Showing papers presented at "Mediterranean Conference on Control and Automation in 2012"

A survey on Underwater Wireless Sensor Networks and applications

Abstract: In this article a survey on the different technologies in the area of Underwater Wireless Sensor Networks (UWSN) will be presented. The characteristics of these networks are different from those found in the terrestrial ones, while their architecture is vulnerable to various issues such as large propagation delays, mobility of floating sensor nodes, limited link capacity and multiple messages receptions due to reflections on the sea ground and sea surface. This article will present an overview of the underlying technologies in UWSN and will focus in presenting the most important research approaches towards UWSNs' architecture, routing, MAC and localization protocols, energy consumption and security, while highlighting their most illustrative real-life applications.

A model predictive control approach to the load shifting problem in a household equipped with an energy storage unit

Abstract: This paper deals with the load shifting problem in a household equipped with smart appliances and an energy storage unit with conversion losses. The problem is faced by establishing an event driven Model Predictive Control framework aiming to meet the real life dynamics of a household and to keep low the impact of the control system on the total electric energy consumption. The proposed approach allows the consumer to minimize the daily energy cost in scenarios characterized by Time of Use tariffs and Demand Side Management, by dynamically evaluating the best time to run of the appliances and the optimal evolution of the battery level of charge. A proper set of realistic simulations validates the proposed approach, showing the relevance of the energy storage unit in the domestic load shifting architecture.

Analysis of drilling vibrations: A time-delay system approach

Abstract: The main purpose of this study is the description of the qualitative dynamical response of a rotary drilling system with a drag bit, using a model that takes into consideration the axial and the torsional vibration modes of the bit. The studied model, based on the interface bit-rock, contains a couple of wave equations with boundary conditions consisting of the angular speed and the axial speed at the top additionally to the angular and axial acceleration at the bit whose contain a realistic frictional torque. Our analysis is based on the center manifold Theorem and Normal forms theory whose allow us to simplify the model.

Multipurpose autonomous underwater intervention: A systems integration perspective

Abstract: Nowadays, autonomous intervention is getting more attention in the underwater robotics community. Few research projects on this matter are currently under development. In this context, and after a first successful experience in the RAUVI Spanish project (2009–2011), the authors are currently involved in the TRIDENT project (2010–2013), funded by the European Commission. To succeed in autonomous intervention, an AUV endowed with a manipulator and with a high degree of autonomy is essential. The complexity of the required robotic system is very high and the system integration process becomes critical. This paper presents the problems being solved in TRIDENT, from a systems integration perspective. As a case study, some results, achieved during the last experiments carried out in the Roses harbor (Girona) in October 2011 will be presented, to demonstrate the capabilities exhibited by the AUV for Intervention under development. The experiments were focused on the problem of autonomously searching and recovering a black-box mock-up that was previously thrown to an unknown position. This paper presents the hardware and software integration aspects that were necessary in order to address such a challenging problem.

Sensor placement for fault diagnosis performance maximization in Distribution Networks

Abstract: The success of any diagnosis strategy critically depends on the sensors measuring process variables. This paper presents a strategy based on diagnosability maximization for optimally locating sensors in distribution networks. The goal is to characterize and determine the set of sensors that guarantees a maximum degree of diagnosability taking into account a given sensor configuration cardinality constraint. The strategy is based on the structural model of the system under consideration. Structural analysis is a powerful tool for determining diagnosis possibilities and evaluating whether the number and the location of sensors are adequate in order to meet some diagnosis specifications. The proposed approach is successfully applied to leakage detection in a Drinking Water Distribution Network.

Advances in observer design for Takagi-Sugeno systems with unmeasurable premise variables

Abstract: This paper proposes a new approach of observer design for nonlinear systems described by a Takagi-Sugeno model. Its main contribution concerns models with premise variables depending on the system states which are completely or partially unknown. This case is more difficult than when the premise variables are known or measured. Indeed, in this case, weighting functions of the observer depend on state estimates and the state estimation error is then governed by a Lipschitz nonlinear system. Here, two main results are established. Firstly, relaxed stability conditions are provided, using a nonquadratic Lyapunov function, to guarantee asymptotic stability of the observer. This aims to reduce the conservativeness compared to the existing works and enhance the maximal admissible Lipschitz constant for which the linear matrix inequality (LMI) conditions are feasible. Secondly, the Input-to-State Stability concept combined to a nonquadratic Lyapunov function are used to guarantee a bounded state estimation error which relaxes the conservativeness related to the Lipschitz constant. The robustness aspect is dealt with respect to some bounded modeling uncertainties and additive bounded perturbations. The stability conditions are expressed in terms of LMI.

Trust-based multi-agent filtering for increased Smart Grid security

Abstract: We address the problem of state estimation of the power system for the Smart Grid. We assume that the monitoring of the electrical grid is done by a network of agents with both computing and communication capabilities. We propose a security mechanism aimed at protecting the state estimation process against false data injections originating from faulty equipment or cyber-attacks. Our approach is based on a multi-agent filtering scheme, where in addition to taking measurements, the agents are also computing local estimates based on their own measurements and on the estimates of the neighboring agents. We combine the multi-agent filtering scheme with a trust-based mechanism under which each agent associates a trust metric to each of its neighbors. These trust metrics are taken into account in the filtering scheme so that information transmitted from agents with low trust is disregarded. In addition, a mechanism for the trust metric update is also introduced, which ensures that agents that diverge considerably from their expected behavior have their trust values lowered.

On the Lyapunov and Bohl exponent of time-varying discrete linear system

Abstract: This note studies the problem of bounds for the Lyapunov exponent of a parameter perturbed system when the perturbation has finite average value. Such a bound is presented in terms of Bohl exponent of the unperturbed system. In particular, it has been shown that the Lyapunov exponent of perturbed system is not greater than the Bohl exponent of the unperturbed system if the average value of perturbations is zero. The obtained result is illustrated by a numerical example.

Sliding mode control for Turbocharged Diesel Engine

Abstract: For modern Diesel engines, accurate fuel-air ratio AFR and Exhaust Gas Recirculation (EGR) rates control is important for manufacturers to face a more restrictive legislation levels. To fulfill the requirements, hardware devices such (EGR) and Variable Geometry Turbochargers (VGT) valves have been introduced, and sophisticated control algorithms were designed. The main objective of the air path controller is to regulate in the intake manifold the AFR ratio and the EGR fraction rates to their desired values. Earlier EGR PID controllers needed a fastidious and time consuming calibration step for each engine operating point. Nonlinear control algorithms has quickly appeared as a promising way to provide an efficient air path controllers, since they don't need a calibration step. The main drawback of such controllers is coming from the fact that they are based upon a diesel engine model which handles parameters uncertainties and signal measurements errors, that affects the control performance. In this paper we propose a novel control scheme for controlling the diesel engine air path. Control design is carried out under the sliding mode framework. The proposed controller has been tested on the Jankovic Tubocharged Diesel Engine (TDE) model. To demonstrate the robustness of the proposed controller, simulation results showing the tracking of the compressor flow W c and the exhaust manifold pressure p 2 variables are presented.

Distributed strategies for balancing a weighted digraph

Abstract: A weighted digraph is balanced if, for each node, the sum of the weights of the edges outgoing from that node is equal to the sum of the weights of the edges incoming to that node. Weight-balanced digraphs play a key role in a number of applications, including cooperative control, distributed optimization, and distributed averaging problems. We address the weight-balance problem for a distributed system whose components (nodes) can exchange information via interconnection links (edges) that form an arbitrary, possibly directed, communication topology (digraph). We develop two iterative algorithms, a centralized one and a distributed one, both of which can be used to reach weight-balance, as long as the underlying communication topology forms a strongly connected digraph (or is a collection of strongly connected digraphs). The centralized algorithm is shown to reach weight-balance after a finite number of iterations (bounded by the number of nodes in the graph). The distributed algorithm operates by having each node adapt the weights on its outgoing edges and is shown to asymptotically lead to weight-balance. We also analyze the rate of convergence of the proposed distributed algorithm and obtain a (graph-dependent) worst-case bound for it. Finally, we provide examples to illustrate the operation, performance, and potential advantages of the proposed algorithms.

Towards a high-end unmanned tri-TiltRotor: design, modeling and hover control

Abstract: The design process, development considerations as well as hover modeling and control of a high-end capability unmanned Tri-TiltRotor aerial vehicle are the subjects of this article. Tilt-Rotor UAVs efficiently perform as fixed-wing air-craft and as rotorcrafts, being capable of both high-endurance high-speed flight as well as of vertical take-off and landing and precision manoeuvring. The additional integration of high-end computational power and sensors on such a type of aircraft provide a UAV platform with multiple operational usefulness capabilities. The system's hovering nonlinear dynamics and linearized model based on an identification of its actuation sub-system are presented, along with hovering attitude stabilization experimental results.

Wind turbine power maximisation based on adaptive sensor fault tolerant sliding mode control

Abstract: This paper presents a new strategy to robust fault tolerant control (FTC) to optimise the wind energy captured by a wind turbine operating at low wind speeds, using an adaptive gain Sliding Mode Control (SMC). In addition to the inherent robustness of SMC against matched model uncertainty, the proposed method involves a robust descriptor observer design that can provide robust simultaneous estimation of states and the “unknown outputs” (sensor faults and noise) in order to guarantee the robustness of the sliding surface against unknown output effects. Moreover, the sliding surface is designed to achieve the required objectives by utilizing the nonlinear flexible two mass model of the variable speed wind turbine. The proposed FTC SMC method is applied to a 5 MW wind turbine benchmark model.

On line solar irradiation forecasting by minimal resource allocating networks

Abstract: The paper describes an on-line prediction algorithm to estimate, over a determined time horizon, the solar irradiation of a specific site The learning algorithm is based on a radial basis function network and combines the growing criterion and the pruning strategy of the minimal resource allocating network technique An Extended Kalman Filter (EKF) is used to update all the parameters of the network The on-line algorithm is able to avoid the initial training of the neural network A comparison of the performance obtained by the MRAN EKF RBF Neural Network with respect to the standard RBF Neural Network is presented

Resilient planning of PowerLine Communications networks over Medium Voltage distribution grids

Abstract: In this paper a network planning problem aiming to enable underground Medium Voltage (MV) power grids to resilient PowerLine Communications (PLCs) is faced. The PLC network is used to connect PLC End Nodes (ENs) located into the secondary substations to the energy management system of the utility by means of PLC network nodes enabled as Access Points. An optimization problem is formulated, aiming to optimally allocate the Access Points to the substations and the repeaters to the MV feeders. A multi-objective optimization approach is used, in order to keep in balance the needs of minimizing the cost of equipment allocation and maximizing the reliability of PLC network paths. Resiliency and capacity constraints are properly modeled, in order to guarantee the communications even under faulted link conditions. As a byproduct, the optimization algorithm also returns the optimal routing. Simulations performed on a realistic underground MV distribution grid validate the proposed approach

Self-triggered linear quadratic Networked Control

Abstract: This paper deals with Networked Control Systems (NCS) design, under the constraint of limited bandwidth on the communication channel. A linear quadratic regulator for a fixed sampling period is solved and this result is used for the development of ℌ 2 and ℌ ∞ performance indexes, yielding to the statement and solution of ℌ 2 and ℌ ∞ optimal control problems. Finally, a self-triggered controller is designed with a switched system approach in order to improve performance. Examples are presented to illustrate the validity of the theory.

Switching model predictive control for an articulated vehicle under varying slip angle

Abstract: In this article a switching model predictive control scheme for an articulated vehicle under varying slip angles is being presented. For the non-holonomic articulated vehicle, the non-linear kinematic model that is able to take under consideration the effect of the slip angles is extracted. This model is transformed into an error dynamics model, which in the sequence is linearized around multiple nominal slip angle cases. The existence of the slip angles has a significant effect on the vehicle's path tracking capability and can significantly deteriorate the performance of the overall control scheme. Based on the derived multiple error dynamic models, the varying slip angle is being considered as the switching rule and a corresponding switching mode predictive control scheme is being designed that it is also able to take under consideration: a) the constrains on the control signals and b) the state constraints. Multiple simulation results are being presented that prove the efficacy of the overall suggested scheme.

Position funnel control for rigid revolute joint robotic manipulators with known inertia matrix

Abstract: This paper presents funnel control for position control of rigid, revolute joint, n-degree-of-freedom (DOF) robotic manipulators with known inertia matrix. The multi-input multi-output (MIMO) funnel controller assures reference tracking with ‘prescribed transient accuracy’, i.e., for each joint, the absolute value of position and speed tracking error (difference between reference and actual value) is bounded by a prescribed, positive (possibly non-increasing) function of time (the ‘funnel boundary’), respectively.

Sigma-point Unscented Kalman Filter used for AUV navigation

Abstract: This paper presents an implementation of the Sigma-point Unscented Kalman Filter (SP-UKF) used in the simulated task of open-water navigation of two types of AUV. The first simulated vehicle is a large cruise-type vehicle modeled after the Instituto Superior Tecnico, Lisbon vehicle Infante. The second is a small, almost fully actuated vehicle with tunnel thrusters modeled after the SSC Pacific, San Diego vehicle CETUS II. The SP-UKF shows itself, after properly taking care of implementation details, to be a robust methodology which allows for efficient and correct navigation, aided by several typical sensors (DVL, USBL hydroacoustic localization systems, AHRS). The influence of currents on the navigation, and the ability of estimating the current components is also researched. The navigation fix is fed back to the low-level control loops aboard each vehicles to achieve sane and rational navigation of the waterspace that follows stably and robustly the command signals.

Quasi-LPV modelling and non-linear identification of a Twin Rotor System

Abstract: This paper describes the experimental identification of the parameters of the Twin Rotor MIMO System (TRMS) non-linear model using data collected from the real lab set-up. From this non-linear model, a quasi-linear parameter varying (quasi-LPV) model has also been derived using a state transformation. This quasi-LPV model is approximated with a polytopic model using the bounding box approach. Such a model can later be used for control design. The model parameters have been calibrated by means of non-linear least-squares identification approach. Once the calibrated non-linear model has been obtained, a simulator has been built and validated against real data showing satisfactory results when compared to real data.

Self-triggering in nonlinear systems: A small gain theorem approach

Abstract: This paper investigates stability of nonlinear control systems under intermittent information. Building on the small-gain theorem, we develop self-triggered control yielding stable closed-loop systems. We take the violation of the small-gain condition to be the triggering event, and develop a sampling policy that precludes this event by executing the control law with up-to-date information. Based on the properties of the external inputs to the plant, the developed sampling policy yields regular stability, asymptotic stability and L p -stability. Control loops are modeled as interconnections of hybrid systems, and novel results on L p -stability of hybrid systems are presented. Prediction of the triggering event is achieved by employing L p -gains over a finite horizon. In addition, L p -gains over a finite horizon produce larger intersampling intervals when compared with standard L p -gains. Furthermore, a novel approach for calculation of L p -gains over a finite horizon is devised. Finally, our approach is successfully applied to a trajectory tracking control system.

Optimal electric vehicles to grid power control for active demand services in distribution grids

Abstract: In this paper we outline a novel approach for the design of an electric vehicle (EV) aggregator, a controller whose objective is to optimally manage the charging operations of an EV fleet. The control strategy we derive is based on model predictive control and allows to achieve costs minimization, also enabling the aggregator (hence, the EV fleet) to participate to the provisioning of active demand services to upper level market players. Explicative simulations are presented and discussed in order to show the effectiveness of the approach and also to investigate the role of vehicle to grid power.

Leader-follower string formation using cascade control for mobile robots

Abstract: This paper proposes a simple yet effective control algorithm for a platoon of “car-like” robots. The formation used is a line, leader-follower formation, i. e. each robot is a follower for the previous robot and a leader for the next robot. The formation is implemented using speed measurements from optical encoders and distance measurements from image processing, but there is no communication between robots. String stability analysis on the resulting formation is performed. The results of the proposed method are verified by both simulations in Matlab® as well as measured data from the actual mobile robots.

A proactive link-failure resilient routing protocol for MANETs based on reinforcement learning

Abstract: Mobile-Ad-Hoc-Networks (MANET) are self-configuring networks of mobile nodes, which communicate through wireless links. One of the main issues in MANETs is the mobility of the network nodes: routing protocols should explicitly consider network changes into the algorithm design. MANETs are particularly suited to guarantee connectivity in disaster relief scenarios, which are often impaired by the absence of network infrastructures. This work proposes a proactive routing protocol, developed via Reinforcement Learning (RL) techniques, to dynamically choose the most stable path, basing on GPS information, among the feasible ones and to consequently increase resiliency to link failures. Simulations show the effectiveness of the proposed protocol, through comparison with the Optimized Link State Routing (OLSR) protocol.

Fault detection in a wastewater treatment plant based on neural networks and PCA

Abstract: In this paper, a neural network PCA method that integrates neural networks (NN) and principal component analysis (PCA) is used to detect faults in a wastewater treatment plant. The neural networks are used to calculate a non-linear and dynamic model of the process in normal operating conditions. PCA is used to generate monitoring charts based on the residuals calculated as the difference between the process measurements and the output of the networks. It can evaluate the current performance of the process and detects the faults. This technique has been applied to the simulation of a benchmark of a biological wastewater treatment process, a highly non-linear process. The simulation results clearly show the advantages of using this NNPCA monitoring in comparison with classical PCA monitoring.

Implementation and experimental validation of classic MPC on Programmable Logic Controllers

Abstract: Over the last years, a number of publications were written about Model Predictive Control (MPC) on industrial Programmable Logic Controllers (PLC). They focussed on explicit MPC strategies to provide a fast solution. When sufficient time is available to solve a classic MPC problem, an online solution to the corresponding Quadratic Problem (QP) can be provided. This paper investigates the use of an online quadratic programming solver to exploit MPC on a PLC. This will be illustrated with the classic Hildreth QP algorithm and qpOASES, a recently developed online active set strategy. These algorithms will be investigated on a MISO system.

A new Saturation function to convert an output constraint into an input constraint

Abstract: The main result of our new method is to convert an output constraint into an input constraint for a nonlinear system. Then, standard saturated input design can be used as an anti-windup design for instance which is known to be efficient to enlarge a closed-loop system stability domain in presence of control saturations. We only address the single input single ‘saturated’ output problem in this paper.

Repetitive sliding mode control for nondecouplable multivariable systems: Periodic disturbances rejection

Abstract: Disturbances rejection is an important field of control theory. In this context, our paper is proposed to deal with asymptotic rejection of periodic disturbances affecting discrete multivariable systems with an interactor matrix. A multivariable repetitive sliding mode control is proposed to cancel the disturbances when the system is nondecouplable. To synthesis this control an interactor matrix is used. A numerical example shows that the proposed strategy gives good performance in terms of rejecting periodic disturbances for nondecouplable multivariable systems.

Nonlinear model predictive substrate feed control of biogas plants

Abstract: Optimal substrate feed control of biogas plants is a complex and challenging task due to the nonlinearity of the anaerobic digestion process, which produces biogas from biodegradable input material. In this paper a nonlinear model predictive control (NMPC) scheme is applied to optimally control the substrate feed of an agricultural biogas plant. The implemented algorithms are investigated in a simulation study using a validated simulation model of a full-scale biogas plant. Process states are estimated using a recently developed state estimator. Results show that this approach is very feasible providing the plant operator with a gain of 550 € per day compared to previous operation.

Distributed macro calibration in sensor networks

Abstract: In this paper a novel consensus-based distributed algorithm for blind macro-calibration in sensor networks is proposed. It is proved, on the basis of an originally developed methodology for treating higher order consensus schemes, that the algorithm achieves asymptotic agreement for sensor gains and offsets in the mean square sense and with probability one. In the case of a given reference, all sensors are asymptotically calibrated. Simulation results illustrate properties of the algorithm.

An artificial intelligence approach to forward kinematics of Stewart Platforms

Abstract: The Stewart Platform, one of the most successful and popular parallel robots, has attracted the attention of many researchers in recent decades. The solution of the forward kinematics problem in real-time is one of the key aspects that continues to garner interest. In this paper we propose a new approach for solving this particular case using Support Vector Machines, a popular Machine Learning method for classification and regression. The algorithm involves a data generation and preprocessing off-line phase, and a fast on-line evaluation. The experiments show that this method is very accurate and suitable for use in real-time.