stop motor operation. Experimental results demonstrate the reliability and effectiveness of the technique for diagnosis and prognosis of stator and speed sensor faults. The Texas Instrument TMS 320F28335 is used for the real-time implementation of the proposed wavelet-based prognosis and the FTC of the IM drive under different fault situations.