Accession Number : ADA531658
Title : Human-Automation Collaborative RRT for UAV Mission Path Planning
Descriptive Note : Master’s thesis
Corporate Author : MASSACHUSETTS INST OF TECH CAMBRIDGE DEPT OF ELECTRICAL ENGINEERING AND Information Technology
Personal Author(s) : De Jesus Caves, Americo
Report Date : JUN 2010
Pagination or Media Count : 113
Abstract : Future envisioned Unmanned Aerial Vehicle (UAV) missions is going to be transported in dynamic and sophisticated environments. Human-automation collaboration is going to be needed to be able to distribute the elevated mission workload which will naturally arise from all of these interactions. One sector of great interest during these missions may be the supervision of multiple UAVs with a single operator, and it is advisable to know how individual operators can supervise a small group of vehicles performing semi-autonomous path planning while staying away from no-fly zones and replanning quickly. Regrettably, real-time planning and replanning could be a computationally troublesome task, especially in the high density obstacle environments which are envisioned later on urban applications. Recent work has suggested using a randomized formula referred to as Quickly exploring Random Tree (RRT) formula for path planning. While able to find achievable solutions rapidly, it’s unclear how good an individual operator can supervise a small group of UAVs which are planning according to this type of randomized formula, particularly because of the unpredictable nature from the generated pathways. This thesis is definitely the outcomes of a test that tested an adjustment from the RRT formula to be used in human supervisory charge of UAV missions. The experiment tested how human operators socialized and performed when given techniques used in getting together with an RRT to supervise UAV missions in environments with dynamic obstacle fields of various densities.
The experimental results shown that some variants from the RRT increase subjective workload, but didn’t provide conclusive evidence for whether utilizing an RRT formula for path planning is preferable to manual path planning when it comes to overall mission occasions. Research into the data and behavior observations hint at directions for possible future work.
Descriptors : *MISSIONS, *PLANNING, *AIRCRAFT, *UNMANNED, HUMANS, MISSION PROFILES, DYNAMICS, RANDOM VARIABLES, TEAMS(PERSONNEL), THESES, VARIATIONS, MANUAL OPERATION, VEHICLES, WORKLOAD, Real-time, ENVIRONMENTS, OPERATORS(PERSONNEL), ALGORITHMS, CONTROL
Subject Groups : Pilotless Aircraft
Distribution Statement : APPROVED FOR PUBLIC RELEASE