Emmett Lalish Department of Aeronautics & Astronautics University of Washington PhD Student Controls emmettl@u.washington.edu Resume

Biography Research Publications Activities Nonlinear Dynamics & Control Lab

Research My research focuses on tactical control algorithms for Unmanned Aerial Vehicles (UAVs). Tactical means higher level than inner-loop stabilization and lower level than strategic mission planning. A common vehicle model for these applications is known as the unicycle model, which is basically a point that moves forward and turns. As such, it is a reasonable approximation for many vehicles, including airplanes, ships and cars. I create control algorithms for the unicycle model that follow a particular mission goal while taking into account the speed limitations of an aircraft as well as control saturation, such as maximum turn rate. My focus is on nonlinear feedback control as opposed to linear systems theory or optimization. So far the missions that I am investigating fall primarily into two categories: Collision Avoidance and Target Tracking. Coordination of multi-agent systems is a focus of the Nonlinear Dynamics & Control Lab, so my work also ties into that. Collision Avoidance is inherently multi-agent, but I have also studied tracking a target with multiple agents. Collision Avoidance (or Deconfliction) This is my current research and what I hope to do my Dissertation on. I am working to find a feedback method which can guarantee safety of all vehicles while taking into account the directions they would like to go and the actuation limitations present. A big part of the motivation for this work is to integrate UAVs into the airtraffic control system, and perhaps also automate more of the existing airtraffic control system (i.e. Freeflight). Additionally, this research could be used as collision avoidance for ships or teams of mobile robots. Target Tracking and Coordination My earlier research focused on target tracking, which is basically making one vehicle stay near another. For my research, one application could be to keep one or more UAVs near a convoy to act as a wireless uplink, to perform reconnaissance on the convoy, or to patrol the area around the convoy. In the case of UAVs, this can be difficult because the target may move slower than the airplane can. I solved this problem using an oscillatory control law, which my Master's thesis was written on.