Publications

International refereed journals

Building energy consumption used for internal heating and cooling purposes is one of the most viral research topics. Retrofitting and renovation activities in building applications aim towards utilizing modern construction materials, with improved thermal and insulation characteristics. It is more than evident that such an approach leads to an improved thermal shield for the building (improving passive building elements). In addition well calibrated rule based control designs are also being adopted in the last decades as a way to improve the energy efficiency in buildings (improving active elements management). Both of the above approaches though are considered as time static since disturbances with high uncertainty (weather conditions, human presence and activity) along with the unavoidable construction material aging phenomena affect building behavior and HVAC dynamics. As a result control recalibration activities seem more than necessary to maintain energy efficiency. Followed by the rapid evolution in the computing machines sector and simulation software kits, research effort has been focused on model-assisted and co-simulation based control strategies which utilize the available computational power of modern machines towards improving energy efficiency and comfort levels through appropriate design Building Optimization and Control (BOC) systems, utilizing available system models. However the main drawback in model-assisted strategies is the fact that they heavily rely on the available building model which requires a tedious offline pre-application period including many simulation tests and/or field experiments so as to fine tune and tailor manually the model and consequently the control logic implemented. Moreover, no matter how elaborate the building model is, aging characteristics and uncertain disturbances are factors which call for re-designing (periodically) the available simulation model and the respective control strategies.
This paper considers an alternative approach to BOC system design. The main attribute of the proposed methodology is that it can provide automated fine-tuning of the BOC system: no human intervention or a simulation model are required for the initial deployment of the controller as well as for the continuously applied fine-tuning procedure. Real-life experiments performed in a highly energy demanding building in Tel Aviv Israel, during spring time, demonstrate that the proposed approach can effectively provide intelligent decisions that none of the currently employed rule/event-based strategy can replicate.

Iakovos T. Michailidis, Christos D. Korkas, Elias B. Kosmatopoulos, Evyatar Nassie, “Automated Control Calibration Exploiting Exogenous Environment Energy: An Israeli Commercial Building Case Study”,Energy and Buildings, Elsevier, June 2016

Integration of renewable energy sources in microgrids can be achieved via demand response programs, which change the electric usage in response to changes in the availability and price of electricity over time. This paper presents a novel control algorithm for joint demand response management and thermal comfort optimization in microgrids equipped with renewable energy sources and energy storage units. The proposed work aims at covering two main gaps in current state-of-the-art demand response programs. The first gap is integrating the objective of matching energy generation and consumption with the occupant behavior and with the objective of guaranteeing thermal comfort of the occupants. The second gap is developing a scalable and robust demand response program. Large-scale nature of the optimization problem and robustness are achieved via a two-level supervisory closed-loop feedback strategy: at the lower level, each building of the microgrid employs a local closed-loop feedback controller that processes only local measurements; at the upper level, a centralized unit supervises and updates the local controllers with the aim of minimizing the aggregate energy cost and thermal discomfort of the microgrid. The effectiveness of the proposed method is validated in a microgrid composed of three buildings, a photovoltaic array, a wind turbine, and an energy storage unit. Comparisons with alternative demand response strategies reveal that the proposed strategy efficiently integrates the renewable sources; energy costs are reduced and at the same time thermal comfort of the occupants is guaranteed. Furthermore, robustness is proved via consistent improvements achieved under heterogeneous conditions (different occupancy schedules and different weather conditions).

Christos D. Korkas, Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos, “Occupancy-based demand response and thermal comfort optimization in microgrids with renewable energy sources and energy storage”,Applied Energy, Elsevier, 1st February 2016

This paper deals with the problem of autonomous exploration of unknown areas using teams of Autonomous X Vehicles (AXVs)—with X standing for Aerial, Underwater or Sea-surface—where the AXVs have to autonomously navigate themselves so as to construct an accurate map of the unknown area. Such a problem can be transformed into a dynamic optimization problem which, however, is NP-complete and thus infeasible to be solved. A usual attempt is to relax this problem by employing greedy (optimal one-step-ahead) solutions which may end-up quite problematic. In this paper, we first show that optimal one-step-ahead exploration schemes that are based on a transformed optimization criterion can lead to highly efficient solutions to the multi-AXV exploration. Such a transformed optimization criterion is constructed using both theoretical analysis and experimental investigations and attempts to minimize the “disturbing” effect of deadlocks and nonlinearities to the overall exploration scheme. As, however, optimal one-step-ahead solutions to the transformed optimization criterion cannot be practically obtained using conventional optimization schemes, the second step in our approach is to combine the use of the transformed optimization criterion with the cognitive adaptive optimization (CAO): CAO is a practicably feasible computational methodology which adaptively provides an accurate approximation of the optimal one-step-ahead solutions. The combination of the transformed optimization criterion with CAO results in a multi-AXV exploration scheme which is both practically implementable and provides with quite efficient solutions as it is shown both by theoretical analysis and, most importantly, by extensive simulation experiments and real-life underwater sea-floor mapping experiments in the Leixes port, Portugal.

A. Ch. Kapoutsis, S. A. Chatzichristofis , L. Doitsidis, J. Borges de Sousa, J. Pinto , J. Braga and E. B. Kosmatopoulos , “Real-time Adaptive Multi-Robot Exploration with Application to Underwater Map Construction”, Submitted to Autonomous Robots,30 October 2015.

Energy efficient passive designs and constructions have been extensively studied in the last decades as a way to improve the ability of a building to store thermal energy, increase its thermal mass, increase passive insulation and reduce heat losses. However, many studies show that passive thermal designs alone are not enough to fully exploit the potential for energy efficiency in buildings: in fact, harmonizing the active elements for indoor thermal comfort with the passive design of the building can lead to further improvements in both energy efficiency and comfort. These improvements can be achieved via the design of appropriate Building Optimization and Control (BOC) systems, a task which is more complex in high-inertia buildings than in conventional ones. This is because high thermal mass implies a high memory, so that wrong control decisions will have negative repercussions over long time horizons. The design of proactive control strategies with the capability of acting in advance of a future situation, rather than just reacting to current conditions, is of crucial importance for a full exploitation of the capabilities of a high-inertia building. This paper applies a simulation-assisted control methodology to a high-inertia building in Kassel, Germany. A simulation model of the building is used to proactively optimize, using both current and future information about the external weather condition and the building state, a combined criterion composed of the energy consumption and the thermal comfort index. Both extensive simulation as well as real-life experiments performed during the unstable German wintertime, demonstrate that the proposed approach can effectively deal with the complex dynamics arising from the high-inertia structure, providing proactive and intelligent decisions that no currently employed rule-based strategy can replicate.

Iakovos T. Michailidis, Simone Baldi, Elias B. Kosmatopoulos, Martin F. Pichler, Juan R. Santiago, “Proactive Control for Solar Energy Exploitation: a German High-inertia Building Case Study”,Applied Energy, Elsevier, October 2015

Electrical smart microgrids equipped with small-scale renewable-energy generation systems are emerging progressively as an alternative or an enhancement to the central electrical grid: due to the intermittent nature of the renewable energy sources, appropriate algorithms are required to integrate these two typologies of grids and, in particular, to perform efficiently dynamic energy demand and distributed generation management, while guaranteeing satisfactory thermal comfort for the occupants. This paper presents a novel control algorithm for joint energy demand and thermal comfort optimization in photovoltaic-equipped interconnected microgrids. Energy demand shaping is achieved via an intelligent control mechanism for heating, ventilating, and air conditioning units. The intelligent control mechanism takes into account the available solar energy, the building dynamics and the thermal comfort of the buildings’ occupants. The control design is accomplished in a simulation-based fashion using an energy simulation model, developed in EnergyPlus, of an interconnected microgrid. Rather than focusing only on how each building behaves individually, the optimization algorithm employs a central controller that allows interaction among the buildings of the microgrid. The control objective is to optimize the aggregate microgrid performance. Simulation results demonstrate that the optimization algorithm efficiently integrates the microgrid with the photovoltaic system that provides free electric energy: in particular, for each building composing the microgrid, the energy absorbed from the main grid is minimized, the energy demand is balanced with the solar energy delivered to each building, while taking into account the thermal comfort of the occupants.

Simone Baldi, Athanasios Karagevrekis, Iakovos T. Michailidis, Elias B. Kosmatopoulos, “Joint energy demand and thermal comfort optimization in photovoltaic-equipped interconnected microgrids”,Energy Conversion and Management, Elsevier, September 2015

Energy efficient operation of microgrids, a localized grouping of controllable loads with distributed energy resources like solar photovoltaic panels, requires the development of energy management systems (EMSs) with the capability of controlling the loads so as to optimize the aggregate performance of the microgrid. In microgrids comprising of buildings of different nature (residential, commercial, industrial, etc.), where the occupants exhibit heterogeneous occupancy schedules, the objective of an effective management strategy is to optimize the aggregate performance by intelligently exploiting the occupancy schedules and the intermittent production of solar energy. This paper presents a simulation-based optimization approach for the design of an EMS in grid-connected photovoltaic-equipped microgrids with heterogeneous occupancy schedule. The microgrid exchanges energy, buying or selling it, with the main grid and the EMS optimizes an aggregate multi-objective criterion that takes into account both the energy cost and the thermal comfort of the occupants of the microgrid. Simulative results obtained using a microgrid test case developed in EnergyPlus demonstrate the effectiveness of the proposed approach: the proposed EMS strategy is shown to take advantage of the occupancy information, intelligently and automatically changing the energy demand of each building according to the occupants’ behavior, and achieving relevant improvements with respect to alternative EMS strategies.

Christos D. Korkas, Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos, “Intelligent energy and thermal comfort management in microgrids with heterogeneous occupancy schedule”,Applied Energy, Elsevier, 1st July 2015

This article describes a computationally efficient simulation-based control design approach that has the capability of handling optimization problems arising from large-scale nonlinear systems, with fast convergence properties and low computational requirements. The purpose of this article is to describe the main features of the PCAO algorithm, analyze its convergence and stability properties, and demonstrate its efficiency using simulations of two large-scale, real-life systems (a traffic network and an energy-efficient building) for which conventional optimization techniques fail to provide an efficient simulation-based control design.

Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos, Petros A. Ioannou,”A “plug and play” computationally efficient approach for control design of large-scale nonlinear systems using cosimulation: a combination of two “ingredients””,IEEE Control Systems Magazine, October 2014

Accurate maps are essential in the case of robot teams, so that they can operate autonomously and accomplish their tasks efficiently. In this work we present an approach which allows the generation of detailed maps, suitable for robot navigation, from a mesh of sparse points using Cellular Automata and simple evolutions rules. The entire map area can be considered as a 2D Cellular Automaton (CA) where the value at each CA cell represents the height of the ground in the corresponding coordinates. The set of measurements form the original state of the CA. The CA rules are responsible for generating the intermediate heights among the real measurements. The proposed method can automatically adjust its rules, so as to encapture local morphological attributes, using a pre-processing procedure in the set of measurements. The main advantage of the proposed approach is the ability to maintain an accurately reconstruction even in cases where the number of measurements are significant reduced. Experiments have been conducted employing data collected from two totally different real-word environments. In the first case the proposed approach is applied, so as to build a detailed map of a large unknown underwater area in Oporto, Portugal. The second case concerns data collected by a team of aerial robots in real experiments in an area near Zurich, Switzerland and is also used for the evaluation of the approach. The data collected, in the two aforementioned cases, are extracted using different kind of sensors and robots, thus demonstrating the applicability of our approach in different kind of devices. The proposed method outperforms the performance of other well-known methods in literature thus enabling its application for real robot navigation.

A. Ch. Kapoutsis, S. A. Chatzichristofis, G. Ch. Sirakoulis, L. Doitsidis, and E. B. Kosmatopoulos, “Employing Cellular Automata for Shaping Accurate Morphology Maps Using Scattered Data from Robotic Missions”, Robots and Lattice Automata, Emergence, Complexity and Computation Volume 13, Springer-Verlag, G. Sirakoulis, A. Aadamatzky (Eds), pp 229-246, 12 October 2014.

The smart grid technology is emerging progressively, due to environmental and energy standards and to the forecasted exhaustion of non-renewable energy sources: currently, many smart microgrids small-scale renewable-energy generation systems are connected to the traditional grid, allowing the two grids to coexist. To perform efficiently load control and demand response management, appropriate algorithm for controlling smart microgrids must be developed. In addition to load control for reduction of energy consumption, climate control to optimize the thermal comfort of the occupants must be performed. This paper presents a microgrid simulation test case to evaluate the performance of different strategies for load management and thermal comfort optimization. The main contribution of the paper are: (1) contrary to many state-of the-art approaches that rely on simplified model, in this papers a realistic microgrid, modelled using an elaborate energy building simulation program (EnergyPlus) is employed for the synthesis and evaluation of the control strategies; (2) a realistic thermal comfort model, the Fanger index, is adopted and the thermal satisfaction of the end user is part of the performance index to be optimized; (3) the optimization of the energy load and of the solar energy distribution is performed jointly; far from optimal solutions can be achieved if one of the two tasks is neglected. Simulations using historical summer data demonstrate that via optimization and control we can effectively integrate the microgrid test case with the photovoltaic system that provides renewable solar energy, i.e. we can optimally balance the energy demand with the solar energy, while taking into account the thermal comfort of the occupants.

Athanasios Karagevrekis, Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos,”Interconnected Microgrids: An Energyplus Simulation Test Case”, Machines Review, July 2014

This paper describes a new control scheme for approximately optimal control (AOC) of nonlinear systems, convex control design (ConvCD). The key idea of ConvCD is to transform the approximate optimal control problem into a convex semi-definite programming (SDP) problem. Contrary to the majority of existing AOC designs where the problem that is addressed is to provide a control design which approximates the performance of the optimal controller by increasing the “controller complexity,” the proposed approach addresses a different problem: given a controller of “fixed complexity” it provides a control design that renders the controller as close to the optimal as possible and, moreover, the resulted closed-loop system stable. Two numerical examples are used to show the effectiveness of the method.

Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos, Antonis Papachristodoulou,”Convex Design Control for Practical Nonlinear Systems”,IEEE Transactions on Automatic Control, 3rd March 2014

Multi-objective robot exploration constitutes one of the most challenging tasks for autonomous robots performing in various operations and different environments. However, the optimal exploration path depends heavily on the objectives and constraints that both these operations and environments introduce. Typical environment constraints include partially known or completely unknown workspaces, limited-bandwidth communications, and sparse or dense clattered spaces. In such environments, the exploration robots must satisfy additional operational constraints, including time-critical goals, kinematic modeling, and resource limitations. Finding the optimal exploration path under these multiple constraints and objectives constitutes a challenging non-convex optimization problem. In our approach, we model the environment constraints in cost functions and utilize the cognitive-based adaptive optimization algorithm to meet time-critical objectives. The exploration path produced is optimal in the sense of globally minimizing the required time as well as maximizing the explored area of a partially unknown workspace. Since obstacles are sensed during operation, initial paths are possible to be blocked leading to a robot entrapment. A supervisor is triggered to signal a blocked passage and subsequently escape from the basin of cost function local minimum. Extensive simulations and comparisons in typical scenarios are presented to show the efficiency of the proposed approach.

Amanatiadis, S. A. Chatzichristofis, K. Charalampous, L. Doitsidis, E. B. Kosmatopoulos, P. Tsalides, A. Gasteratos and S. Roumeliotis, “A MULTI-OBJECTIVE EXPLORATION STRATEGY FOR MOBILE ROBOTS UNDER OPERATIONAL CONSTRAINTS”, “IEEE Access”, Volume 1, October 2013, pp 691-702, IEEE.

This paper deals with the problem of deploying a team of flying robots to perform surveillance-coverage missions over a terrain of arbitrary morphology. In such missions, a key factor for the successful completion of the mission is the knowledge of the terrain’s morphology. The focus of this paper is on the implementation of a two-step procedure that allows us to optimally align a team of flying vehicles for the aforementioned task. Initially, a single robot constructs a map of the area using a novel monocular-vision-based approach. A state-of-the-art visual-SLAM algorithm tracks the pose of the camera while, simultaneously, autonomously, building an incremental map of the environment. The map generated is processed and serves as an input to an optimization procedure using the cognitive, adaptive methodology initially introduced in Renzaglia et al. (Proceedings of the IEEE international conference on robotics and intelligent system (IROS), Taipei, Taiwan, pp. 3314–3320, 2010). The output of this procedure is the optimal arrangement of the robots team, which maximizes the monitored area. The efficiency of our approach is demonstrated using real data collected from aerial robots in different outdoor areas.

L. Doitsidis, S. Weiss, Al. Renzaglia, M. W. Achtelik, E. Kosmatopoulos, R. Siegwart, D. Scaramuzza, Optimal Surveillance Coverage for Teams of Micro Aerial Vehicles in GPS-Denied Environments using Onboad Vision, Autonomous Robots, vol. 33, no.1-2, pp. 173-188, 2012.

The problem of deploying a team of flying robots to perform surveillance coverage missions over an unknown terrain of complex and non-convex morphology is presented. In such a mission, the robots attempt to maximize the part of the terrain that is visible while keeping the distance between each point in the terrain and the closest team member as small as possible. A trade-off between these two objectives should be fulfilled given the physical constraints and limitations imposed at the particular application. As the terrain’s morphology is unknown and it can be quite complex and non-convex, standard algorithms are not applicable to the particular problem treated in this paper. To overcome this, a new approach based on the Cognitive-based Adaptive Optimization (CAO) algorithm is proposed and evaluated. A fundamental property of this approach is that it shares the same convergence characteristics as those of constrained gradient-descent algorithms (which require perfect knowledge of the terrain’s morphology and optimize surveillance coverage subject to the constraints the team has to satisfy). Rigorous mathematical arguments and extensive simulations establish that the proposed approach provides a scalable and efficient methodology that incorporates any particular physical constraints and limitations used to navigate the robots into an arrangement that (locally) optimizes surveillance coverage.

A. Renzaglia, L. Doitsidis, A. Martinelli, E. B. Kosmatopoulos, Multi-Robot 3D Coverage of Unknown Areas, International Journal of Robotics Research, vol. 31, no.6, pp. 738-752, May 2012.

Buildings nowadays are increasingly expected to meet higher and more complex performance requirements: they should be sustainable; use zero-net energy; foster a healthy and comfortable environment for the occupants; be grid-friendly, yet economical to build and maintain. The essential ingredients for the successful development and operation of net zero- and positive-energy buildings (NZEB/PEB) are: thermal simulation models, that are accurate representations of the building and its subsystems; sensors, actuators, and user interfaces to facilitate communication between the physical and simulation layers; and finally, integrated control and optimization tools of sufficient generality that using the sensor inputs and the thermal models can take intelligent decisions, in almost real-time, regarding the operation of the building and its subsystems. To this end the aim of the present paper is to present a review on the technological developments in each of the essential ingredients that may support the future integration of successful NZEB/PEB, i.e. accurate simulation models, sensors and actuators and last but not least the building optimization and control. The integration of the user is an integral part in the dynamic behavior of the system, and this role has to be taken into account. Future prospects and research trends are discussed.

D. Kolokotsa, D. Rovas, E. Kosmatopoulos, K. Kalaitzakis, A roadmap towards intelligent net zero- and positive-energy buildings, Solar Energy, vol. 85, issue 12, pp. 3067-3084, 2011.

In this paper, we study the problem of optimizing (fine-tuning) the design parameters of large-scale traffic control systems that are composed of distinct and mutually interacting modules. This problem usually requires a considerable amount of human effort and time to devote to the successful deployment and operation of traffic control systems due to the lack of an automated well-established systematic approach. We investigate the adaptive fine-tuning algorithm for determining the set of design parameters of two distinct mutually interacting modules of the traffic-responsive urban control (TUC) strategy, i.e., split and cycle, for the large-scale urban road network of the city of Chania, Greece. Simulation results are presented, demonstrating that the network performance in terms of the daily mean speed, which is attained by the proposed adaptive optimization methodology, is significantly better than the original TUC system in the case in which the aforementioned design parameters are manually fine-tuned to virtual perfection by the system operators.

A. Kouvelas, K. Aboudolas, E.B. Kosmatopoulos and M. Papageorgiou, Adaptive Performance Optimization for Large-Scale Traffic Control Systems, IEEE Transactions on Intelligent Transportation Systems, vol. 12, issue 4, pp. 1434-1445, 2011.

Learning mechanisms that operate in unknown environments should be able to efficiently deal with the problem of controlling unknown dynamical systems. Many approaches that deal with such a problem face the so-called exploitation-exploration dilemma where the controller has to sacrifice efficient performance for the sake of learning “better” control strategies than the ones already known: during the exploration period, poor or even unstable closed-loop system performance may be exhibited. In this paper, we show that, in the case where the control goal is to stabilize an unknown dynamical system by means of state feedback, exploitation and exploration can be concurrently performed without the need of sacrificing efficiency. This is made possible through an appropriate combination of recent results developed by the author in the areas of adaptive control and adaptive optimization and a new result on the convex construction of control Lyapunov functions for nonlinear systems. The resulting scheme guarantees arbitrarily good performance in the regions where the system is controllable. Theoretical analysis as well as simulation results on a particularly challenging control problem verify such a claim.

E. B. Kosmatopoulos, Control of Unknown Nonlinear Systems with Efficient Transient Performance using Concurrent Exploitation and Exploration, IEEE Transactions on Neural Networks, vol. 21, no. 8, pp. 1245-1261, Aug. 2010.

This technical note tackles the problem of constructing state-feedback stabilizers that guarantee good transient closed-loop performance when applied to general unknown nonlinear multiinput state-feedback stabilizable systems. An adaptive Control Lyapunov Function-based control scheme is proposed in order to address this problem. Mathematical analysis establishes that the proposed control scheme guarantees good closed-loop transient performance outside the regions where the system is uncontrollable, provided the controlled system admits a controllability-like property.

E. B. Kosmatopoulos, CLF-based Control Design for Unknown Multi-Input Nonlinear Systems with Good Transient Performance, IEEE Transactions on Automatic Control, vol. 55, no. 11, Nov. 2010.

Adaptive optimization (AO) schemes based on stochastic approximation principles such as the Random Directions Kiefer–Wolfowitz (RDKW), the Simultaneous Perturbation Stochastic Approximation (SPSA) and the Adaptive Fine-Tuning (AFT) algorithms possess the serious disadvantage of not guaranteeing satisfactory transient behavior due to their requirement for using random or random-like perturbations of the parameter vector. The use of random or random-like perturbations may lead to particularly large values of the objective function, which may result to severe poor performance or stability problems when these methods are applied to closed-loop controller optimization applications. In this paper, we introduce and analyze a new algorithm for alleviating this problem. Mathematical analysis establishes satisfactory transient performance and convergence of the proposed scheme under a general set of assumptions. Application of the proposed scheme to the adaptive optimization of a large-scale, complex control system demonstrates the efficiency of the proposed scheme.

E.B. Kosmatopoulos, An adaptive optimization scheme with satisfactory transient performance, Automatica, Vol. 45, No. 3, pp. 716-723, 2009.

Multi-objective robot exploration, constitutes one of the most challenging tasks for autonomous robots performing in various operations and different environments. However, the optimal exploration path depends heavily on the objectives and constraints that both hese operations and environments introduce. Typical environment constraints include partially known or completely unknown workspaces, limited-bandwidth communications and sparse or dense clattered spaces. In such environments, the exploration robots must satisfy additional operational constraints including time-critical goals, kinematic modeling and resource limitations. Finding the optimal exploration path under these multiple constraints and objectives constitutes a challenging non-convex optimization problem. In our approach, we model the environment constraints in cost functions and utilize the Cognitive-based Adaptive Optimization (CAO) algorithm in order to meet time-critical objectives. The exploration path produced is optimal in the sense of globally minimizing the required time as well as maximizing the explored area of a partially unknown workspace. Since obstacles are sensed during operation, initial paths are possible to be blocked leading to a robot entrapment. A supervisor is triggered to signal a blocked passage and subsequently escape from the basin of cost function local minimum. Extensive simulations and comparisons in typical scenarios are presented in order to show the efficiency of the proposed approach.

A. Amanatiadis, S. A. Chatzichristofis, K. Charalampous, L. Doitsidis, E. B. Kosmatopoulos, P. Tsalides, A. Gasteratos and S. Roumeliotis, A Multi-Objective Exploration Strategy for Mobile Robots under Operational Constraints, IEEE Access, IEEE, Accepted for Publication

Despite the continuous advances in the fields of intelligent control and computing, the design and deployment of efficient large scale nonlinear control systems (LNCSs) requires a tedious fine-tuning of the LNCS parameters before and during the actual system operation. In the majority of LNCSs the fine-tuning process is performed by experienced personnel based on field observations via experimentation with different combinations of controller parameters, without the use of a systematic approach. The existing adaptive/neural/fuzzy control methodologies cannot be used towards the development of a systematic, automated fine-tuning procedure for general LNCS due to the strict assumptions they impose on the controlled system dynamics; on the other hand, adaptive optimization methodologies fail to guarantee an efficient and safe performance during the fine-tuning process, mainly due to the fact that these methodologies involve the use of random perturbations. In this paper, we introduce and analyze, both by means of mathematical arguments and simulation experiments, a new learning/adaptive algorithm that can provide with convergent, an efficient and safe fine-tuning of general LNCS. The proposed algorithm consists of a combination of two different algorithms proposed by Kosmatopoulos (2007 and 2008) and the incremental-extreme learning machine neural networks (I-ELM-NNs). Among the nice properties of the proposed algorithm is that it significantly outperforms the algorithms proposed by Kosmatopoulos as well as other existing adaptive optimization algorithms. Moreover, contrary to the algorithms proposed by Kosmatopoulos , the proposed algorithm can operate efficiently in the case where the exogenous system inputs (e.g., disturbances, commands, demand, etc.) are unbounded signals.

E.B. Kosmatopoulos and A. Kouvelas, Large-Scale Nonlinear Control System Fine-Tuning through Learning, IEEE Transactions Neural Networks, Vol. 20, No. 6, pp. 1009-1023, 2009.

Recently, we introduced an adaptive control design for linearly parameterized multi-input nonlinear systems admitting a known control Lyapunov function (CLF) that depends on the unknown system parameters. The main advantage of that design is that it overcomes the problem where the estimation model becomes uncontrollable (at regions of the state space where the actual system is controllable). However, the resulted adaptive control design is quite complicated and, moreover, it exhibited poor transient behavior in various applications. In this technical note, we propose and analyze a new computationally efficient adaptive control design that overcomes the aforementioned shortcomings. The proposed design is based on an adaptive optimization algorithm introduced recently by the author, which makes sure that the parameters to be optimized (which correspond to the controller parameters in this technical note) are modified so as to both lead to a decrease of the function to be minimized and satisfy a persistence of excitation condition. The main advantage of the proposed adaptive control design is that it can produce arbitrarily good transient performance outside (a) the regions of the state space where the actual system becomes uncontrollable and (b) a region of the parameter estimates space which shrinks exponentially fast. It is also worth noting that the class of systems where the proposed algorithm is applicable is more general than that of our previous work; however, it has to be emphasized that, due to the fact that the proposed algorithm involves the use of random sequences, all the established stability and convergence results are guaranteed to hold with probability one.

E.B. Kosmatopoulos, Adaptive Control Design based on Adaptive Optimization Principles, IEEE Transactions on Automatic Control, Vol. 53, No. 11, pp. 2680-2685, 2008.

Conference abstracts and proceedings

The objective of this research is to evaluate the performance of a system of systems optimization algorithm, namely, L4G-PCAO, in building energy systems. Since the test bed of this research is an office building with more than two hundred occupiers, the heating and cooling demands of the building must always be fully satisfied. Consequently, changes in the currently-installed control system cannot be made forthrightly. Therefore, fresh ideas like implementation of new control strategies or optimization algorithms should be firstly put to the test via dynamic simulation, which makes engineers capable of examining new control and optimization strategies. The performance should then be analyzed and evaluated before implementing in the use case. This paper presents a strategy for simulative-based implementation of L4G-PCAO in a building energy system and also evaluates its performance. The results show that it is not only possible to conserve energy by applying this newly-developed optimization algorithm to existing control systems, but also it can shift the usage of energy sources in a more environment-friendly direction.

Roozbeh Sangi, Thomas Schild, Magnus Daum, Johannes Fütterer, Rita Streblow, Dirk Müller, Iakovos T. Michailidis, Elias B. Kosmatopoulos, “Simulation–based implementation and evaluation of a system of systems optimization algorithm in a building control system”, MED’16: The 24th Mediterranean Conference on Control and Automation, Athens, Greece.

Microgrids equipped with small-scale renewable energy generation systems and energy storage units offer challenging opportunity from a control point of view. In fact, in order to improve resilience and enable islanded mode, microgrid
energy management systems must dynamically manage controllable loads by considering not only matching energy generation and consumption, but also thermal comfort of the occupants. Thermal comfort, which is often neglected or
oversimplified, plays a major role in dynamic demand response, especially in front of intermittent behavior of the renewable energy sources. This paper presents a novel control algorithm for joint demand response management and thermal comfort optimization in a microgrid composed of a block of buildings,
a photovoltaic array, a wind turbine, and an energy storage unit. In order to address the large-scale nature of the problem, the proposed control strategy adopt a two-level supervisory strategy: at the lower level, each building employs a local controller that processes only local measurements; at the upper
level, a centralized unit supervises and updates the three controllers with the aim of minimizing the aggregate energy cost and thermal discomfort of the microgrid. Comparisons with alternative strategies reveal that the proposed supervisory strategy efficiently manages the demand response so as to sensibly improve independence of the microgrid with respect to the main grid, and guarantees at the same time thermal comfort of the occupants.

Christos D. Korkas, Simone Baldi, Iakovos T. Michailidis, Ioannis Boutalis, Elias B. Kosmatopoulos, “A Supervisory Approach to Microgrid Demand Response and Climate Control”, MED’16: The 24th Mediterranean Conference on Control and Automation, Athens, Greece

The improvement of traffic flow in cities, traditionally involves signal control strategies of a centralized logic. In this context, several methodological approaches have been proposed and used. These approaches have been proven quite efficient in cases where elaborate models are available and regularly updated, information from all corners of the system can be collected and managed in a central manner, and the scale of the resulting control problem is medium to small. Otherwise, they may fail to respond efficiently to the prevailing traffic conditions; thus stipulating the necessity of developing techniques able to handle efficiently large-scale control problems with steadily evolving and highly complex, nonlinear and stochastic dynamics. This is the main reason for the recently observed shift towards the development of scalable approaches, which, though based on a decentralized structure (by junction), will be capable to improve the traffic flow efficiency at network level with low design effort and control infrastructure investment. Such an approach is presented in this paper along with some preliminary microscopic simulation results and future plans.

Diamantis Manolis, Iakovos Michailidis, Christina Diakaki, Elias Kosmatopoulos, Ioannis Papamichail, Markos Papageorgiou, “An adaptive decentralized approach to the signal control of urban networks”, 9th Triennial Symposium On Transportation Analysis (TRISTAN IX), 13th June, Oranjestad, Aruba, 2016

Over the recent past years research effort has been dedicated towards addressing a generic solution in System of Systems (SoS) control problems. Two are the main obstacles that have to be bypassed in such problem cases, especially in real life applications, rendering the optimization problem into a complicated/challenging one: (i) modelling/simulation tools are usually used in order to construct an as-close-as-possible to reality accurate model, whose construction though requires a considerable amount of time and effort and (ii) furthermore, standard control system designs when applied to SoS exhibit poor performance as they are required to handle very high-dimensional problems. In this paper, we present a first attempt towards addressing these issues. More precisely, a new adaptive optimal control methodology is presented and evaluated. The main attributes of the proposed control methodology is its local nature with minimum requirements for coordination between the constituent system of the SoS and its model-free nature.

Elias B. Kosmatopoulos, Iakovos T. Michailidis, Christos D. Korkas, Christos Ravanis, “Local4Global Adaptive Optimization and control for System-of-Systems”,European Control Conference (ECC), 15th July, Linz, Austria, 2015

The demand-side energy management of microgrids comprising of buildings of heterogeneous nature (residential, commercial, industrial, etc.) and thus exhibiting heterogeneous occupancy pattern, requires the development of appropriate energy management systems (EMSs) that can integrate the maximum exploitation of the distributed energy resources like photovoltaic panels with the thermal comfort of the occupants. This paper presents a simulation-based optimization approach for the design of an EMS in grid-connected photovoltaic-equipped microgrids with heterogeneous buildings and occupancy schedules. The EMS optimizes a multi-objective criterion that takes into account both the energy cost and the thermal comfort of the aggregate microgrid. A three-building microgrid test case is used to demonstrate the effectiveness of the proposed approach: comparisons with alternative rule-based and optimization-based EMSs show that the proposed EMS strategy exploits the occupancy information to automatically change the energy demand of each building, resulting in improved energy cost and thermal comfort.

Christos D. Korkas, Simone Baldi, Iakovos T. Michailidis, Elias B. Kosmatopoulos, “Multi-objective control strategy for energy management of grid-connected heterogeneous microgrids”, American Control Conference (ACC), July 1st 2015, Chichago IL, USA

Suitable control measures and strategies must be taken to counteract the reduced throughput and the degradation of the network infrastructure caused by traffic congestion in urban networks. This paper studies and analyzes the performance of an adaptive traffic-responsive strategy that manages the traffic light parameters (the cycle time and the split time) in an urban network to reduce traffic congestion. The proposed traffic-responsive strategy adopts a nearly-optimal control formulation: first, an (approximate) solution of the HJB is parametrized via an appropriate Lyapunov positive definite matrix; then, the solution is updated via a procedure that generates candidate control strategies and selects at each iteration the best one based on the estimation of close-to-optimality and the information coming from the simulation model of the network (simulation-based design). Simulation results obtained using an AIMSUN model of the traffic network of Chania, Greece, an urban traffic network containing many varieties of junction staging, demonstrate the efficiency of the proposed approach.

Simone Baldi, Iakovos T. Michailidis, Vasiliki Ntampasi, Elias B. Kosmatopoulos,”Simulation-based synthesis for approximately optimal urban traffic light management”, American Control Conference (ACC), July 1st 2015, Chichago IL, USA

Within the project NOPTILUS, a fully functional system/methodology had been developed that allows the cooperative, fully-autonomous navigation of teams of AUVs when deployed in Static or Dynamic Underwater Map Construction (SDUMC) or Dynamic Underwater Phenomena Tracking (DUPT) missions. The key ingredient of this fully functional system/methodology (called the NOPTILUS Planning, Assignment and Navigation Module – NOPTILUS PAN) is an optimal control algorithm – called Parametrized Cognitive Adaptive Optmization – (PCAO) – developed by one of the NOPTILUS partners (CERTH). PCAO is firstly tailored and modified so as to be applicable to the problem of autonomous navigation of teams of AUVs when deployed in SDUMC or DUPT missions. For this purpose, a nonlinear model is developed so as to capture the dynamics of the AUVs, their sensors and the underwater environment. More precisely, the original PCAO-based approach is revised so as to be able to efficiently handle information coming from the localization module, the underwater acoustic communication module, the situation understanding module as well as instructions from the operator. The information coming from these modules is handled by the NOPTILUS PAN module without the need to enter in tedious re-design tasks. Two real-life experiments (involving teams of AUVs deployed in static mapping or simultaneous static mapping and dynamic target taking) demonstrate the efficiency and practicability of the NOPTILUS PAN module.

A. Ch. Kapoutsis, G. V. Salavasidis, S. A. Chatzichristos, J. Braga, J. Pinto, J. Borges de Sousa, Elias B. Kosmatopoulos “The NOPTILUS Project Overview: A Fully-Autonomous Navigation System of Teams of AUVs for Static/dynamic Underwater Map Construction”, Proceedings IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles, NGCUV2015, Girona, Spain, April 2015.

In this paper the experimental results coming from the application of a novel Building Optimization and Control (BOC) technique in two different large-scale buildings are exposed. This novel BOC technique, described in a previous companion paper (part I: optimization algorithms) and denominated as PCAO, shows (i) its ability to provide extremely fast and in a “plug-n-play” fashion an efficient BOC system no matter how large-scale and complex is the building and (ii) its ability to rapidly and efficiently adapt and calibrate itself so as to “quickly learn the best BOC policy” even in the case where a poor or no model for the building dynamics is available. It has to be emphasized that the two different buildings where the PCAO approach is applied and evaluated correspond to “difficult to control buildings”, i.e., efficient BOC system design for these two buildings is significantly more complex than that of an “average building”.

Iakovos T. Michailidis, Simone Baldi, Elias B. Kosmatopoulos, Ioannis Boutalis,”Optimization-based Active Techniques for Energy Efficient Building Control Part II: Real-life Experimental Results”,International Conference on Buildings Energy Efficiency and Renewable Energy Sources, BEE RES 2014, June 1st-3rd, Kozani, Greece, 2014

Building Optimization and Control (BOC) deals with the deployment of active techniques, aiming at efficiently controlling the active elements of a building (HVACs, concrete activation elements, etc.): these techniques can result in tremendous energy savings, without the need for passive expensive solutions (new materials for passive insulation, the use of glass facades etc.). In this paper, we present a new approach for BOC system design – abbreviated as PCAO – with the following attributes: (i) providing extremely fast and in a “plug-n-play” fashion an efficient BOC system no matter how large-scale and complex the building is (ii) rapidly and efficiently adapting and calibrating itself so as to “quickly learn the best BOC policy” even in the case where a poor or no model for the building dynamics is available. This paper mainly deals with the algorithmic details of the proposed approach: real-life results coming from the application of PCAO in two large scale buildings are presented in a companion paper (part II: real-life experimental results).

Iakovos T. Michailidis, Simone Baldi, Elias B. Kosmatopoulos, Ioannis Boutalis,”Optimization-based Active Techniques for Energy Efficient Building Control Part I: Optimization Algorithms”,International Conference on Buildings Energy Efficiency and Renewable Energy Sources, BEE RES 2014, June 1st-3rd, Kozani, Greece, 2014

Recently, there has been a growing interest towards simulation-based control design (co-simulation), where the controller utilizes an optimizer to minimize or maximize an objective function (system performance) whose evaluation involves simulation of the system to be controlled. However, existing simulation-based approaches are not able to handle in a computationally efficient way large-scale optimization problems involving hundreds or thousands of states and parameters. In this paper, we propose and analyze a new simulation-based control design approach, employing an adaptive optimization algorithm capable of efficiently handle large-scale control problems. The convergence properties of the proposed algorithm are established. Simulation results exhibit efficiency of the proposed approach when applied to large-scale problems. The simulation results employ two large-scale real-life systems for which conventional popular optimization techniques totally fail to provide an efficient simulation-based control design.

Simone Baldi, Iakovos T. Michailidis, Hossein Jula, Elias B. Kosmatopoulos,”A ‘Plug-n- Play’ Computationally Efficient Approach for Control Design of Large-Scale Nonlinear Systems using co-Simulation”,52nd IEEE Conference on Decision and Control, 10th December, Florence, Italy, 2013

Many Control Design approaches presuppose the knowledge of a state space description of the system. As in many real and large scale systems the construction of a state space model using analytical equations and rough assumptions is a time consuming process leading most of the times to questionable precision. System simulation software has been developed for several applications over the recent past decades. For this purpose a System Identification approach is proposed in the present paper. The idea of the System Identification from a simulation model using statistical measurements from the simulation output data has been of great interested for optimal control schemes which require accurate and furthermore, agile, simple and practical models to handle. In practice real systems and accurate simulation models, show nonlinear dynamic behavior which concludes to linear approximation-identification models being unsuitable for this purpose. The proposed Multi-Linear State Space Model Identification process is using the idea of mixing signals as weighted contribution factors for the final model behavior. A common Least Square error minimization process is used to minimize the model approximation error.

Iakovos T. Michailidis, Martin F. Pichler, Elias B. Kosmatopoulos,”Multi-Linear State Space Model Identification for Large Scale Building Systems”,Sustainable Building Conference, Graz, Austria, 1480, 25th September 2013

In this paper we present a solution to the problem of positioning a team of Micro Aerial Vehicles for a surveillance task in an environment of arbitrary and unknown morphology. The problem is addressed taking into account physical and environmental constraints like limited sensor capabilities and obstacle avoidance. The goal is to maximize the area monitored by the team, by identifying the best configuration of the team members. The proposed method is a distributed extension of our previous work based on the Cognitive Adaptive Optimization (CAO) algorithm. This distributed and scalable approach allows us to obtain coordinated and safe trajectories to accomplish the task in 3D environments. The different formulation of the problem considered in this paper allows also dealing with communication constraints. We provide extensive experimental results using data collected by a team of aerial robots and compare the efficiency of the distributed and centralized approach.

A. Renzaglia, L. Doitsidis, S. A. Chatzichristofis, A. Martinelli and E. B. Kosmatopoulos, “DISTRIBUTED MULTI-ROBOT COVERAGE USING MICRO AERIAL VEHICLES”, 21st Mediterranean Conference on Control and Automation, (MED’13), pp. 963 – 968, June 25 – 28, 2013, Platanias-Chania, Crete – Greece.

In this paper, we present a new approach that is able to efficiently and fully-autonomously navigate a team of Unmanned Aerial or Underwater Vehicles (UAUV’s) when deployed in exploration of unknown static and dynamic environments towards providing accurate static/dynamic maps of the environment. Additionally to achieving to efficiently and fully-autonomously navigate the UAUV team, the proposed approach possesses certain advantages such as its extremely computational simplicity and scalability, and the fact that it can very straightforwardly embed and type of physical or other constraints and limitations (e.g., obstacle avoidance, nonlinear sensor noise models, localization fading environments, etc).

A. Ch. Kapoutsis, S. A. Chatzichristofis, L. Doitsidis, J. Borges de Sousa and E. B. Kosmatopoulos, “AUTONOMOUS NAVIGATION OF TEAMS OF UNMANNED AERIAL OR UNDERWATER VEHICLES FOR EXPLORATION OF UNKNOWN STATIC & DYNAMIC ENVIRONMENTS”, 21st Mediterranean Conference on Control and Automation, (MED’13), pp. 1181 – 1188, June 25 – 28, 2013, Platanias-Chania, Crete – Greece.

M. Achtelik, M. Achtelik, Y. Brunet, M. Chli, S. A. Chatzichristofis, J.D. Decotignie, K.M. Doth, F. Fraundorfer, L. Kneip, D. Gurdan, L. Heng, E. Kosmatopoulos, L. Doitsidis, G.H. Lee, S. Lynen, A. Martinelli, L. Meier, M. Pollefeys, D. Piguet, A. Renzaglia, D. Scaramuzza, R. Siegwart, J. Stumpf, P. Tanskanen, C. Troiani and S. Weiss, “SFly: Swarm of Micro Flying Robots”,in proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2012), Vilamura, Algarve, Portugal, October 7 – October 12, 2012. (finalist for the best video award)

Current multi-AUV systems are far from being capable of fully autonomously taking over real-life complex situation-awareness operations. As such operations require advanced reasoning and decision-making abilities, current designs have to heavily rely on human operators. The involvement of humans, however, is by no means a guarantee of performance; humans can easily be over-whelmed by the information overload, fatigue can act detrimentally to their performance, properly coordinating vehicles actions is hard, and continuous operation is all but impossible. Within the European funded project NOPTILUS we take the view that an effective fully-autonomous multi-AUV concept/system, is capable of overcoming these shortcomings, by replacing human-operated operations by a fully autonomous one. In this paper, we present a new approach that is able to efficiently and fully-autonomously navigate a team of AUVs when deployed in exploration of unknown static and dynamic environments towards providing accurate static/dynamic maps of the environment. Additionally to achieving to efficiently and fully-autonomously navigate the AUV team, the proposed approach possesses certain advantages such as its extremely computational simplicity and scalability, and the fact that it can very straightforwardly embed and type of physical or other constraints and limitations (e.g., obstacle avoidance, nonlinear sensor noise models, localization fading environments, etc).

S. Chatzichristofis, A. Kapoutsis, E. Kosmatopoulos, L. Doitsidis, D. Rovas , and J. Borges de Sousa, “The NOPTILUS project: autonomous multi-AUV navigation for exploration of unknown environments”, Proceedings of the IFAC Workshop – Navigation, Guidance and Control of Underwater Vehicles (NGCUV 2012), Porto, April 2012.

Book chapters

Nowadays, most energy savings and thermal comfort improvements in buildings rely on passive design and construction techniques like superinsulation, passive solar heat gain and shading. Despite the benefits of passive design, active techniques for Building Optimization and Control (BOC) can result in relevant energy savings in everyday buildings without the need for expensive investments associated with passive solutions. The essence of BOC algorithms lies in controlling the Heating, Ventilation, and Air Conditioning (HVAC) system of a building to manage energy demands and improve thermal comfort conditions for occupants. In order to develop efficient BOC systems two aspects that must be handled efficiently are the large-scale nonlinear nature of the optimization problem and the uncertainty in the dynamics of the building. This chapter addresses these two aspects by proposing an adaptive optimization approach aiming at adaptively solving the optimal energy efficient control building problem. Real-life experimental results coming from the application of the proposed control approach in two different office buildings are presented. The two buildings differ for climate zone (Mediterranean vs Continental), HVAC system (air conditioners vs radiant slabs), seasonal climate control problem (cooling vs heating) and construction techniques (poor thermal insulation vs well insulated building). The proposed optimization strategy is able to handle the large-scale nature of the optimization problem, taking advantage of the peculiar characteristics of each building, and self-tuning so as achieve significant improvements in both in energy savings and thermal comfort.

Iakovos T. Michailidis, Simone Baldi, Elias B. Kosmatopoulos, Martin F. Pichler, Juan R. Santiago, “Improving Energy Savings and Thermal Comfort in Large-scale Buildings via Adaptive Optimization”, Nova Science Publishers, Francisco Miranda (Ed.), pp. 315-336, 2015. ISBN: 978-1-63482-707-22

Other publications

adminPublications