*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Po-Nien (Thomas) Lin
(Advisor: Prof. Dimitri Mavris]
will propose a doctoral thesis entitled,
Autonomous Framework for Operational Decision-Making
of An Unmanned Aerial System in Cluttered and GPS Absent Scenarios
On
Friday, April 24 at 2:30 p.m.
Blue Jeans (https://bluejeans.com/356607879)
Abstract
With the advance of human society, many occupations and their working conditions are hazardous to a certain degree for human operators. These operational challenges or constraints can generally be categorized into three major aspects: safety, time, and space. An unmanned aerial system (UAS) has the capability to execute its designated missions or tasks with safety, efficiency, mobility, and flexibility given the support of on-board sensors and computation units. These capabilities allow an to gather and analyze the situation within a short amount of time while having aerial maneuverability and allowing human counterparts to shield away from the hazardous working environment or conditions. As a result, the UAS is the perfect candidate to perform various tasks and operations in extreme weather conditions, geographically challenging locations while protecting people from hazardous jobs. In order to fully develop the capabilities of future UAS applications, it is essential to design an autonomous UAS framework with the capabilities to sense objects in the operating environment, perform designated operations accurately, and make strategic decisions according to the safety factors of the current situation.
UAS-related studies have grown dramatically in the past twenty years; among these research topics, the most important development is to increase the autonomy level of UAS operations. Therefore, three basic requirements to achieve this higher autonomy level for autonomous UAS operations have been concluded: 1) sense-and-avoid, 2) decision-making, and 3) accurate operation. The current challenges faced by modern UAS development have been summarized in this proposal as a series of high-level research questions and need to be answered by an UAS framework seeking to achieve a high level of operation autonomy. As a result, four essential building blocks (or functionalities) are considered, including: target detection/identification, navigation without the support of GPS signals, the ability to configure new operating paths or zones, and making operational decisions.
These four building blocks are used as the evaluation criteria for current UAS frameworks to determine if a framework fulfills the requirements. Since almost none of the current frameworks meet all four criteria, a new framework for advanced UAS autonomous operations has been proposed in this research. By establishing this proposed framework, higher autonomy levels and better operational performances can be achieved through a more complete system composition and lower manufacturing cost without implementing expensive hardware. Therefore, the ultimate goal of this study is to establish an advanced UAS framework with the capabilities to perform accurate operations in cluttered and GPS absent/denied scenarios under the support of decision-making and operation reconfiguration functions.
Committee
