Categorized review of drive simulators and driver behavior analysis focusing on ACT-R architecture in autonomous vehicles

Driving a vehicle in a safe manner is a highly specialized task that depends on the driver’s cognitive ability, skills, attention, fast processing, interpretation of traffic situations and rules, and above all, sound decision making. By driver, we imply either a human or an intelligent agent. In parallel to numerous yet fragmented studies that have addressed the individual effect of the above factors, there has been a continuous quest towards a unifying perspective to articulate the driving task within the framework of a cognitive architecture. Adaptive Control of the Thought—Rational (ACT-R) as a theory of human cognition is among the most promising and complete cognitive architectures that has been widely used to model human behavior and decision making. It is sensible to suggest that it can realistically be employed to model human behavior in driving, and by the same token, it can be configured as architecture for an autonomous driving system. In this paper, we present a directed review of drive simulators, and cognitive architectures, notably ACT-R architecture. Resources including but not limited to Google, Google Scholar, Crossref, Science Direct, and Mendeley are examined, compiled, and presented in this document. The results are classified into seven categories: (1) simulation environment; (2) review of ongoing drive simulators; (3) driving simulation data analysis; (4) driving performance analysis; (5) driving simulation validation; (6) driver behavior analysis; and (7) driving simulation experiments. The review could be beneficial to researchers and stakeholders in Advanced Driver Assistance Systems (ADAS) and Driverless cars.