Anastasios Kouvelas

Large Scale Nonlinear Control System Fine-Tuning Through Learning

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.

Adaptive Fine-Tuning of Nonlinear Control Systems With Application to the Urban Traffic Control Strategy TUC

Practical large-scale nonlinear control systems require an intensive and time-consuming effort for the fine-tuning of their control parameters in order to achieve a satisfactory performance. In most cases, the fine-tuning process may take years and is performed by experienced personnel. The purpose of this paper is to introduce and analyze a systematic approach for the automatic fine-tuning of the control parameters of practical large-scale nonlinear control systems and investigate its efficiency when applied to the recently developed urban traffic control strategy traffic-responsive urban control. The proposed approach is based on a concept similar to the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm. The difference between the SPSA algorithm and the proposed approach is that, SPSA employs an approximation of the gradient of an appropriate objective function using only the most recent fine-tuning experiments, while in the proposed approach the approximation of the gradient is performed by using a linear-in-the-parameters approximator that incorporates information of a user-specified time-window of the past experiments. Mathematical analysis of the proposed approach establishes its convergence properties and that SPSA can be regarded as a special case of the proposed approach. Simulation results using the traffic network of the city of Chania, Greece-a typical urban traffic network containing all possible varieties of complex junction staging-demonstrate the efficiency of the proposed approach.

Adaptive Performance Optimization for Large-Scale Traffic Control Systems

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.

Maximum Pressure Controller for Stabilizing Queues in Signalized Arterial Networks

In this paper the problem of arterial signal control is considered. Urban intersections face serious congestion problems, but the installation and maintenance of centralized systems is deemed cumbersome. A decentralized approach that is relatively simple to implement is studied. The recently proposed maximum pressure controller, demonstrated to stabilize queues in arterial traffic systems, is tested in simulations. Modifications of the controller are analyzed and compared under the same demand scenarios. The mesoscopic model used for the simulation experiments is an extended version of the store-and-forward model and emulates the arterial traffic network as a queuing system. The results demonstrate the efficiency of the maximum pressure algorithm, which, under certain conditions, can stabilize all queues in the system.

Real-time merging traffic control for throughput maximization at motorway work zones

Work zones on motorways necessitate the drop of one or more lanes which may lead to significant reduction of traffic flow capacity and efficiency, traffic flow disruptions, congestion creation, and increased accident risk. Real-time traffic control by use of green–red traffic signals at the motorway mainstream is proposed in order to achieve safer merging of vehicles entering the work zone and, at the same time, maximize throughput and reduce travel delays. A significant issue that had been neglected in previous research is the investigation of the impact of distance between the merge area and the traffic lights so as to achieve, in combination with the employed real-time traffic control strategy, the most efficient merging of vehicles. The control strategy applied for real-time signal operation is based on an ALINEA-like proportional–integral (PI-type) feedback regulator. In order to achieve maximum performance of the control strategy, some calibration of the regulator’s parameters may be necessary. The calibration is first conducted manually, via a typical trial-and-error procedure. In an additional investigation, the recently proposed learning/adaptive fine-tuning (AFT) algorithm is employed in order to automatically fine-tune the regulator parameters. Experiments conducted with a microscopic simulator for a hypothetical work zone infrastructure, demonstrate the potential high benefits of

Feedback Perimeter Control for Heterogeneous Urban Networks Using Adaptive Optimization

A control scheme for heterogeneous transportation networks is presented. The methodology is based on the concept of the Macroscopic Fundamental Diagram (MFD) integrated with an adaptive optimization technique. The heterogeneous transportation network is first partitioned into a number of regions with homogeneous traffic conditions and well-defined MFDs. A macroscopic MFD-based model is used to describe the traffic dynamics of the resulting multi-region transportation system. A multivariable proportional integral (PI) feedback regulator is implemented to control the nonlinear system in real-time. The control variables consist of the inter-transferring flows between neighbourhood regions and the actuators correspond to the traffic lights of these areas (e.g. boundaries between regions). The recently proposed Adaptive Fine-Tuning (AFT) algorithm is used to optimize the gain matrices as well as the vector with the set-points of the PI controller. AFT is an iterative adaptive algorithm that optimizes the values of the tuneable parameters of the controller (e.g. gains and set-points) based on measurements of a performance index (e.g. total delay) for different perturbations of the parameters. The overall control scheme is tested in simulation and different performance criteria are studied. The performance of a fixed-time policy is compared to the final controller that is obtained after the convergence of AFT.

Real-time estimation of critical vehicle accumulation for maximum network throughput

Perimeter traffic flow control has recently been found to be a practical and efficient control scheme in mitigating traffic congestion in urban road networks. This control scheme aims at stabilising the accumulation of vehicles of the socalled network fundamental diagram near critical accumulation to achieve maximum network throughput. Nevertheless, the maximum throughput in urban road networks may be observed over a range of accumulation-values. In this work, an adaptive perimeter flow control strategy is proposed that allows the automatic monitoring of the critical accumulation to help maintain the accumulation near the optimal range of accumulation-values, while networks throughput is maximised. To this end, we design a Kalman filter-based estimation scheme that utilises real-time measurements of circulating flow and accumulation of vehicles to produce estimates of the currently prevailing critical accumulation. We use real data from an urban area with 70 sensors and show that the area exhibits a network fundamental diagram with low scatter. We demonstrate that the fundamental diagram is reproduced under different days but its shape and critical occupancy depend on the applied semi-real-time signal control and the distribution of congestion in the network. Results from the application of the estimation algorithm to the experimental data indicate good estimation accuracy and performance, and rapid tracking behaviour.

Application of the Signal Control Strategy TUC in Three Traffic Networks: Comparative Evaluation Results

The recently developed network-wide real-time signal control strategy TUC has been implemented in three traffic networks with quite different traffic and control infrastructure characteristics: Chania, Greece (23 junctions); Southampton, U.K. (53 junctions); and Munich, Germany (25 junctions), where it has been compared to the respective resident real-time signal control strategies TASS, SCOOT and BALANCE. The paper describes the three application networks; the application, demonstration and evaluation conditions; as well as the comparative evaluation results. The main conclusions drawn from this undertaking is that TUC is an easy-to-implement, inter-operable, low-cost real-time signal control strategy whose performance, after very limited fine-tuning, proved to be better or, at least, similar to the ones achieved by long-standing strategies that were in most cases very well fine-tuned over the years in the specific networks

AFT2: An Automated Maintenance and Calibration Tool for Traffic Management & Control Systems

Currently, a tremendous amount of human effort and time is spent for maintenance and calibration of operations of Transport Management & Control Systems (TMCSs). TMCS maintenance and calibration is usually performed by experienced personnel in the lack of an automated and systematic approach with no guarantee that the overall maintenance procedure will end-up successfully. Severe congestion, delay and safety problems may occur during manually-based maintenance and calibration activities, which usually take from several months to few years until completed. AFT2 is aiming at replacing the manually-based TMCS maintenance and calibration by a fully-automated procedure applicable to general TMCSs. The approach of AFT2 is based on a recently introduced Adaptive Optimization (AO) methodology which was proven - using rigorous mathematical arguments - to provide with safe and reliable, efficient and rapid maintenance and calibration of general TMCSs. The objective of the present paper is to demonstrate AFT2 efficiency through its application - in simulation - to three different complex, large-scale road TMCSs.

Exploring the impact of autonomous vehicles in urban networks and potential new capabilities for perimeter control

Recently there has been an enormous effort towards the development and field deployment of autonomous vehicles (AV). These developments have a direct impact on road traffic characteristics, as AV can also be used as sensors, actuators, and generally as effective components of an online traffic management system. In this work, we explore through microsimulation how AV can affect the network performance on urban networks. It is demonstrated through simulation experiments, that for mixed traffic (i.e. AV and conventional vehicles), different penetration rates of AV can reduce the density (i.e. congestion) and increase the capacity (i.e. throughput) of the network. Furthermore, new emerging capabilities for perimeter control in urban regions are explored, by utilizing new sources of data that can become available. A new feedback-based regulator is developed, which uses AV speed measurements as the state-feedback variables. The regulator is demonstrated to work equivalently well with a well fine-tuned regulator that uses loop detectors occupancy measurements (i.e. accumulation estimates) from all the links of the network. The speed-based feedback perimeter flow regulator performs significantly well even for very low penetration rates of AV (e.g. 5%), resulting in reduced delays for the users and less congestion in the urban region. Although the regulator has sparse measurements only from AV it can achieve the same improvement with the accumulation-based regulator that has the whole picture of the network.