Reliable Risk Management for Autonomous Vehicles based on Sequential Bayesian Decision Networks and Dynamic Inter-Vehicular Assessment

Abstract

Guaranteeing the safety of an autonomous vehicle (AV) is a challenging task, especially if the perceived environment is highly uncertain and other road users deviate from their expected trajectories. In this paper, we propose a probabilistic overall strategy for risk assessment and management of AV in highway through a Sequential Level Bayesian Decision Network (SLBDN) and an appropriate analytical formalization of criteria for anomaly detection based on a Dynamic Predicted Inter-Distance Profile (DPIDP) between vehicles. Accordingly, the proposed system is designed to take the suitable maneuver decision, have a safety retrospection and verification over the current maneuver risk and take appropriate evasive action autonomously from moving obstacles. Moreover, this probabilistic framework accounts for measurements uncertainty through an Extended Kalman Filter (EKF) and for vehicles' maximum capacities. Since the proposed strategy has a short response time, integrating safety verification in the decision-making process makes real time evasive decisions possible. Several simulation results show the good performance of the overall proposed control architecture, mainly in terms of efficiency to handle probabilistic decision-making even for risky scenarios.

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

“…To decrease the measurement uncertainties, the Extended Kalman Filter is used. The method is proven to have satisfactory results even in risky situations [29].…”

“…-Only small traffic situations can be handled in real-time on state-ofthe-art computing hardware Probabilistic-based methods [28], [29] -Ability to interact with the behaviors of other dynamic objects on the road.…”

Detection and Risk Analysis with Lane-Changing Decision Algorithms for Autonomous Vehicles

“…In intelligent transportation systems (ITS) [ 11 ], it is usually qualified by the dangerousness of a situation for passengers who may be caused physical injury. For this, different metrics are commonly used: Inter-vehicular time [ 12 ]: Refers to the time separating two successive vehicles in the same traffic lane. The traffic law defines a two-second safety Inter-Vehicular Time (TIV).…”

“…Inter-vehicular time [ 12 ]: Refers to the time separating two successive vehicles in the same traffic lane. The traffic law defines a two-second safety Inter-Vehicular Time (TIV).…”

A design method based on Bayesian decision for routing-based digital microfluidic biochips

“…Next, it was widely applied in the fields of physics, chemistry, biology and other disciplines. 38,39 Currently, its concepts and methods have good development prospects in the fields of engineering technology, 40 management science, 41 system operation research, 42 and medical diagnosis. 43 As a classic decision theory, the Bayesian decision theory not only considers the loss caused by making decisions under various circumstances but also considers the probability of occurrence of various circumstances.…”