next day
all days

View: session overviewtalk overview

08:30-09:30 Session 1: Welcome and Plenary - Arend Schwab
Bicycle Dynamics and Control
SPEAKER: Arend Schwab

ABSTRACT. Riding a bicycle is an acquired skill. At rest the system is highly unstable yet, given some forward speed, it is easy to stabilize. Over the past 140 years scores of people have been attracted to this subject, either for a dissertation, a hobby or sometimes as part of a life’s work. Unfortunately, few results agree and there is little generality on the basic features that make a bicycle stable and why some bicycle are easier to control then others. In this talk I will focus first on the dynamics of the bicycle after which I will address some interesting and open control issues.

10:15-12:15 Session 2A: Aerospace Control
Application of a Fuzzy Logic Controller with Manipulation Scaling Factor For Speed Control On a Small Scale Turbojet Engine
SPEAKER: Sinan Ekinci

ABSTRACT. There are various studies to control turbojet engines with FLC (Fuzzy Logic Control) due to their highly nonlinear behavior with respect to engine spool speed, but the control approaches for defined problems were typical PI (Proportional-Integral) type FLC for many of them without considering the engine spool speed. In this paper, a second order function which gives manipulation scaling factor according to actual spool speed is added to PI (Proportional-Integral) type FLC for speed control on a small turbojet engine to improve control performance. Also the number of configuration parameters of FLC is decreased by this way. The parameters of FLC and manipulation scaling factor function are optimized by using genetic algorithm to obtain desirable transient response and steady state performance. The designed controller is implemented on an in-house, embedded electronic control unit designed for SSTE (Small Scale Turbojet Engine). The designed FLC’s are tested on a SSTE and performance comparison between the controllers is discussed.

Efficient Aeroservoelastic Modeling and Control using Trailing-Edge Flaps of Wind Turbines
SPEAKER: Bing Feng Ng

ABSTRACT. The increase in size of wind turbines necessitates the accurate predictions of aeroelastic effects and requires the use of localized active control techniques to overcome unnecessary loadings and oscillations. This paper presents a computationally efficient aeroservoelastic modeling approach for dynamic load alleviation in large wind turbines with trailing-edge aerodynamic control surfaces. The resulting dynamic aeroelastic model is written directly in a state-space formulation suitable for model reduction and control synthesis. Trailing-edge flaps are modeled directly in the unsteady aerodynamics and the linear model of a single rotating blade is used to design a Linear-Quadratic-Gaussian regulator for minimizing the root-bending moments, which is shown to provide load reductions of about 20\% in closed-loop on the full wind turbine non-linear aeroelastic model.

A Class of Nonlinear Unknown Input Observer for Fault Diagnosis: Application to Fault Tolerant Control of an Autonomous Spacecraft
SPEAKER: Robert Fonod

ABSTRACT. In this paper, the problem of Nonlinear Unknown Input Observer (NUIO) based Fault Detection and Isolation (FDI) scheme design for a class of nonlinear Lipschitz systems is studied. The proposed FDI method is applied to detect, isolate and accommodate thruster faults of an autonomous spacecraft involved in the rendezvous phase of the Mars Sample Return (MSR) mission. Considered fault scenarios represent fully closed thruster and thruster efficiency loss. The FDI scheme consists of a bank of NUIOs with adjustable error dynamics, a robust fault detector that is based on judiciously chosen frame and an isolation logic. The bank of observers is in charge of confining the fault to a subset of possible faults and the isolation logic makes the final decision about the faulty thruster index. Finally, a thruster fault is accommodated by re-allocating the desired forces and torques among the remaining healthy thrusters and closing the associated thruster valve. Monte Carlo results from ”hight-fidelity” MSR industrial simulator demonstrate that the proposed fault tolerant strategy is able to accommodate thruster faults that may have effect on the final rendezvous criteria.

Robust Model Following Control Design for Missile Roll Autopilot

ABSTRACT. This paper represents a robust model following control method augmented with error integration and Luenberger observer for anti-air missile roll autopilot designed using optimal control laws. The design is shown to be robust to external disturbance, noisy measurements and sensor lags by frequency domain analysis. The regulation performance of the controller is presented by simulations.

H infinity Control for Quadrotor Attitude Stabilization
SPEAKER: Wesam Jasim

ABSTRACT. This paper proposes a state feedback controller for the attitude stabilization problem of quadrotors. The quadrotor attitude is represented by unit quaternion and the disturbance of the attitude model is taken into consideration. A robust controller is synthesized via H infinity optimal design approach. Solving the nonlinear H infinity optimal control problem using state feedback is meltdown to finding a solution to a Hamilton-Jocobi inequality. Based on the quadrotor attitude dynamics, an appropriate parameterized Lyapunov function is selected and the corresponding state feedback controller is derived. Then the parameters are found from the Hamilton-Jocobi inequality. The resultant state feedback controller can lead to the closed nonlinear system having L2 gain less than or equal to a constant gamma, and establish the asymptotically stability of the closed nonlinear system without external disturbance. The simulation provides the results to show the stability and the robust performance against to disturbance.

Quadrotor Control for Trajectory Tracking in Presence of Wind Disturbances
SPEAKER: Oualid Araar

ABSTRACT. Quadrotor helicopters have been the focus of many research works during the last decade. Thanks to their numerous appealing features, their applications are increasingly reported in the literature. Their main limitation, however, is their lightweight which makes them sensitive to external disturbances. The aim of this paper is to achieve robust trajectory following for a Quadrotor UAV, in the presence of wind disturbances. Two nonlinear controllers are presented, the first controller is based on Feedback Linearisation, while the second is designed using a Backstepping approach. Numerical simulations are provided to support the controllers design and assess their robustness to wind gust disturbances.

10:15-12:15 Session 2B: Mobile Robotics
Stabilisation and Manoeuvre of electrically powered pedestrian controlled Uniaxial Vehicles for Goods Transport

ABSTRACT. Uniaxial vehicles for goods transport such as hand trucks offer intuitive manoeuvrability with little space requirements. This is why this class of vehicles is popular with many transport tasks. The disadvantage of this pedestrian controlled means of transportation is the need for the user to apply force for stabilisation and propulsion. A solution approach is to equip the uniaxial vehicle with a controlled drive system providing force for both stabilisation without external support and pedestrian controlled propulsion with low interaction forces even with high payload. To realise this automatic balancing the frame's pitch angle must be adopted with every change of load. This is a fundamental difference to uniaxial vehicles without payload or for passenger transportation. To make possible intuitive and convenient pedestrian controlled manoeuvre the conroller must set translational and rotational speed based on low interaction forces applied by the user. To allow for cheap production we developed a balance and manoeuvre controller not requiring sensor information about both load's absolute weight and user interaction forces. Main components of this concept are a Kalman state estimator for identification of frame's and load's combined center of gravity's pitch angle and a sophisticated program sequence switching automatically between operating modes for reload and manoeuvre. The control concept was designed and implemented using our new uniaxial vehicle system for urban goods transport. This vehicle is the first electrically powered hand truck that both balances adaptive to changing load and allows for pedestrian controlled manoeuvre without need to operate control levers for setting speed or steering angle. The experimental results show the correct operation of the approach. The results presented in this paper provide a basis for further research in the field of control strategies for stable and convenient operation of this type of vehicle under real-life conditions.

Design and Implementation of Networked Real-time Control System with Image Processing Capability
SPEAKER: Guo-Ping Liu

ABSTRACT. This paper describes the design and implementation of an OMAP3530-based networked control system which has real-time image processing capability. It integrates an ARM core and a DSP core in the controller. The ARM core with Linux operating system is responsible for applying functional services and the DSP core is responsible for executing image processing algorithms. The due-core structure makes the novel controller available for advanced control and image processing so that the designed control system expands the application of the existed networked controllers to machine vision. Finally, a reliable application for vision-based mobile robot location demonstrates the effectiveness of the designed system .

Noncausal Finite Time Interval Iterative Learning Control Law Design
SPEAKER: Xuan Wang

ABSTRACT. Iterative learning control has been developed for processes or systems that complete the same finite duration task over and over again. The exact mode of operation is that after each execution is complete the system resets to the starting location ready for the start of the next one. Each execution is known as a trial and the duration the trial length. Once each trial is complete the information generated is available for use in computed the control input for the next trial. This paper uses a two-dimensional systems setting to give some design oriented results on how to maximize the use of previous trial information.

Neural Network based Reinforcement Learning Control of Autonomous Underwater Vehicles with Control Input Saturation
SPEAKER: Rongxin Cui

ABSTRACT. In this paper, the trajectory tracking control of the autonomous underwater vehicle (AUV) is investigated in discrete time, for ease of digital computer calculation. A reinforcement learning technique in the controller and a second neural network to estimate the reinforcement signals is used in this paper to guarantee the best possible tracking performance of the AUV. Simulation results show that the errors convergence to a adjustable neighborhood around zero, and optimization has been achieved in the sense of reinforcement learning.

Inverse kinematics for a redundant robotic manipulator used for nuclear decommissioning
SPEAKER: Pierre Besset

ABSTRACT. The article develops a generic framework for the control of a dual-manipulator mobile robotic system for nuclear decommissioning, with a particular focus on the inverse kinematics and trajectory planning. A six Degrees-Of-Freedom (DOF) kinematic model for each manipulator is described, including the rotation of the end-effector. On this basis, the forward kinematic equations are relatively straightforward to determine. However, the redundant nature of the manipulator and its particular geometry exclude the possibility of a closed-form analytic solution to the inverse kinematics. Hence, previous studies into advanced automatic control using the device have locked off selected control valves. For example, recent research into nonlinear, state-dependent control systems for the hydraulic actuators have been limited to just three DOF. By contrast, this new study considers all the joints of each manipulator, basing the resolution of the inverse kinematics on the iterative Jacobian transpose method. Preliminary experimental and simulation results highlight the potential utility of the system by performing standard tasks such as pick and place operations.

Skid Steering Mobile Robot Modeling and Control

ABSTRACT. In this paper, a Linear Quadratic Regulator (LQR) controller augmented with a feed-forward part is designed for controlling the dynamics of a skid steering mobile robot. The controller is simple in terms of design and implementation in comparison with complex nonlinear control schemes that are usually designed for this system. Moreover, it provides good performance for the plant comparable with a nonlinear controller based on the inverse dynamics which depends on the availability of an accurate model describing the system. Simulation results are included.

10:15-12:15 Session 2C: Adaptive, Nonlinear & Time Varying Control
LPV Gain-Scheduling Control with Time-Varying Sampling Time for Rejecting Nonstationary Harmonically Related Multisine Disturbances

ABSTRACT. A method to design linear parameter-varying (LPV) controllers with time-varying sampling time to reject nonstationary harmonically related multisine disturbances is presented in this work. Unlike common approaches, the design considers the disturbance as a time-invariant model while the plant is an LPV system with sampling time as the scheduling parameter in linear fractional transformation (LFT) form. Only one scheduling parameter is used for this approach, independently of the number of frequencies. Good disturbance rejection is achieved by simulation.

Adaptive unknown input reconstruction scheme for Hammerstein-Wiener systems

ABSTRACT. In this paper an adaptive time-varying filter for unknown/unmeasurable input reconstruction is proposed. The algorithm is based on parity-equations and is applicable to Hammerstein-Wiener systems, i.e. systems composed of a linear dynamic part followed and preceded by a memoryless nonlinearity. An error-in-variables case is considered, i.e. known input and output signals are both subjected to measurement uncertainties. The scheme forms an extension to a filter previously proposed by the authors. As the input reconstruction involves transformation of noisy signals through memoryless static functions, measurement noise is either amplified or reduced, depending on the gradient of the nonlinear function. Thus, in the proposed scheme the bandwidth of the filter is adjusted depending on the operating point allowing for a trade-off between noise attenuation and a phase lag.

Improving Lateral Stability of a Motorcycle via Assistive Control of a Reaction Wheel

ABSTRACT. A simplified non-linear motorcycle model forms the plant. The control goals are roll angle tracking, disturbance rejection and minimal energy consumption. The control is designed to cooperate with the rider, either by designing it as a parallel control schemer, or by using the reaction wheel to improve the motorcycle stability. Two control designs are tested: linear optimal control applied to the nonlinear model, and a structured controller approach based on an uncertainty model. The next step is to test this control an electric bicycle fitted with a reaction wheel.

Online identification of piecewise affine systems

ABSTRACT. In this note, we present a novel procedure for online identification of piecewise affine systems. We show how to identify the submodels of a piecewise affine system online, under the assumption of known switching boundaries. The adaptively identified subsystem parameters can be used for indirect adaptive control or online plant diagnosis and fault detection. The procedure proposed in this note extends the well-studied series-parallel parameter identifiers in adaptive control to piecewise affine systems. Challenges that had to be solved for this extension are the additional constant input of the affine subsystems and the switching behavior which is a potential source of instability. After the problem description, we sketch the proof for both stability and parameter convergence of the proposed algorithm. Finally, numerical simulations validate the results and analyze the convergence speed of the estimated parameters.

A Generalized Barbalat Lemma Based on a Persistently Exciting Condition
SPEAKER: Ti-Chung Lee

ABSTRACT. This paper studies attractivity employing a new persistently exciting (PE) condition. Under an integral condition, the proposed PE condition is shown to be a sufficient condition to guarantee attractivity. Then, it is applied to derive a generalized Barbalat lemma. Based on this result, several generalizations of Barbalat lemma are then proposed. In particular, the standard assumption that requires uniform continuity on the positive real axis can be relaxed by admitting countable discontinuous points. Moreover, the integrand of the assumed integral condition can be a composition of a targeted function and a time-varying function. An interesting example is provided to illustrate the usefulness of the proposed results.

Adaptive Tracking Control for a New Mobile Manipulator Model

ABSTRACT. The adaptive trajectory and force tracking control is considered for a new mobile manipulator under both holonomic and affine constraints with the presence of uncertainties and disturbances in this paper. A new state transformation is proposed to deal with the affine constraints, and then based on a suitable reduced dynamic model, adaptive controller is presented to ensure that the states of closed-loop system asymptotically track to desired trajectories while the constraint force remains bounded by tuning design parameters. Detailed simulation results confirm the effectiveness of the control strategy.

10:15-12:15 Session 2D: Special Session - Predictive Control for Aerospace: Lessons Learned
Predictive Control for Aerospace: Lessons Learned

ABSTRACT. The session will bring together MPC experts from across 3 different areas of application within the field of aerospace: space, air traffic management, and UAV control. Their unifying challenge is that the MPC optimization now involves the design of a trajectory for the motion of a vehicle. This requires the framework of MPC to be combined with motion planning ideas from robotics. Common challenges include handling non-convex operating regions, limited computing power, and the need to be scalable with the number of vehicles involved. The session will aim to take a tutorial and reflective approach, using examples to illustrate lessons learned.

10:15-12:15 Session 2E: Special Session - Hardware Connectivity with MATLAB and Simulink
Hardware Connectivity with MATLAB and Simulink
SPEAKER: Owen McAree

ABSTRACT. MATLAB® and Simulink® connect to a wide range of hardware platforms for project-based learning, signal processing, computer vision, communications, data acquisition, instrument control, embedded systems, and more. These include low-cost hardware platforms like Arduino® and Raspberry Pi™, embedded systems like ARM®- and Zynq®-based architectures, or high-end systems for real-time testing. This session showcases some exciting projects made possible with this connectivity.

13:00-15:00 Session 3A: Flight Control, Guidance & Operations
Accounting for the Effect of Ground Delay on Commercial Formation Flight
SPEAKER: Thomas Kent

ABSTRACT. This paper explores the impact ground delay can have on designing and executing rendezvous operations such as formation flight. For a given formation pairing the route is fixed, speed policies are then generated to compensate for uncertainty in take-off times. Value Iteration is used to solve both the deterministic and stochastic Dynamic Programming problem for an entire state-space. The final optimal policies determine what course of action aircraft should take for any realization of delay. Finally a comparative case study shows that even with delay, formation flights can reach expected savings of 6.1% against flying solo

Switched Control for a Fighter Aircraft
SPEAKER: Emre Kemer

ABSTRACT. This paper presents the design of switched feedback controllers for load factor tracking. The controller consists of a discrete set of state feedback gains which are switched according to some system states such as Mach number and altitude. Attention is restricted to the design of a longitudinal controller for the ADMIRE fighter benchmark model. Each state feedback gain is designed to track the load factor, nz, where integral action is added to enforce zero steady state error. The stability of the overall closed-loop system is established thanks to piecewise linear quadratic Lyapunov functions and dwell time theory.

Real-time Obstacle Collision Avoidance for Fixed Wing Aircraft Using B-splines

ABSTRACT. A real-time collision avoidance algorithm is developed based on parameterizing an optimal control problem with B-spline curves. The optimal control problem is formulated in output space rather than control or input space, this is feasible because of the differential flatness of the system for a fixed wing aircraft. The flat output trajectory is parameterized using a B-spline curve representation. In order to reduce the computational time of the optimal problem, the aircraft and obstacle constraints are augmented in the cost function using a penalty function method. The developed algorithm has been simulated and tested in MATLAB/Simulink.

Longitudinal guidance of UAVs using sliding mode approach

ABSTRACT. This paper presents a sliding mode based longitudinal guidance scheme for Unmanned Aerial Vehicles (UAVs). A nonlinear sliding surface is proposed here for altitude control of UAVs. Longitudinal guidance law based on traditional linear sliding surface cannot provide good performance for both large and small errors in altitude, hence a non-linear surface is proposed here. The proposed sliding surface gives good performance in level cruising as well as during climb/decent phases. The Proposed guidance scheme is implemented on a 6-degrees-of-freedom (6-dof) simulation of a UAV and different scenarios of climb, decent and level cruise phase is simulated.

Low Order H-infinity Controllers for a Quadrotor UAV
SPEAKER: Hasan Basak

ABSTRACT. This paper addresses the designs of low-order controllers for an Unmanned Aerial Vehicle (UAV) quadrotor using fixed and structured H-infinity optimization techniques. Closed-loop performance, computational efficiency and tracking performance of low order controllers are compared with those of a standard full order H-infinity controller.

Full Linear Control of a Quadrotor UAV, L2 vs L∞ Norm
SPEAKER: Oualid Araar

ABSTRACT. Quadrotors low level control has attracted the attention of many research works over the last decade. In most of previous works, position stabilisation was achieved using nonlinear control techniques. In this paper, two full linear techniques based on optimal control theory are presented. The first, based on the L2 norm, is a Linear Quadratique Servo (LQ- Servo) controller. The second one, optimizes the L norm and is designed using H∞ control approach. Emphasis is placed on the robustness to external disturbances and particularly wind gusts. Simulation results based on the full nonlinear model of the quadrotor are presented as a basis of comparison between the two controllers.

13:00-15:00 Session 3B: Robust and Optimal Control
Synthesis of Biquadratic Impedances with A Specific Seven-Element Network

ABSTRACT. The object of this paper is to identify the non-regular biquadratics that can be realized by a specific seven-element network. Renewed interest in the classical network synthesis arises in the synthesis of passive mechanical impedances. Recent work has shown that five-element networks are capable of realizing all regular biquadratics and a small region of non-regular biquadratics. Based on the study of series-parallel six-element synthesis of biquadratics, a specific seven-element network was proposed, which is totally to realize a large range of nonregular biquadratics. The complete realizability conditions for this network will be defined and expressed in canonical form for biquadratics. The non-regular realizable region will then be explicitly characterized.

Robust fault detection for switched systems with time-varying delay using delta operator approach

ABSTRACT. This paper addresses the problem of fault detection for switched systems with time-varying delay using delta operator approach. By means of the delta operator systems, a fault detection observer is used as the residual generator. Based on the switched Lyapunov function approach, sufficient conditions for the existence of the fault detection observer are given in terms of linear matrix inequalities (LMIs). An example is provided to show the effectiveness and applicability of the proposed method.

Guaranteed cost control of uncertain impulsive switched systems with nonlinear disturbances
SPEAKER: Cunjia Liu

ABSTRACT. This paper discusses a guaranteed cost control problem of an uncertain impulsive switched system with nonlinear disturbances. The contribution of this study is to extend guaranteed cost control from traditional systems to uncertain impulsive switched systems with nonlinear disturbances. Firstly, an uncertain impulsive switched system with nonlinear disturbances is stated and some important definitions are given. Secondly, the stability of the closed-loop uncertain impulsive switched system with a linear quadratic guaranteed cost control law is proved and the upper boundary of performance cost is discussed. Especially, different from traditional stability analysis of impulsive switched systems without disturbances, above conclusions are given under a nonlinear disturbance. Thirdly, Linear Matrix Inequality (LMI) formulation is utilized to design the optimal guaranteed cost control law of the uncertain impulsive switched system. Finally, simulations are carried on to demonstrate the proposed algorithm.

Output Feedback Stabilization of an Inertia Wheel Pendulum Using Sliding Mode Control
SPEAKER: Nasir Khalid

ABSTRACT. This paper investigates the application of a novel robust output feedback stabilizing controller to an Inertia Wheel Pendulum (IWP)- a bench mark problem in the control of under-actuated mechanical systems. The proposed scheme uses a Sliding Mode Control (SMC) design for the stabilization of an IWP having unstable zero dynamics. Starting with the system equations of motion in standard form, a normal form representation of the system is derived, and a sliding mode controller is synthesized for robust stabilization of the system. The state feedback control design is extended to an output feedback control using a high gain observer. It is shown that the system response under the output feedback controller converges to that of the state feedback controller. It is also shown that estimated states converge to actual states arbitrarily fast.

Robust and Optimal Stabilization of Uncertain Linear Systems using LQR Methods
SPEAKER: Salman Zaffar

ABSTRACT. Robust and Optimal stabilization of a class of a Linear Time Invariant (LTI) systems is discussed which exhibit linear time varying (LTV) behavior due to the presence of parametric variations and uncertainties. Linear quadratic methods offer global optimal control solutions for LTI systems. Such methods offer optimal solutions only locally for systems which become LTV due to parametric uncertainties. We propose that such an LTV system can be divided into two or more LTI systems in terms of the operating conditions ranging from nominal to most uncertain. In our proposed approach, two linear quadratic regulators would each be separately designed for nominal operating conditions and the uncertain conditions in a system. It is shown that the switching between the two regulators depend upon the size of the uncertainty. A machine learning algorithm such as the support vector machine has been used to design a switching surface as a function of only the parametric uncertainties of the system. Extended high gain observers are used to estimate the parametric uncertainties needed for switching between the two regulators. Simulation results are included to demonstrate the performance of the proposed approach.

Mixed H2/H∞ Feedback Control of Multivariable Dynamically Substructured Systems
SPEAKER: Chi-Lun Wang

ABSTRACT. Dynamically substructured system testing method divides an original system into several substructures. In the numerical substructure, linear components are simulated via real-time computation. In the physical substructure, a transfer system, which includes actuators and sensors, is installed to interface the numerical and physical parts. During the test, unwanted disturbances and noise from the actuator inevitably cause synchronization errors between the outputs of numerical and physical substructures at the interface, and consequently result in unsuccessful tests. Therefore, advanced control using the mixed H2/H∞ algorithm is proposed in this paper to ensure optimal and robust synchronization. The synchronization problem is transformed to tracking design according to a numerical-substructure-based framework and is solved based on Riccati-like equations and linear matrix inequality. A multivariable mass-spring-damper substructured system is developed to verify the proposed control strategy via numerical studies.

13:00-15:00 Session 3C: Modelling, Identification and Estimation
Adaptive Neuro-Fuzzy Method to Estimate virtual SI Engine Fuel Composition using Residual Gas Parameters
SPEAKER: Kinyip Chan

ABSTRACT. This paper addresses computer engine simulation with an adaptive neuro fuzzy method to estimate the fuel composition by using the residual gas information. The availability of fuel composition is divided into. The actual ready mixed composition provided from suppliers, named in gasoline, contains different name hydro-carbon atoms. The composition is unknown, but varies with standard based on engine combustion profile. This study researches the idea of further engine control on fuel composition to improve engine performance and reduce on emissions. Fuel composition can be estimated using combustion product after gas exchange. This study investigates a computer based engine model which uses Adaptive Neuro-Fuzzy Interface System (ANFIS) to identify the fuel composition. The residual composition contains the level of Carbon Dioxide (CO2), Oxygen (O), Carbon Monoxide (CO) and Nitric Oxide (NO) which developed the network to estimate the Hydrocarbon level of original fuel input. Results show that ANFIS control is reasonably distinguish three different fuel compositions in the tests.

Operating point adjustment procedure for a class of continuous-time bilinear models
SPEAKER: Ivan Zajic

ABSTRACT. This paper presents a proposal of a novel parameter adjustment procedure for a class of continuous-time bilinear models. This procedure allows the user to adjust the operating point dependent dynamic characteristics of the considered bilinear model to match the dynamic characteristics of the same model however operated at different working point. This procedure finds its application in system identification and simulation of bilinear models as well as in the area of bilinear model based control.

A 2-Dimensional Hammerstein model for Heating and Ventilation Control of Conceptual Thermal Zones

ABSTRACT. The research behind this article aims to reduce the operational costs and energy consumption of closed-environment growing systems, or grow-cells. Essentially a sealed building with a controlled environment, and insulated from outside lighting, grow-cells are configured to suit the particular crop being produced. The article briefly reviews the concept and preliminary work in relation to a prototype being developed by the authors and collaborating industry partner. Here, limitations in the temperature control system can lead to significant thermal gradients and poor efficiency. In this regard, the main focus of the article concerns a novel approach to thermal modelling that includes quantitative identification of spatial zones with similar thermal characteristics and the estimation of steady state temperature functions based on the heater and fan input voltages. These are combined with either a linear or a state-dependent parameter model, to represent the transient response. This approach yields a Hammerstein type model which, in this article, is optimised and evaluated using experimental data collected from 30 thermocouples distributed around an environmental test chamber.

Efficient Gaussian Process modelling of section weights in polymer stretch blow moulding

ABSTRACT. The injection stretch blow moulding (ISBM) has been widely applied in polyethylene terephthalate bottles production process. The modelling of the blowing conditions is important for the control of the process. In this paper, a nonparametric modelling method namely Gaussian Process is employed and applied to estimate the section weights during the process. A key issue in Gaussian Process modelling is to iteratively compute the covariance matrix inversion which is often extremely time-consuming as the optimization of the Guassian process requires to call the marginal likelihood function and its gradient function. The matrix decomposition methods like Cholesky and QR decomposition have shown to be able to significantly reduce the computation complexity in solving linear problems, and they are also used in Gaussian process modelling in computing the product of a matrix inversion with a column vector in this paper. The algorithm complexities are analysed and compared, and it is shown that the proposed method is particularly useful as the matrix size increases. The proposed method is then used to model the section weights in polymer stretch blow moulding, and the results confirm the efficiency of the proposed method in significantly saving the computation resources.

Output observer design for fault detection

ABSTRACT. In this paper, we propose an output observer design methodology for linear single output systems for fault detection purposes. Unlike traditional observers that are based on the the state-space representation of the system, the proposed observer design is based on the input-output representation of the system. The main advantage of the proposed observer is its simplicity of design and it has a lower order that the original system.

Fuzzy Universal Model Approximator for Distributed Collector Solar Field Control

ABSTRACT. This paper studies the problem of controlling a concentrating parabolic solar collector which consists in making the outlet oil temperature to track a set reference. For this purpose, a novel fuzzy universal approximate model is introduced in order to accurately reproduce the system dynamics behavior. It is a low order state space representation that has been derived from the partial differential equation describing the oil temperature evolution using fuzzy transform theory. The resulting set of ordinary differential equations simplifies the system analysis and the control law design and is suitable for real time controller implementation. A non linear control has been synthesized based the proposed fuzzy model resorting to Lyapunov Control functions. Simulation results show good performance of the proposed model.

13:00-15:00 Session 3D: Invited Session - Flow Control
Wall-transpiration feedback control of laminar streaks using an adjoint approach

ABSTRACT. This paper presents theoretical and numerical results on the penetration of small-amplitude free-stream vortical disturbances into an incompressible laminar boundary layer, and the wall-based adjoint feedback control of the resulting streamwise-elongated, low-frequency fluctuations within the boundary layer. The theoretical formulation of the low-frequency disturbances, also called laminar streaks or Klebanoff modes, follows the work of Leib, Wundrow & Goldstein [13] and is based on the incompressible linearised unsteady boundary region equations. The outer boundary conditions of this system synthesize the mutual interaction between the boundary layer and the free-stream vortical fluctuations, which generated the low-frequency streaks. Adjoint theory is applied to the equations of motion, and wall-transpiration is employed to attenuate the Klebanoff modes. Our results extend the findings of the adjointbased work by Cathalifaud & Luchini [5], which utilised the steady boundary region equations without including the effect of free-stream disturbances in the formulation.

Optimal state feedback control of streaks and Gortler vortices induced by free-stream vortical disturbances

ABSTRACT. A framework for active wall-transpiration control of spatially developing boundary-layer flows over flat and concave walls (in the streamwise direction) is developed. The controller is designed to suppress the energy growth of streaks and G\"{o}rtler vortices (with emphasis on the latter) induced by free-stream vortical disturbances within an incompressible boundary layer. The control framework uses the primitive variables, velocity and pressure. The flow model is based on the linearised unsteady boundary-region (LUBR) equations, which are the rigorous asymptotic form of the Navier-Stokes equations in the limit of low frequency and long-streamwise wavelength. At a particular wavenumber, the effect of free-stream turbulence appears as explicit forcing of these equations which is obtained by asymptotic matching with the far field conditions. An optimal control problem that accounts for this explicit forcing is formulated and solved. The objective cost function that is minimised comprises the weighted energy of the streaks/ Gortler vortices and of the actuation. Results show the effectiveness of the developed control approach.

Feedback control for reducing the pressure drag of bluff bodies terminated by a backward-facing step
SPEAKER: unknown

ABSTRACT. Four flows over a backward-facing step are considered. These exhibit a range of flow physics, and are of relevance to the flow over the rear of ``squareback" road vehicles. Computational flow simulations are used as a test-bed to devise a linear feedback control strategy which achieves a mean base pressure recovery (equivalent to reducing pressure drag) for all four flows. The strategy is based on the premise that reducing pressure drag fluctuations improves mean pressure recovery. Thus the feedback control objective is to attenuate base pressure force fluctuations. The response of each of the flows to actuation is characterised via harmonic forcing system identification. Feedback control is found to successfully achieve a mean pressure recovery for all four of the flows, and is particularly effective for 2-D geometries (with either laminar or turbulent separation). The approach uses only body-mounted sensing and actuation and could be applied experimentally.

Passivity-Based Feedback Control of a Channel Flow For Drag Reduction
SPEAKER: Peter Heins

ABSTRACT. A new active-closed-loop linear time-invariant robust control strategy for reducing drag in a channel flow is presented. Using the theory of passive systems, controllers were designed so as to make the closed-loop system as close to passive as is possible. Physically, this means minimising the maximum bound on perturbation energy production within the system. Although the resulting controllers are linear, the nonlinearity, inherent in all fluidic systems, is accounted for and included as an exogenous feedback forcing. Passivity-based controllers were generated for a Re=2230 channel flow for a range of spatial wavenumber pairs and then analysed. They were shown to be capable of greatly restricting transient energy growth.

Recent progress in turbulent skin friction drag reduction: an overview

ABSTRACT. The overview leads to the following conclusions.

Recent research on fundamental understanding of flows with drag reduction has been successful: there are now tractable (that is yielding to intuitive understanding) and computationally-efficient methods for predicting both the structures and the drag reduction.

The focus should now be shifted to two issues: high-Re effects and practical feasibility.

High-Re effects appear to be mostly negative, however, there are results suggesting that in some cases increase in Re can even improve the drag-reduction effect.

There is at least one proposal (oblique wavy wall) of a practically-possible method of reducing skin friction.

Energy Amplification of Poiseuille Flow in a Pipe Lined with Riblets

ABSTRACT. Using as a passive method for drag reduction, riblets are structures that run parallel to one another and are aligned longitudinally to fluid flows. It has been shown experimentally and numerically that the drag coefficients over the surface can be reduced by up to 10% when the shape, spacing and height of the riblets are optimised. While the benefits of riblets have been well known, the mechanism of drag reduction is not fully understood. This paper analyses the effects of riblet structures on energy amplification in streamwise constant Poiseuille flow in a circular pipe. The linearised governing equations of the flow are written into the two-dimensional/three-component form. Through a change of coordinates, the rough domain is transformed into a smooth one. The equations are discretised using spectral methods and the finite-dimensional approximation of the system is represented by a state space model. The transient growth in the flow without external forcing and H2 norm of the system subject to stochastic noise are calculated. Our computation shows that the presence of riblets can reduce the transient growth and H2 norm in pipe flow.

13:00-15:00 Session 3E: Tutorial Session - Zames-Falb multipliers
Tutorial on Zames-Falb Multipliers

ABSTRACT. This session will provide a tutorial on both the underlying theory of Zames-Falb multipliers and their use in modern control analysis and synthesis. The class of multipliers has received steady attention in the literature, not least due to their application to systems with saturation, including anti-windup. Our treatment will be sufficiently introductory to be accessible to PhD students. In particular we only consider continuous single-input single-output systems. Nevertheless we will include up-to-date research questions and the tutorial should be of interest and use to all researchers in this and related fields.

15:30-17:30 Session 4A: Posters 1
Implementations of Predictive Controller for NMAS Consensus with Neighbour’s Agents Delay
SPEAKER: Guoping Liu

ABSTRACT. This paper discussed the implementation of prediction algorithm in solving the external consensus problem for networked multi-agent system (NMAS) with asymmetric network-induced communication delay. By considering that communication delay between agents is different for each neighboring agents, the output predictor based on the recursive equation is developed. The developed strategy is simulated with single-input single-output (SISO) NMAS model and validated through practical experiment with water level test rigs under intranet network connection.

Formal Verification of Control System Properties with Theorem Proving

ABSTRACT. This paper presents the formal verification of high- level properties of control systems with theorem proving (through the Why3 tool). Properties that can be verified with this approach include stability, feedback gain, and robustness, among others. The systems are modelled in Simulink and we propose how to specify the properties of interest over the signals using Simulink blocks. A library of assertion blocks (logic expressions) to annotate the Simulink model, and currently in development, is presented. The functionality and specification of properties in the blocks of the Simulink models are automatically translated from Simulink to Why3 as ‘theories’ and verification goals, respectively, by our tool implemented in MATLAB. A library of theories in Why3 has been developed to facilitate the process of translation of the different Simulink blocks. The goals are automatically verified in Why3 with relevant theorem provers. A first-order discrete system is used to exemplify both, the translation process from Simulink to the Why3 formal logic language, and the verification of Lyapunov stability through added assertion blocks over signals in the model.

Fault Detection and Isolation for an Inertial Navigation System using a bank of Unscented H-infinity filters
SPEAKER: unknown

ABSTRACT. Demand for unmanned aerial vehicles (UAVs) is on the rise and likely to further increase as new uses for this far-reaching technology continue to emerge in both civil and military airspace. In order to ensure safe operation and meet reliability standards at a safety-critical level, instrument failures have to be robustly handled through effective fault diagnosis. A popular approach to fault detection for non-linear systems is the extended Kalman filter (EKF). It has, however, been shown to lack robustness in the face of non-Gaussian noise disturbances and modelling errors. An alternative to the EKF is the extended H (EHF) filter, which is capable of robust estimation even in the presence of coloured (non-Gaussian) noise, though it does inherit certain shortcomings from the EKF, which it is modelled on. A very recent addition to the H family of filters is the unscented H filter (UHF), which holds out the promise of delivering both robustness and excellent estimation performance in non-linear settings. This paper presents arguably the first application of the UHF to an FDI task: sensor fault detection and isolation (FDI) in a strap-down inertial navigation system (INS) of the type commonly mounted on smaller unmanned aircraft. We apply the UHF in a bank of dedicated observers within an analytical redundancy framework. The results are competitive against the EKF.

Multivariable Control of a Lighter than Air System

ABSTRACT. This paper describes the control theory behind a new group of wind energy systems known as Airborne Wind Energy Systems (AWES). The control architecture of a kite system and rigid body system is explored and this is then compared to a lighter than air system developed by Altaeros Energies. The plant model of this system is described in detail and the main dynamics of the system are outlined. This paper concludes with a simulation study looking at pitch and altitude control using a multivariable PID control strategy. This was developed using the Pettinen-Koivo method and tuned on a linearized model of the Altaeros System. Results show good tracking in both altitude and pitch set-points

High Gain Observer with Algorithm Transformation to Extended Jordan Observable Form for Chaos Synchronization Applications

ABSTRACT. This work is concerned with the observer design for a class of chaotic nonlinear systems for chaos communications applications. For this, we consider mainly the Rossler chaotic model as drive oscillator. We start by transforming the Rossler model into the so-called Extended Jordan Controllable Form in order to facilitate the construction of the observer’s gain. It is shown that synchronization is achieved under a typical unidirectional master-slave configuration by using the receiver as the observer. As robustness against delay and different initial conditions are key for telecommunication applications, the designed observer is simulated with Simulink under the presence of delay in the transmission channel and different initial conditions for transmitter and receiver, showing good convergence performance. The results obtained show potential of application to Visible Light Communication, a free space optical communication based on LEDs.

Model Based Hysteresis Compensation for IPMC Sensors
SPEAKER: Yonghong Tan

ABSTRACT. Ionic Polymer-Metal Composite (IPMC) is a kind of smart material which can be used as sensors or actuators. As IPMC is very flexible and can produce larger electric signal when it is deformed, it is more suitable to be used as sensors to measure deformation, displacement and flow rate. However, hysteresis existing in IPMC will deteriorate the performance of the sensor. In this paper, a neural network model based compensator is proposed to reduce the effect of hysteresis. In the compensation scheme, a method of expanded input space is introduced to transform the multi-valued mapping of hysteresis to a one-to-one mapping. The corresponding theory of the construction of the expanded input space is illustrated. Then, based on the expanded input space, the inverse model based compensator is then constructed. Moreover, a geometric compensation method is proposed to compensate for the measuring error of the laser sensor on IPMC chip. Finally, experimental results are presented to validate the proposed method of hysteresis compensation for IPMC sensors.

Jitter sensitivity of a self-tuning input-constrained predictive controller

ABSTRACT. In this paper, the jitter sensitivity of a real-time embedded implementation of a self-tuning generalized predictive control algorithm will be experimentally investigated using a hardware-in-the-loop testing technique. The aim of the study was to explore the potential impacts of slew rate limits of the input of the controlled system on the performance of the controller when jitter is present. The paper also examines the effects of different preemption levels of the underlying earliest deadline first scheduler on the performance of the control system.


ABSTRACT. A MATLAB GUI is presented which is used to help students learn to design controllers in the frequency domain. It complements the author’s two previous GUIs for plotting and identification of systems in the frequency domain. It also incorporates the concept used in the “electronic calculator that makes students think” to assist learning.

A Behavioural Control Strategy of Human-Human Interaction in an Object Transfer Task
SPEAKER: unknown

ABSTRACT. This paper presents an outline of human-human interaction (HHI) to establish a framework to understand how a behaviour based approach can be developed in the design of a human-robot interaction (HRI) strategy. To approach the conceptual design guidelines for a HRI control strategy, the human dynamic model of human behaviour during performing an object transfer task without any types of communication has been strategically analysed. The extended crossover model proposed by McRuer has been applied to identify the human arm characteristic models under various conditions when executing cooperative tasks. A set of compliant-object-handover tasks have been designed and conveyed (based on Box-Behnken test design), along with the influence variables affecting the human forces, consisting of mass, friction and target displacement. According to the results, the McRuer crossover models were appropriately estimated and reported to be in good matching with the actual experimental data. Additionally, it can be found that a loop gain (KH) is inversely proportional to the object distance moved and is associated with a faster response. The best-fit percentages of human force profiles are almost 100%; therefore, the proposed models can be used to present the human arm characteristics effectively.

Point localisation of a subsea cable using particle filters

ABSTRACT. Point localisation of subsea cables are necessary as a starting point in the search of a particular section or whole length of a cable and become a demanding and challenging task in an uncertain environment such as sea. The authors propose a novel method of using particle filters for estimating the position of a subsea cable in a highly uncertain environment. The method was tested on data collected from a buried cable in the Baltic sea, Denmark and shown to have close approximation to the true location of the subsea cable. The method can be used to localise a subsea cable in an off-shore noisy and uncertain environment and provides an inexpensive alternative to the use of a diver or a remotely operated platform.

15:30-17:30 Session 4B: Posters 2
A Review of Driver Modelling
SPEAKER: unknown

ABSTRACT. Increasingly accurate vehicle simulations are required by automotive manufacturers and researchers in order to develop new ideas while minimising the use of costly prototype vehicles. The development in vehicle simulations has been the focus of much research; however the progress of driver modelling development for use in these simulations has not been as rapid. In areas such as fuel economy it has been shown that the behaviour of the driver plays a significant role and as such a vehicle simulation used to investigate fuel economy should include a driver model that can mimic different driver behaviour. Historically fuel economy simulations have been undertaken with a reference speed profile followed accurately with classical closed loop control and this approach will neglect the characteristics of real world driving that come from having a human driver. In this paper the development of driver modelling will be summarised up to the current state of the art and further applications for driver models are detailed.

Deployment of model-based product engineering for embedded automotive control systems

ABSTRACT. Software development and test methods for automotive embedded systems are increasing dramatically since the deployment of the first electronic control unit in 1970s. The increase of automotive embedded systems will continue to grow due to the deployment of complex vehicle features realised by software in order to meet the voice of the customer demands for higher quality, safety, reliability and comfort. This challenge makes imperative the need for new and innovative development and validation engineering methods in order to enable the creation of robust vehicle control systems while managing the pressures of increased system complexity and reduced time to market. This paper presents the roll-out of a new development process for automotive control systems and software followed by an insight of challenges and opportunities faced during deployment. The development process is called model-based product engineering. The process is integrated with a design verification interface for automated test script generation throughout the development cycle. The deployment of model-based product engineering is portrayed through examples of vehicle control systems and software development.

Hybrid (DEBBO) Fuzzy Logic Controller for Quarter Car Model

ABSTRACT. In this paper a hybrid Differential Evolution based Biogeography Based Optimization (DEBBO) which converge quickly rather than algorithms without hybridization has been proposed for the tuning of Fuzzy Logic Controller (FLC) applied to a Quarter Car (QC) model with the RMS value of body acceleration as the performance index. It has been proven that the proposed controller works better than the conventional Proportional Integral Differential (PID) controller, DEBBO based PID (DEBBOPID) and intelligent FLC with the same control structure by performing the simulation in MATLAB/ Simulink environment.

SI Engine Combustion Wall Thermal Management Potential without the Present of Control Limitation

ABSTRACT. Tight future CO2 emission targets have encouraged extensive research in options for improving internal combustion engine efficiency. Amongst those, engine thermal management is a promising area to both reduce losses and improve engine power. Earlier studies have shown that engine thermal management was not just protecting engine from overheating but it also can improve engine performance, fuel consumption and even emissions. However, the effects and limits of thermal management are highly complex, and a better understanding is required to reach the full potential. The aim of this paper is to demonstrate the potential of manipulating combustion wall temperature for improving engine efficiency. A 1D numerical model of a 2.2L natural aspirated engine was developed using GT-Suite software for this purpose. The spark timing and fueling in the engine model was also recalibrated to explore the indirect influence of thermal management influence on engine efficiency. The model assumes that the optimal temperature can be achieved at all times, ignoring some of the control implementation issues for now. The results show that optimized combustion wall temperature produces significant fuel consumption improvements at low to medium engine speed at both low and high load. The comparison with conventional temperature control was made using 7 legislated and academic test cycles. The highest fuel economy improvement of about 4% was recorded in urban test cycles. A smaller improvement of more than 2% was found for motorway driving.

A Fuzzy Logic Approach for Vehicle Collision Energy Distribution

ABSTRACT. This paper presents the concept of a novel active buckling control strategy aimed at alleviating the compatibility problem arising when two vehicles of dissimilar mass and stiffness values encounter a collision. The approach assumes that the properties of the vehicle body structures may be changed via actively controlled materials; in particular the smallest positive eigenvalue which corresponds to the point at which buckling commences. The approach is based on a multi-dimensional look-up table combined with fuzzy logic for interpolating between pre-calculated levels of energy absorption related to the stiffness values. By modelling the force versus deformation characteristic, the energy to be absorbed is more appropriately apportioned, thus enhancing vehicle safety

Non linear optimization of a sport motorcycle’s suspension interconnection system.
SPEAKER: unknown

ABSTRACT. A high fidelity nonlinear model of a sport motorcycle is modified to include interconnected suspension forces between the front and rear ends. The comfort and the suspensions efficiencies have been study for a wide range of interconnection stiffness and damping coefficients. An optimization of these coefficients is performed considering different possible mechanical implementations: full active interconnection, semi-active interconnection and passive interconnection. Finally the system is analyzed from the stability point of view to ensure that the oscillating modes are not strongly modified and the system stability is not compromised.

Improving the Velocity Obstacle Approach for Obstacle Avoidance in Indoor Environments
SPEAKER: Ahmad Alsaab

ABSTRACT. Abstract--- The velocity obstacle approach (VO) is considered an easy and simple method to avoid moving obstacles, where the collision cone principle is used to detect the collision situation between two circular-shaped objects. The VO approach has two challenges when applied in indoor environments. The first challenge is to extract collision cones of non-circular objects from sensor data, where applying fitting circle methods generally produce large and inaccurate collision cones specially for line-shaped obstacle such as walls. The second challenge is that the mobile robot cannot sometimes move to its goal because all its velocities to the goal are located within collision cones. The contribution of this paper is that a method was demonstrated to extract the collision cones of circular and non-circular objects using a laser sensor, where the obstacle size and the collision time are considered to weigh the velocities of the robot. The experiments within an indoor environment showed that the mobile robot successfully avoided dynamic obstacles which have different abilities to avoid other obstacles.

Friction Compensation for a Force Controlled Electric Actuator with Unknown Sinusoidal Disturbance Motion
SPEAKER: unknown

ABSTRACT. This paper presents a method of friction compensation for a linear electric motor subjected to unknown sinusoidal disturbance motions. The method uses a Coulomb friction model and applies a feedforward step signal when velocity zero crossing occurs. Velocity zero crossing estimation is achieved using an algorithm based on measured feedback velocity and force.

Research & Design on a control system for a disk-type flying robot with multiple rotors
SPEAKER: Ikuo Yamamoto

ABSTRACT. The paper describes the development of a disk-type flying robot with multiple rotors for observation from sky.The effective modeling method for control system design is established.High tracking performance of the reference trajectory and good robustness against environmental disturbance are confirmed by the optimal robust control system.

Control and Energy Management for Quadrotor

ABSTRACT. The availability of sufficient energy reserves and the performance of the autopilot constitute serious limiting factors in order to guarantee the completion of missions by battery-powered quadrotor helicopters. In this context, this article proposes methods aimed at improving the mission-level functional reliability through enhanced system self-awareness and adaptive mission planning. The central idea is to use system prognosis to estimate the energy available along the mission. In view of the possibility of occurrence of faults, the controller adopted for the quadrotor is of cascade switching multi-model predictive type, based on a set of piecewise affine models for the helicopter's displacement dynamics. The mission reconfiguration is carried out by casting it in the form of a Mixed Integer Linear Programming (MILP). The proposed methods were evaluated by numerical simulation under a variety of realistic scenarios. The results were satisfactory in terms of mission reconfiguration while performing accurate reference tracking.

15:30-17:30 Session 4C: Posters 3
Wind Turbine Gust Estimation Using Remote Sensing Data
SPEAKER: unknown

ABSTRACT. The offshore wind energy industry is experiencing sustained growth as governments around the world look to secure low-carbon sources of energy. In order to capture more energy from the wind, larger turbines are being designed, leading to the structures becoming increasingly vulnerable to damage caused by violent gusts of wind. Advance knowledge of such gusts will enable turbine control systems to take preventative action, reducing the cost of turbine maintenance. Therefore, in this paper we present a methodology to estimate the velocity profile of an oncoming wind field, given only limited spatio--temporal measurements from typical light detection and ranging (LiDAR) instruments, mounted on the turbine nacelle. The main contribution of this paper lies in the derivation of a simplified deterministic model of atmospheric boundary--layer flows, based on the Navier--Stokes equations, that enables subsequent implementation of an unscented Kalman Filter. Results are presented that compare the accuracy of the estimated wind field to actual wind-data produced from large eddy simulations of the atmospheric boundary layer.

Active Control of Fluid-borne Noise in Hydraulic Systems Using In-series and By-pass Structures
SPEAKER: unknown

ABSTRACT. The nature of digital hydraulic systems may cause severe fluid-borne noise problems because of the pulsed nature of the flow. An effective method to reduce the noise that does not impair the system performance and efficiency is needed. This article reports on initial investigations of an active valve for pressure pulsation attenuation in switched inertance hydraulic systems (SIHS) based on in-series and by-pass structures. The in-series structure represents a valve arranged in line between the SIHS and the load providing a controlled pulsating pressure drop, whilst for the by-pass structure the valve was arranged in parallel with the load providing a controlled pulsating bleed-off flow. A high-performance piezoelectric valve was used as the active controller. Adaptive notch filters with the filtered-X least mean square algorithm were applied for pressure pulsation attenuation, while a frequency-domain least mean square filter was used for secondary path identification. Simulated and experimental results show that excellent cancellation was achieved using the proposed methods, which have several advantages over passive noise control systems. Comparison of the in-series and by-pass structures is discussed in terms of system performance, robustness and advantages. The proposed control structures are very promising for fluid-borne noise cancellation in fluid power systems or other fluid systems with severe noise or vibration problems.

Comparison of RBF and Local Linear Model Networks for Nonlinear Identification of a pH Process

ABSTRACT. Abstract— This paper focuses on the nonlinear identification of an experimental pH neutralisation process using real data. The performances of radial basis function (RBF) and local linear model networks (LLMN) for identifying this significantly nonlinear process are compared. Results are presented to illustrate the choice of the various network parameters in the model structures for network training and validation data. The overall results demonstrate the practical ability of the two network structures for nonlinear system identification.

A LabVIEW-based PI controller for controlling CE 105 coupled Tank System
SPEAKER: Hawre Hussein

ABSTRACT. In this paper, use of Proportional-Integral (PI) controller to monitor and control liquid level in an interconnected CE 105 model coupled tank is investigated. To achieve a system which can instantaneously and accurately control the liquid level in a coupled tank, two different PI controllers have been tested. The LabVIEW library for the PI controller is used to measure liquid levels in the coupled tank. The PI SubVI already exists in the LabVIEW library that gives reasonable performance but to get a better system performance and monitor the liquid levels more accurately another SubVI is derived from the PI controller mathematical equations. The practical results and the system performance of the second SubVI show a faster response and more accurate instantaneous data which minimises the error in the measurements to ±1 mm. Furthermore, the robustness of the controller to change in the system’s parameters is also investigated and established

Enumerative Nonlinear Model Predictive Control for Linear Induction Motor using Load Observer
SPEAKER: Jean Thomas

ABSTRACT. Enumerative nonlinear model predictive control for speed tracking problem of linear induction motors has been presented in [1], where the authors show that this control scheme has better performance as compared to direct torque control. In this paper, the authors show that using a load observer for integral action, the performance can be further improved. Specifically simulation results show that a load observer results in better tracking properties and offers more robust control.

Diagnosis of bearing fault of brushless DC motor under stationary operating condition
SPEAKER: Wathiq Abed

ABSTRACT. This paper presents a new approach for predicting the rolling element bearing defects in brushless DC motor. The stator current and the vibration are selected as fault indicators. The Discrete wavelet transform (DWT) is used to extract the best features from the collected data followed by feature reduction by the application of orthogonal fuzzy neighbourhood discriminative analysis approach. A dynamic neural network classifies the different faults under stationary operating condition. The results obtained from real time simulation demonstrate the effectiveness and reliability of the proposed methodology in accurately classifying the different faults.

Modelling and Building of Experimental Rig for High Redundancy Actuator

ABSTRACT. The high redundancy actuator (HRA) concept is a novel approach to fault tolerant actuation. It refers to an actuator that consists of relatively large number of actuation elements, connected both in series and parallel to form a single actuator. This configuration improves the reliability and availability of the actuator and thus provides a high degree of fault tolerance. The HRA also reduces the need for over-sizing an actuator especially in safety-critical applications such as aerospace. HRA is suitable to a wide range of actuation technology but this paper focuses on a linear electromechanical actuator (EMA) due to its extensive application not only in industrial machinery but in aircraft and the aerospace industry in general. This paper presents ongoing and future work to demonstrate the concept of a fault tolerant system high redundancy actuator through a 3-by-4 series-in-parallel linear electromechanical actuator.

A New Approach to Design Optimal Excitation Trajectories for Parameter Estimation of Robot Dynamics
SPEAKER: unknown

ABSTRACT. This paper presents a new method for designing optimized excitation trajectories for the estimation of robot manipulator model parameters. Current optimization methods require large number of parameters and solving nonlinear optimization problems. This work utilizes the Schroeder Phased Harmonic Sequence (SPHS) in the trajectory design process which has three advantages over previous methods: (i) a guaranteed optimal solution, (ii) a controllable frequency spectrum and a low peak factor signal, (iii) a reduced number of parameters to optimize. The method here proposed is experimentally tested on a small three Degree of Freedom robot. The results show that the model estimated using the new method is more accurate at predicting the dynamics of the robot when compared to a non-optimized excitation.

ANFIS Based Jacobian for a Parallel Manipulator Mobility Assistive Device
SPEAKER: Ahmed Asker

ABSTRACT. Nowadays, parallel manipulators are used widely in bioengineering applications; this leads to many exciting expectations as well as challenges. The kinematic analysis of parallel manipulators with their differential kinematics yielding the Jacobian in a closed form is not a trivial task. In this paper a parallel manipulator based mobility assistive device called EJADII is analyzed to determine forward kinematics, inverse kinematics and closed-form Jacobian. An Adaptive Neuro-Fuzzy Inference System (ANFIS) is trained to estimate the Jacobian. This system would be useful when determination of the Jacobian in a closed-form is difficult to determine. The human motion during sit to stand captured by VICON experiment is used with two assisting scenarios to train and verify this system. Computer simulations show relatively good results of the proposed system.

Disclaimer | Powered by EasyChair Smart Program