We propose a new approach for navigating people in a ubiquitous computing environment by using digital pheromone trails, similar to ants being led by pheromones to a food source. Unlike ants, humans can use their intelligence in selecting routes. Our idea is to compile such intelligence by accumulating the history of people's rational behaviors and leaving this history as digital pheromones in the environment for later use. In simulations of navigation services, we found that the original ant colony optimization (ACO), which is a metaheuristic based on the foraging activity of ants, does not completely fit our purpose. Therefore, two modifications were made to the original ACO. Our simulation results show that people can be successfully navigated by simulated services implemented using these modified ACOs.
展开▼
