Aerospace Control and Guidance Systems Committee

Announcements

You must first log in to access prior meeting presentations, register for a meeting, or nominate some for the Ward Award.

If you do not have a login account, or cannot remember the email address associated with your account, please click on the Application Form link below.

Login

Site Search

Search our site:

Upcoming Events

Photos

Meeting Highlights New!

Subcommittee S

Prior Meetings

Abstracts may be viewed by anyone. Presentations are only available to active members who have logged in.

Meeting 132
(coming soon)

Home

Ward Memorial Award

Planning Advisory Board

Downloads

Constitution and By-Laws

About

History

Contact Us

← Return to agenda

Meeting	ACGS Committee Meeting 121 - Tucson, AZ - April 2018
Agenda Location	7 SUBCOMMITTEE E – Flight, Propulsion, and Autonomous Vehicle Control Systems 7.3 Flight Testing of Intelligent Motion Video Guidance for Unmanned Air System Ground Target Surveillance
Title	Flight Testing of Intelligent Motion Video Guidance for Unmanned Air System Ground Target Surveillance
Presenter	John Valasek
Affiliation	Texas A&M University
Available Downloads*	presentation
	Downloads are available to members who are logged in and either Active* or attended this meeting.
Abstract	Unmanned Aircraft Systems (UAS) are gaining increased use for a variety of defense and civilian roles, and are predicted to participate more as effective actors in future surveillance and reconnaissance work. To efficiently and accurately collect intelligence data many current UAS require supervision from a team of two to four operators. Operating a Small UAS with a non-gimbaled or fixed camera increases operator workload since the entire vehicle must be steered to visually track targets. This presentation details the implementation and flight testing of a machine learning algorithm for the autonomous tracking of ground targets by UAS with a non-gimbaled or fixed camera that offers the potential to reduce operator workload and the number of operators. The Reinforcement Learning agent uses the Q-Learning algorithm and learns a control policy to keep a target within the camera image frame without user intervention. The Reinforcement Learning agent is trained offline in a simulation environment and learns different control policies to successfully track targets based upon target trajectory types, and crosswinds. The control law is implemented entirely onboard the vehicle rather than the ground control station, and functions as an outer-loop controller that commands the autopilot. Performance of the system is demonstrated with flight results for stationary and randomly moving targets, in addition to tracking randomly moving targets in unstructured environments.