Ibrahim Volkan Isler

Research Overview

The ultimate goal of my research is to develop algorithms for robotic sensor networks to perform sensing tasks that take place in complex, dynamic environments in a fully autonomous fashion. Toward this goal, my students and I are developing algorithms with provable performance guarantees for solving sensing automation tasks in such a way that sensing, actuation and communication issues are addressed simultaneously. In the long run, we are planning to demonstrate the utility of these algorithms with real-life deployments. At the moment, we are building a robotic sensor network testbed so as to validate our algorithms experimentally as well.

Currently, our  focus is on the following research problems:

Pursuit-Evasion Games

In a pursuit-evasion game, one or more pursuers try to capture an evader who, in turn, tries to avoid capture. Currently, we are studying the role of information available to the players on the outcome of the game.
For more information, please visit this page
This work is supported by NSF CCF-0634823.
New!! Check out Garcias playing pursuit-evasion in our lab [.avi].

Sensor Placement and Management

In accomplishing a sensing task, the network should decide on who goes where (motion planning), who senses what (sensor planning), and who communicates with whom (topology management). At the moment, we are studying these problems in the context of a network composed of robots equipped with cameras. The ultimate goal is to design planning algorithms which solve these problems simultaneously and in a distributed fashion.
This work is supported in part by NSF CNS-0707939 and NSF IIS-0745537.

Check out recent results on placement and selection.

Human-Robot Interaction

Many applications require robots to share the same workspace with humans. In such settings, it is important that the robots perform in a ``human-friendly'' way. The primary challenge here is to model what human-friendly means.  In our current work, we are investigating ways of incorporating biometric data from the human into the design of a robotic algorithm in such a way that the robot maintains the human's state (e.g. stress levels) at desirable levels while performing the robotic task.
For more information, you can either visit this page or read this paper.

More information about my  research is available through the RSN lab web page (currently under construction).