Robotic Sensor Networks The major goal of the research is to develop, test and characterize algorithms for scalable, application-driven, wireless network services using a heterogeneous collection of communicating mobile nodes. Some of these nodes are autonomous robots in that their movements are not human controlled. The others are portable thus making them dependent on humans for transportation. We emphasize that most (though not all) of the mobile nodes have modest sensing, computational and communication resources. The collection as a whole is an example of a wireless sensor network.

	AVATAR: Autonomous Vehicle Aerial Tracking and Reconnaissance Initially, the focus of our research was in creating a reliable control mechanism for a model helicopter. Once that had been achieved we focused on performing higher lever tasks with the helicopter. Besides stable autonomous flight, today we are able to perform tasks such as GPS waypoint navigation, autonomous vision-based landing and autonomous sensor deployment. We are currently researching areas such as autonomous landing on a moving target, deployment on a moving target, stealthy target pursuit and vision-based obstacle avoidance in 3D.

	Multi-Robot Coordination There are many robotics application areas that can benefit from the use of multi-robot systems, i.g., environment monitoring, autonomous construction, search and rescue, etc. The main focus of our research is to develop an online, de-centralized, coordinated motion strategy for a group of robots so that they achieve a given task as a team. Our interest is expanded to multi-robot problems that demand a tightly-coupled coordination, i.g., formation assembly and maintenance. Our approach uses a distributed behavior-based strategy whereby we dynamically grow formations from single robots into complex geometries.

	Estimation: Mapping and Localization Robotic mapping and localization have been highly active research areas in robotics and AI for at least two decades. Robotic mapping addresses the problem of acquiring spatial models of physical environments through mobile robots. Robotic localization addresses the problem of estimating a robot's location based on sensor data. Those problems are generally regarded as some of the most important problems in the pursuit of building truly autonomous mobile robots.

	Planning and Control Most mobile robots have simple dynamics and are limited to traveling on planar surfaces. Two of our research thrusts are aimed at more sophisticated locomotion modes. Trajectory planning over uneven terrain is a useful capability for outdoor robotics. Dynamically stable legged robots can overcome obstacles that are impossible for standard wheeled or legged robots to negotiate.