Holger Prothmann

Observation and Control of Organic Systems

Organic Computing (OC) assumes that current trends and recent developments in computing, like growing interconnectedness and increasing computational power, pose new challenges to designers and users. In order to tackle the upcoming demands, OC has the vision to make systems more life-like (organic) by endowing them with abilities such as self-organisation, self-configuration, self-repair, or adaptation. Distributing computational intelligence by introducing concepts like self-organisation relieves the designer from exactly specifying the low-level system behaviour in all possible situations. In addition, the user has the possibility to define a few high-level goals, rather than having to manipulate many low-level parameters.

Organic traffic light control for urban road networks

In recent years, autonomic and organic computing have become areas of active research in the informatics community. Both initiatives aim at handling the growing complexity in technical systems by focusing on adaptation and self-optimisation capabilities. A promising application for organic concepts is the control of road traffic signals in urban areas. This article presents an organic approach to traffic light control in urban areas that exhibits adaptation and learning capabilities, allowing traffic lights to autonomously react on changing traffic conditions. A coordination mechanism for neighbouring traffic lights is presented that relies solely on locally available traffic data and communication among neighbouring intersections, resulting in a distributed and self-organising traffic system for urban areas. The organic systems efficiency is demonstrated in a simulation-based evaluation.

Organic traffic control

Urban road networks are an infrastructural key factor for modern cities. To facilitate an efficient transportation of people and goods, it is crucial to optimise the networks’ signalisation and to route drivers quickly to their destination. As road networks are widespread and their traffic demands are dynamically changing, adaptive and self-organising (and therefore organic) control systems are required. This article demonstrates the potential benefits of organic traffic control: It presents an Observer/Controller that optimises an intersection’s signalisation and introduces a self-organising coordination mechanism that allows for the traffic-responsive creation of progressive signal systems (or green waves). All presented mechanisms advance the state of the art and help to reduce the negative environmental and economical impact of traffic.

Decentralised Progressive Signal Systems for Organic Traffic Control

An increased mobility and the resulting rising traffic demands lead to serious congestion problems in many cities. Although there is not a single solution that will solve traffic congestion and the related environmental and economical problems, traffic light coordination is an important factor in achieving efficient networks. This paper presents a new distributed approach for dynamic traffic light coordination that relies on locally available traffic data and communication among neighboring intersections. The coordination mechanism is combined with an organic traffic control approach to form an adaptive, distributed control system with learning capabilities. The efficiency of the resulting organic system is demonstrated in a simulation-based evaluation.

Possibilities and limitations of decentralised traffic control systems

Due to steadily increasing mobility and the resulting rising traffic demands, serious congestion problems can be observed in many cities. One promising approach to alleviate the congestion effects is the coordination of the networks traffic signals in response to the traffic flow. The recently introduced Decentralised Progressive Signal Systems approach is an adaptive coordination mechanism for traffic signals in urban road networks that relies on local traffic data only. Since the decentralised process cannot lead to optimal results in some special cases, it is extended with an optional hierarchical component introduced in this paper. Based on a broader view on the current network traffic, this Regional Manager is responsible for determining which intersections are coordinated. The efficiency of the coordination determined by the Regional Manager is demonstrated in a simulation-based evaluation that considers the decentralised mechanism and an uncoordinated system for comparison.

Improving XCS Performance by Distribution

Learning Classifier Systems (LCSs) are rule-based evolutionary reinforcement learning (RL) systems. Today, especially variants of Wilsons eXtended Classifier System (XCS) are widely applied for machine learning. Despite their widespread application, LCSs have drawbacks: The number of reinforcement cycles an LCS requires for learning largely depends on the complexity of the learning task. A straightforward way to reduce this complexity is to split the task into smaller sub-problems. Whenever this can be done, the performance should be improved significantly. In this paper, a nature-inspired multi-agent scenario is used to evaluate and compare different distributed LCS variants. Results show that improvements in learning speed can be achieved by cleverly dividing a problem into smaller learning sub-problems.

Self-Organised routing for road networks

Increasing mobility and the resulting rising traffic demands cause serious problems in urban regions world-wide. Approaches to alleviate the negative effects of traffic include an improved control of traffic lights and the introduction of dynamic route guidance systems that take current conditions into account. One solution for the former aspect is Organic Traffic Control (OTC) which provides a self-organised and self-adaptive system founded on the principles of Organic Computing. Based on OTC, this paper introduces a novel concept to dynamic route guidance in urban road networks. Inspired by the well-known protocols Distance Vector Routing and Link State Routing from the Internet domain, the major goal of the route guidance mechanism is to increase the networks robustness with respect to congested or blocked roads. The efficiency of the developed approach is demonstrated in a simulation-based evaluation that considers disturbed and undisturbed traffic conditions.

XCS revisited: a novel discovery component for the eXtended classifier system

The eXtended Classifier System (XCS) is a rule-based evolutionary on-line learning system. Originally proposed by Wilson, XCS combines techniques from reinforcement learning and evolutionary optimization to learn a population of maximally general, but accurate condition-action rules. This paper focuses on the discovery component of XCS that is responsible for the creation and deletion of rules. A novel rule combining mechanism is proposed that infers maximally general rules from the existing population. Rule combining is evaluated for single- and multi-step learning problems using the well-known multiplexer, Woods, and Maze environments. Results indicate that the novel mechanism allows for faster learning rates and a reduced population size compared to the original XCS implementation.

Decentralised Route Guidance in Organic Traffic Control

Increasing mobility and the resulting rising traffic demands cause serious problems in urban regions world-wide. Approaches to alleviate the negative effects of traffic include an improved control of traffic lights and the introduction of dynamic route guidance (DRG) systems that take current conditions into account. Based on the self-organising Organic Traffic Control system, we introduce a novel DRG concept for urban road networks. It is inspired by the well-known Distance Vector Routing protocol from the Internet domain and increases the networks robustness with respect to congested or blocked roads. We demonstrate the efficiency of the developed approach in a simulation-based evaluation.