A Study on Human-like Driving Decision-Making Mechanism in Autonomous Vehicles Under Various Road Scenarios

Behavior Planning for Autonomous Driving: Methodologies, Applications, and Future Orientation

“…BP methodologies can be classified into two main categories classical-based and learning-based methods [14], [17], [18]. They have also classified the BP methods based on four main categories: knowledge-based, heuristic-based, approximate reasoning-based, and human-like methods.…”

“…Another method of classification of BP is a classification approach according to human capabilities. These capabilities are rationality, obeying the rules, cognitive abilities, and learning [18]. The existing methodologies is classified into rule-based, cooperative-based, probabilistic-based, game theoretic-based, and learning-based approaches [19].…”

“…Nevertheless, they did not tackle the appropriateness of the methodologies with the different scenarios and applications. Lastly, BP methodologies are studied by [18]. Although they compared the algorithms from the point of view of the performance, they did not mention or refer to any mathematical equations for the method used to measure the performance's parameters.…”

“…while the second level is low level behaviors including angular and longitude velocities and accelerations. Another framework for decision-making is described by [18]. They stated that the input to the decision-making module is perception and vehicle prediction.…”