Autonomous obstacle avoidance and decision-making algorithms for intelligent connected vehicles in a complicated transportation environment are important studies on intelligent driving. However, it is difficult to adapt to a more complicated traffic environment based on safety distance and conventional potential field. Therefore, in this paper, a driving risk field model based on field theory is proposed involving transportation factors and vehicle conditions. A hidden Markov model was used to evaluate and determine the motion state of surrounding vehicles. A safe, feasible, and smooth collision-free path was planned by calculating the magnitude of the potential field forces on the longitudinal and lateral sides of the obstacle vehicles. The results showed that the method can effectively select a suitable path for obstacle avoidance in complex road conditions while satisfying safety and traffic laws.
This paper develops a novel integrated collision avoidance strategy for autonomous vehicles in an emergency based on steering and braking. Specifically, the framework of the collision avoidance strategy is composed of two parts: an up-level decision-making layer and a low-level controller layer. The purpose of the up-level is to select the appropriate control strategy based on the vehicle information, and the low-level is to drive the vehicle according to the instructions generated by the up-level. More concretely, a novel control strategy is proposed by integrating four-wheel steering, active rear steering, and differential braking with guaranteed path-tracking accuracy and driving stability by adaptive model predictive control (AMPC). Finally, extensive co-simulations in MATLAB/Simulink and CarSim are conducted to verify the effectiveness of the proposed collision avoidance strategy in terms of tracking error, yaw rate, and roll angle.
Active obstacle avoidance control strategy is the main question of intelligent vehicles, and active four-wheel steering is gradually used in the intelligent control system. Considering both tracking performance and driving characters, an active controller with four-wheel steering (4WS) and active rear steering (ARS) based on adaptive model predictive theory (AMPC) is designed. The control architecture is composed a supervisor and an AMPC controller. The supervisor is used to select the appropriate control mode and the AMPC is used to calculate the expected steering angle when the stability indexes over the safety threshold. Finally, the proposed control strategy is simulated via Carsim-Simulink co-simulation. The results show that the integrated controller can track the obstacle avoidance path and has good driving performance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.