Bio-inspired Multi-Agent Collaboration for Urban Monitoring Applications

Abstract

Vehicular sensor networks (VSNs) provide a collaborative sensing environment where mobile vehicles equipped with sensors of different nature (from chemical detectors to still/video cameras) inter-work to implement monitoring applications such as traffic reporting, environment monitoring, and distributed surveillance. In particular, there is an increasing interest in proactive urban monitoring where vehicles continuously sense events from streets, autonomously process sensed data (e.g., recognizing license plates), and possibly route messages to vehicles in their vicinity to achieve a common goal (e.g., to permit police agents to track the movements of specified cars). MobEyes is a middleware solution to support VSN-based proactive urban monitoring applications, where the agents (e.g., police cars) harvest metadata from regular VSN-enabled vehicles. Since multiple agents collaborate in a typical urban sensing operation, it is critical to design a mechanism to effectively coordinate their operations to the area where new information is rich in a completely decentralized and lightweight way. We present a novel agent coordination algorithm for urban sensing environments that has been designed based on biological inspirations such as foraging, stigmergy, and Lévy flight. The reported simulation results show that the proposed algorithm enables the agents to move to “information patches” where new information concentration is high, and yet limits duplication of work due to simultaneous presence of agents in the same region.