A System Development for Enhancement of Human-Machine Teaming
Nowadays, complex machines are commonly used as a means for the completion of challenging tasks. These machines involve sophisticated operations and require well trained human operators. Sometimes, operating such complex machines, even for a short period of time, causes operator fatigue and cognitive overload. Human-Machine Teaming concept is defined not only to include the working together of human operators and machines to achieve a common goal but also the human operator's and machine's awareness of operational uncertainties, human operator performance degradation, and/or machine failures with respect to the other member(s) of the team and the ability to compensate for them with respect to mission goal accomplishment. Traditionally, this information exchange between human operator and machine is performed by a Human-Machine Interface. Human-Machine Teaming extends the interface beyond the current Human-Machine Interface methodology, where the information flow is one-directional, to a bi-directional integrated information flow methodology. In this dissertation, an architecture for implementation of Intelligent Human-Machine Interface was introduced. The Intelligent Human-Machine Interface was then designed, developed, and implemented to, first, account for the human operator state; second, use this knowledge to present a measure to enhance humane operators' decision-making process. The state of a human operator was interpolated from human operator's multi-task performance measure and human operator's physiological measurements in a simulated operation of a machine. The simulated machine involved development of a Multi-Attribute Task Battery in MATLAB by Tennessee State University named TSU-MATB. The development of TSU-MATB was based on the Multi-Attribute Task Battery developed by NASA with the added functionalities of decision-making processes for Human-Machine teaming. Additionally, the Inner Beat Interval (IBI) for the human operator heart beat along with his/her eye Blink Rate (BR) were used as physiological measurements. The IBI and BR measurements were provided by the Pulse Sensor Amped and the NeuroSky MindWave Mobile+ respectively. Furthermore, the proposed Intelligent Human-Machine Interface architecture elaborates beyond the current Human-Machine Interface architecture. The Intelligent Human-Machine Interface architecture consisted of integration of three components oppose to two components as traditional Human-Machine Interface architecture. Those components are: 1) Human to Machine Transformation (HMT), 2) Machine to Human Transformation (MHT), and 3) Multi-Modular Sensor Fusion and Decision-Making (MMSF-DM). The MMSF-DM incorporated the use of machine learning and artificial neural networks (ANN) techniques to determine the state of the simulated machine tasks based on the human performance error with respect to a task and his/her physiological measures. The MMSF-DM was based on the concept presented in the Conscious Architecture for State Exploitation (CASE). The Human to Machine Transformation developed to autonomously augment the human operator's commands to the assigned task based on the task state which has been determined by the MMSF-DM module. Similarly, based on the task state, the Machine to Human Transformation developed to determine how information is presented to the human operator. The proposed Intelligent Human-Machine Interface architecture was show performed correctly and the met the allocated requirements.
Computer Engineering|Artificial intelligence|Computer science
Mohammed I Thanoon,
"A System Development for Enhancement of Human-Machine Teaming"
ETD Collection for Tennessee State University.