InterJournal Complex Systems, 1119
Status: Accepted
Manuscript Number: [1119]
Submission Date: 2004
Cartography application for autonomous sensory agents
Author(s): Sarjoun Doumit ,Ali Minai

Subject(s): CX.18



This paper proposes a cartographic application for the rapid and exact identification of topological characteristics of a deployment area by a group of mobile sensory agents. It is assumed that the deployment area’s features are completely unknown and that the deployed agents are capable of limited computational and processing power and to communicate wirelessly. A region’s topography reveals information about its elevations, surface types and gives an idea about its prevailing environments. This information amassed by those ‘scout’ agents is crucial for follow-up exploratory missions that use this information to maximize the energy consumption and prevent redundancy and the loss or damage of agents as with the case of some recent Mars probes. Every agent needs to maintain a representation of the world around it in a useful context that can be used in conjunction with other agents’ maps. The cartographic application is autonomously executed based on sensory input from the agents. The agents are able to summarize their collective information into an optimal structure like a “trail” that allows consequent agents to maximize their coverage of the area. A fundamental principle in our system is that the agents must spread out efficiently in a diffusion-inspired mobility pattern that is determined by the environment itself. The knowledge amassed in this initial operation can be re-used for better topographic information on the area for consequent or explorations of the same or nearby areas. The advantage of this technique is that it can be used in tandem with other functionalities like projecting the timing of occurrences of events or their possible locations simply by knowing the terrain’s topology and discovering the relationship between them.

Retrieve Manuscript
Submit referee report/comment

Public Comments: