Paolo Gaudiano

Evolving behaviors for a swarm of unmanned air vehicles

We have previously reported on a project involving the control of a swarm of unmanned air vehicles (UAVs) carrying out search or search-and-destroy missions. We developed and tested (in simulation) a number of strategies for swarm control, and proposed systematic evaluation techniques and performance metrics. In this paper we report some additional results in which we evolved some of the swarm control parameters using a genetic algorithm (GA). While the improvements were modest, the results show how evolutionary computing algorithms can be used to facilitate the design of swarm control algorithms.

Control of UAV Swarms: What the Bugs Can Teach Us

This article describes research in the control of UAVs. Control strategies exist for individual UAVs, but not for large, coordinated teams. While there is a general agreement that a cooperative, team or swarm approach to UAV control can be beneficial for a number of military and civilian applications, little work has been done to understand if and when it even makes sense to use more than a single UAV, or a few, individually controlled UAVs. We have begun a study of UAV control strategies that are based on swarm intelligence, a methodology that draws its inspiration from the behavior of social insects. Our long-term goal is to gain a systematic understanding of distributed control strategies, and a quantitative methodology to evaluate performance of UAV swarms under a variety of conditions. In this paper we outline our approach, describe an agent-based model we have used to develop and test our methodology, and present some of our results to date.

Resource allocation for a distributed sensor network

In this paper we describe a project undertaken for the Office of Force Transformation (OFT) to investigate alternative resource allocation strategies for Americas armed forces. In particular, OFT is interested in understanding how resource allocation strategies can be used in the context of distributed, network-centric units. To address this problem we have developed a simulation tool using agent-based modeling to explore the emergent properties of a distributed sensor network. We focus on the task of using distributed sensors with varying characteristics and capabilities trying to detect and track the movement of enemy units in an urban environment. The goal of the project is to identify the impact of different resource allocation strategies on the performance of the sensor network.