Applying evolutionary methods for early prediction of sleep onset. Issue 5 (July 2016)

• Record Type: Journal Article
• Authors: Malcangi, Mario
• Is Part Of: Neural computing & applications. Volume 27:Issue 5(2016)
• Page Start: 1165
• Page End: 1173
• Publication Date: 2016-07
• Subjects: Sleep onset -- Fuzzy logic -- Behavior sensing -- Physiologic sensing -- EFuNN; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-1928-6
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract Driving safety can be achieved by predicting imminent falling asleep at the wheel. Several methods of early detection have been investigated by continuous monitoring of physiological and behavioral parameters. Requirements for noninvasive, unattended, personal adaptation need to be met, along with the effectiveness of the detection method, in order to perform reliably when applied. Because wakefulness and sleep are reflected in several human physiological conditions, such as cardiac activity, breathing, movement, and galvanic skin conductance, captured bioelectric signal features were extracted. A fuzzy decision-fusion logic was tuned to make inferences about oncoming driver fatigue and drowsiness. The evolving fuzzy neural network paradigm was applied to the previous developed framework to improve reliability while keeping target system complexity low.