This paper focuses on the problem of orchestrating the movements of mobile autonomous agents and proposes an approach inspired by the robustness and emergent behavior seen in swarms combined with stochastic combinatorial optimization techniques. The main characteristics of this approach are positive/ negative feedback, decentralized computation, and context awareness. Our idea is to have the movements of agents driven by locally perceived the ant-like pheromone, generated by the agents themselves and propagated via some embedded infrastructure. A globally coordinated and adaptive behavior in the agent?s movements can then emerge due to the interrelated effects of agents affected by the pheromone. We apply the proposed approach to the representative scenarios and also discuss parameter selection. Finally we extent this approach to broader uses in pervasive/ubiquitous computing fields.