SUB-OPTIMAL ROUTE PLANNING BY IMMUNO-AGENTS

Vehicular traffic on roads has greatly contributed to the development of the national economy and has played the critical role in modern society. On the other hand, traffic accidents have increased due to the greater number of vehicles. Over 10,000 lives have been lost every year in Japan. According to one estimate, traffic congestion in urban areas has caused economic losses of about 12 trillion yen and time losses of about 5.6 billion person-hours every year. Furthermore, heavy traffic situations cause to occur such problems as environmental deterioration along roads, disharmony with the global environment, wasting energy, etc. We strongly need to solve these problems as a complex system as soon as possible. On the other hand, a paradigm to solve by soft computing such complex systems as a society system and an economy system is being constructed. Soft computing contains fuzzy logic, neurocomputing, genetic algorithm, and chaos theory, and is a way to use them in combination complementarily. It is a natural tendency to use flexible soft computing for the solution and control of the behavior of complex systems. Authors consider that "bio-direction" contributes a great deal to soft computing methodology as its element technology. Immune information processing based on immunology, above all, has just made a start and its future development is expected. As shown in the example of solving the dynamic obstacle evasion problem of multiple AGV (Automated Guided Vehicle)s as an immune information processing mechanism, the decision of priority among plural element behaviors is considered as a reactive planning system, and the non-network type immune information processing mechanism is used for an adaptive algorithm of the system. This paper proposes a sub-optimal route planning mechanism of vehicles in urban areas using the non-network type immune system. This system announces a sub-optimal route to drivers in real time using VICS (Vehicle Information and Communication Systems).