Abstract
Traffic Routing System (TRS) is one of the most important intelligent transport systems which is used to direct vehicles to good routes and reduce congestion on the road network. The performance of TRS mainly depends on a dynamic routing algorithm due to the dynamic nature of traffic on road network. AntNet algorithm is a routing algorithm inspired from the foraging behavior of ants. TAntNet is a family of dynamic routing algorithms that uses a threshold travel time to enhance the performance of AntNet algorithm when applied to traffic road networks. TAntNet-1 and TAntNet-2 adopt different techniques for path update to fast direct to the discovered good route and conserve on this good route. TAntNet-3 has been recently proposed by inspiring the scout behavior of bees to avoid the bad effect of forward ants that take bad routes. This chapter presents a new member in TAntNet family of algorithms called TAntNet-4 that uses two scouts instead of one compared with TAntNet-2. The new algorithm also saves the discovered route of each of the two scouts to use the best of them by the corresponding backward ant. The experimental results ensure the high performance of TAntNet-4 compared with AntNet, other members of TAntNet family.
TopAnt routing algorithms is one of the most promising swarm intelligence (SI) methodologies that are studied in many researches (Di Caro & Dorigo, 1998; Kassabalidis et al., 2002; Kroon & Rothkrantz, 2003; Suson, 2010; Claes, R., & Holvoet 2012; Jabbarpouret al., 2014a; Yousefi & Zamani, 2013 ; Ghazy & Hefny, 2014, Jabbarpour et al., 2014b and Girme, 2015).
AntNet algorithm appears in 1998 by Di Caro & Dorigo (1998) for routing of data communication network. The algorithm with its distributed multi agent characteristics, attracted many researchers to adopt the algorithm in both data communication network and road traffic network.
Key Terms in this Chapter
TAntNet-2Algorithm: A threshold based AntNet algorithm, this algorithm is a modified version of AntNet algorithm, developed essentially for routing on road network, TAntNet uses three types of ants (Forward, Backward and Check ant), this algorithm uses a threshold to detect the discovered good route.
Bee Colony Optimization (BCO): A swarm based meta-heuristic, inspired from the behavior of real bees when searching for nectar, BCO firstly introduced by Karaboga (2005) . BCO uses three types, namely, the employee bee, the onlooker bee, and the scout bee. Each type of bees plays different role in the nectar search process.
Intelligent Route Guiding System: An intelligent transportation system that is used for guide vehicle's drivers to the best route, the system uses a real time data that collected by sensors and routing algorithm to discover best route.
TAntNet-3 Algorithm: A modified version of TAntNet-2 algorithm, TAntNet-3 uses scout behavior, which inspired from the Bee foraging behavior, the algorithm uses two type of threshold one for recognize on the discovered good route and the other to detect the bad route, the algorithm can avoid by this way the bad effect of bad route.
Ant Colony Optimization (ACO): A category of optimization algorithms, inspired from the real ant foraging behavior. ACO is a population-based metaheuristic which used to find approximate solutions to difficult optimization problems like travel sales man problem and dynamic routing problem that work on find optimal or near optimal paths to destinations.
Swarm Intelligence (SI): Swarm intelligence appears in nature with different swarms, like Ants, Bees, flock of birds, and school of fish...etc. swarm intelligence were a source of inspiration for many self-organized systems that used to solve complex problems. Swarm intelligence systems composed of many unsophisticated agents that coordinate using decentralized control.
AntNetAlgorithm: A routing algorithm inspired from the foraging behavior of Ant, and firstly introduces by Di Caro and Dorigo (1998) for routing on data network, after that this algorithm is adopted for dynamic routing on road network.