Sebastian Senge

BeeJamA: A Distributed, Self-Adaptive Vehicle Routing Guidance Approach

We present and evaluate our distributed and self-adaptive vehicle routing guidance approach, termed BeeJamA, which provides drivers safely with routing directions well before each intersection. Our approach is based on a multiagent system, which is inspired by the honey bee foraging behavior. It relies on a distributed vehicle-to-infrastructure architecture. On the basis of microscopic traffic simulations under varying penetration rates, we show that BeeJamA outperforms dynamic shortest path algorithms with respect to average (global) travel times and regarding congestion avoidance.

A novel class of multi-agent algorithms for highly dynamic transport planning inspired by honey bee behavior

Commercial transport planning as well as individual intra-city or inter-city traffic in densely populated regions, both in Europe and the US, increasingly suffer from congestion problems, to an extent which e.g. affects predictable transport planning substantially (except - so far - for overnight tours). Due to the highly dynamic character of congestion forming and dissolving, no static approach like shortest path finding, applied globally or individually in car navigators, is adequate here: Its use even makes things worse as can be frequently observed. In this paper we present a completely decentralized multi-agent approach (termed BeeJamA) on multiple layers where car or truck routing are handled through algorithms adapted from the BeeHive algorithms which in turn have been derived from honey bee behavior. We report on extensive distributed simulation experiments in the BeeJamA project which demonstrate a very substantial improvement over traditional congestion handling.

Highly Dynamic and Adaptive Traffic Congestion Avoidance in Real-Time Inspired by Honey Bee Behavior

Traffic congestions have become a major problem in metropolitan areas world-wide, within and between cities, to an extent where they make driving and transportation times largely unpredictable. Due to the highly dynamic character of congestion building and dissolving this phenomenon appears even to resist a formal treatment. Static approaches, and even more their global management, have proven counterproductive in practice. Given the latest progress in VANET technology and the remarkable commercially driven efforts like in the European C2C consortium, or the VSC Project in the US, allow meanwhile to tackle various aspects of traffic regulation through VANET communication. In this paper we introduce a novel, completely decentralized multi-agent routing algorithm (termed BeeJamA) which we have derived from the foraging behavior of honey bees. It is highly dynamic, adaptive, robust, and scalable, and it allows for both avoiding congestions, and minimizing traveling times to individual destinations. Vehicle guidance is provided well ahead of every intersection, depending on the individual speeds. Thus strict deadlines are imposed on, and respected by, the BeeJamA algorithm. We report on extensive simulation experiments which show the superior performance of BeeJamA over conventional approaches.

2-Way evaluation of the distributed BeeJamA vehicle routing approach

We present and evaluate our adaptive and distributed vehicle routing approach, termed BeeJamA, which provides drivers safely with routing directions well before each intersection. Our approach is based on a multi-agent system which is inspired by the honey bee behavior and relies on a V2I architecture. We report on our extensive simulation experiments verifying for very large systems that BeeJamA substantially outperforms all A*-based algorithms relying on global information systems, in particular under all degrees of penetration rates as well as considering reactive flexibility and easy scalability.

Marginal cost pricing and multi-criteria routing in a distributed swarm-intelligence approach for online vehicle guidance

In this paper we investigate the effects of marginal cost pricing and multi-criteria routing as an addition of our distributed real-time vehicle guidance protocol BeeJamA. BeeJamA is is honey bee-inspired swarm intelligence approach for minimizing individual travel times based on a vehicle-to-infrastructure architecture. We propose changes to the swarm behavior allowing for efficient dissemination of marginal costs and multi-criteria routing information. In literature, marginal cost pricing is known as a possibility for reducing global travel times, potentially to the disadvantage of individual travel times. Although we can confirm an improvement over the plain BeeJamA protocol for a microscopic simulation setup, marginal cost pricing could not outperform a path reservation variant of BeeJamA. Distributed routing protocols for computer networks and particularly for vehicle routing often do not provide for multi-criteria decisions and BeeJamA was no exception so far. So, as a second contribution of this paper, the protocol is complemented by multi-criteria routing concepts.

Bee-Inpired Road Traffic Control as an Example of Swarm Intelligence in Cyber-Physical Systems

Physical parts of a system to interacting software components. In many cyber-physical systems considered nowadays, the systems parts form a network structure with concurrent entities and the need of seamless scaling and fault-tolerant information dissemination and decentralized methods. In this contribution we point out that swarm intelligence approaches may be well suited for CPS with networked components. As an example, we present our self-adaptive and distributed vehicle route guidance approach, termed BeeJamA, which provides drivers safely with routing directions well before each intersection. Our approach is based on a multi-agent system which is inspired by the honey bee behavior and relies on a decentralized vehicle-to-infrastructure architecture.

Minimizing vehicular travel times using the multi-agent system beejama

We present and evaluate our self-adaptive and distributed vehicle route guidance approach, termed BeeJamA, which provides drivers safely with routing directions well before each intersection. Our approach is based on a multi-agent system which is inspired by the honey bee behavior and relies on a decentralized vehicle-to-infrastructure architecture. On the basis of microscopic traffic simulations under varying penetration rates it shows that BeeJamA has the tendency to outperform dynamic shortest path algorithms with respect to (global) travel times.

Bee Inspired Bottom-Up Self-Organization in Vehicular Traffic Management

Traffic in densely populated regions increasingly suffers from congestion problems, to an extent which e.g. substantially affects predictable transport planning. Due to the highly dynamic character of congestion forming and dissolving, no static approach like shortest path finding, applied globally or individually in car navigators, is adequate here. In this paper we outline our current work on a completely decentralized multi-agent bottom-up approach (termed BeeJamAJ) on multiple layers where car routing is handled through algorithms derived from honey bee behavior.

Towards hybrid simulation of self-organizing and distributed vehicle routing in large traffic systems

Traffic congestions have been a major problem in metropolitan areas worldwide, causing enormous economical as well as ecological damage. At the same time, in densely populated areas with high vehicle traffic, central information gathering and distribution to vehicles takes too long for providing accurate, let alone optimal routing directions (which would have to be available in due time before vehicles arrive at road intersections). Accurate information provided too late may even add to congestion problems. In this paper we present a bottom-up, multi-agent online approach termed BeeJamA (Bee-Inspired Traffic Jam Avoidance) for individual vehicle routing which, on its network communication layer, is taking advantage of our novel self — organizing network routing algorithm BeeHive/BeeAdhoc. This Swarm Intelligence based method has been largely inspired by the behavior of honey bees. As a distributed algorithm BeeJamA does not rely on global information, and scalability is not a critical issue. BeeJamA features dynamic deadlines. The quality of the algorithm has a strong impact on the acceptance rate through the drivers, for installing and operating communication features (navigators and routing-related software) as well as on driver adherence to routing directions. This, in turn, requires a high amount of flexibility for routing algorithms considering (unpredictable) resetting of destinations by drivers, making dynamic real-time reactions a critical issue. For a comprehensive and comparative realistic evaluation reflecting the aforementioned aspects/parameters we have developed a generic routing framework (GRF) which allows to run BeeJamA and other routing algorithms on different scientific or commercial traffic simulators. (Each of them serves different purposes and is therefore considerably abstracting from reality.) While we (briefly) report on extensive simulation experiments on the MAT-Sim simulator which verify BeeJamAs superior performance compared to existing models we will also outline — as part of our current research — how to create an incremental procedure for performing realistic field studies where in ever larger areas the abstract simulation is replaced, and observed, through real traffic. This imposes very strict requirements for the real-time, or online, performance of the simulator. Comprehensive results results of these altogether novel experimental investigations will be subject of upcoming publications.

Bienen-inspiriertes Straßenverkehrsrouting

Verkehrsstaus verursachen hohe Kosten — in okonomischer als auch in okologischer Hinsicht. Verkehrsstaus konnen als verteilte und dynamische Probleme betrachtet werden, da keine zentrale Regelinstanz, sondern im Gegenteil jeder Fahrer (im Wesentlichen unabhangig voneinander) uber den Beginn und die Strecke der Fahrt selbst entscheidet. Wir argumentieren, dass ein adaquater Losungsansatz zur Stauvermeidung (sowie zur Minimierung von Reisezeiten und Umweltbelastung) auf globale Informationen verzichten muss, um auf dynamische Anderungen des Verkehrsaufkommen in Echtzeit reagieren zu konnen, in Strasenverkehrsnetzen realistischer Grose. In diesem Beitrag wird das verteilte Routingsystem BeeJamA in seiner Weiterentwicklung als Vehicleto-Infrastructure (V21) Architektur beschrieben, zur inzwischen erfolgenden realitatsnahen Evaluierung im MATSim Simulator.