Enhanced Genetic Path Planning for Autonomous Flight

Guest speaker Dr. H. David Mathias will give a research talk on flight planning for drones: "Enhanced Genetic Path Planning for Autonomous Flight."

He is an assistant professor of computer science at Florida Southern College.

Description: "With the relatively recent explosion in the availability and popularity of affordable robotic vehicles (including micro aerial vehicles, or drones), there is renewed interest in applications of path planning. Due to the computational expense of optimal solutions, approximation techniques, including genetic algorithms, are often applied. 

"In this talk, we will examine path planning for micro aerial vehicles (MAVs) in a known, static but otherwise realistic environment. Unlike many path planning algorithms in the literature, the environment we consider is continuous and direction of travel is unconstrained. This is necessary to allow the algorithm to run on a physical MAV using GPS rather than artificial coordinates. 

"Our focus is on improving rate of convergence through the use of domainspecific operators and a new optimization objective. Individually, and in concert, these new features significantly reduce the number of generations required to find an acceptable solution."