Rutger Claes

Rutger Claes

Auto-generated profile, based on publicly available information

Rutger Claes

Katholieke Universiteit Leuven

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 March 2011 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
RCRutger Claes
Rutger Claes
THTom Holvoet
Tom Holvoet
DWDanny Weyns
Danny Weyns
Advanced vehicle guidance systems use real-time traffic information to route traffic and to avoid congestion. Unfortunately, these systems can only react upon the presence of traffic jams and not to prevent the creation of unnecessary congestion. Anticipatory vehicle routing is promising in that respect, because this approach allows directing vehicle routing by accounting for traffic forecast information. This paper presents a decentralized approach for anticipatory vehicle routing that is particularly useful in large-scale dynamic environments. The approach is based on delegate multiagent systems, i.e., an environment-centric coordination mechanism that is, in part, inspired by ant behavior. Antlike agents explore the environment on behalf of vehicles and detect a congestion forecast, allowing vehicles to reroute. The approach is explained in depth and is evaluated by comparison with three alternative routing strategies. The experiments are done in simulation of a real-world traffic environment. The experiments indicate a considerable performance gain compared with the most advanced strategy under test, i.e., a traffic-message-channel-based routing strategy.

1 May 2011
Verified
Conference Paper
Closed access
RCRutger Claes
Rutger Claes
THTom Holvoet
Tom Holvoet
Finding the shortest path in a road network is a well known problem. Various proven static algorithms such as Dijkstra and A* are extensively evaluated and implemented. When confronted with dynamic costs, such as link travel time predictions, alternative route planning algorithms have to be applied. This paper applies Ant Colony Optimization combined with link travel time predictions to find routes that reduce the time spend by travels by taking into account link travel time predictions. The presented algorithm is evaluated using simulations in real world traffic networks.


1 December 2010
Verified
Conference Paper
Closed access
RCRutger Claes
Rutger Claes
THTom Holvoet
Tom Holvoet
Online simulation of traffic can assist route guidance systems by predicting problems such as congestion. Accurate predictions require accurate status information about vehicles - the fact that the vehicles are distributed over large-scale road infrastructure makes this particularly challenging. Embedding the online simulation in the road infrastructure - by distributing it across road side computing infrastructure - is a partial solution, but also adds additional complexity to the symbiotic relationship between online simulation and the physical system. In this paper we describe an approach that uses delegate MultiAgent Systems to reduce the complexity of such symbiotic relationships. Experimental results in a prototype implementation of the route guidance mechanisms show that the approach is feasible and leads to a proactive route guidance mechanisms with the potential of outperforming current state of the practice non-proactive routing mechanisms.

1 December 2009
Verified
Conference Paper
Closed access
RCRutger Claes
Rutger Claes
THTom Holvoet
Tom Holvoet
Microscopic simulation of traffic, while often necessary to capture the effects of interest, is a computationally expensive simulation strategy. What can be observed, however, is that the accuracy required from the simulation for post simulation analysis can depend on the simulated world and vary over simulated time - roads becoming crowded may be simulated differently than sparsely used roads. In this paper we explore multi-model simulation as an adaptive simulation strategy. Multi-model simulations are capable of selecting and switching to a suitable simulation model at runtime, based on the state of the simulated world. This simulation strategy reduces the computational complexity of traffic microsimulations while still maintaining the desired level of accuracy needed to produce meaningful results. We illustrate the approach via an experimental set-up that allows switching between a road queue model and a fully detailed road simulation model.

1 October 2011
Verified
Conference Paper
Closed access
RCRutger Claes
Rutger Claes
THTom Holvoet
Tom Holvoet
Drivers able to use link traversal time predictions can use this information to choose better routes. Such anticipatory vehicle routing bases routing decisions on link traversal times likely to be experienced by the vehicle instead of historic or real time link traversal times. However, generating such link traversal times is a challenge. This paper describes a multi-agent system in which road agents and vehicle agents cooperate to provide ad hoc link traversal times for the routes the vehicle agents intend to follow. Road agents use observations to learn a model of the road they represent and combine this model with information obtained from vehicle agents to predict future link traversal times. Two such models are described in this paper, namely a polynomial model and an artificial neural network. Both models are evaluated based on a microscopic traffic simulation. The predictions obtained using both models are analyzed and compared with the actual observed link traversal times to estimate the accuracy of the predictions.