Interval type-2 fuzzy-based following control of autonomous vehicles in complex environments

In complex traffic operation environments, uncertainties or interfering factors such as road bumps, external conditions, and sensor detection issues have a non-negligible impact on the safety of autonomous vehicle following behavior. This paper suggests a novel sliding mode control approach based on an interval type-2 fuzzy algorithm, aiming to ensure system safety while improving system stability and control accuracy. Firstly, this method introduces virtual points into the leader-follower model, leveraging the advantages of the “hit-soft-wall” operation mode in train control. This improvement refines the follower system model, balancing both safety and efficiency. Meanwhile, Ackermann’s steering geometry principle is applied to reduce the system’s dimensionality, addressing the dimensionality difference between the control inputs and the controlled variables. Secondly, an improved integral fast nonsingular terminal sliding mode surface (IFNTSMS) is designed, employing a saturation function as the switching function to mitigate the effects of chattering, thereby ensuring smooth driving of autonomous vehicles and enhancing the system’s speed and accuracy. Finally, considering that noise from external environments, sensors, and other factors may interfere with the data of the lead vehicle, intelligent modeling based on an interval type-2 fuzzy system is used to identify noise online and adaptively adjust model parameters. The rationality and effectiveness of the suggested model and control strategy are confirmed by numerical simulations.