Semi-trailers are widely used in highway freight transportation because of their
large payload capacity and high transport efficiency. However, structural
characteristics such as a high center of gravity (CG), heavy loads, and the
dynamic coupling between the tractor and trailer make them prone to yaw
instability and rollover under complex conditions. To solve these problems, this
article proposes a hierarchical stability control architecture for semi-trailers
based on the joint estimation of equivalent parameters. First, a
six-degree-of-freedom (6-DOF) theoretical dynamic model is established. This
model includes the lateral, yaw, and roll motions of both the tractor and
trailer to provide desired reference states. Second, a parameter estimation
method combining a genetic algorithm (GA) with a forgetting-factor recursive
least squares (RLS) algorithm is designed. It dynamically identifies eight
unknown equivalent parameters, specifically the tire cornering stiffness and
suspension damping. Next, a hierarchical controller is developed. The upper
layer uses model predictive control (MPC) to calculate the required additional
yaw moments, while the lower layer allocates these moments through quadratic
programming (QP) based on vehicle steering characteristics. Co-simulation
results, evaluated using error metrics that compare control outputs directly
against TruckSim reference outputs, show that the fusion GA-RLS method offers
better accuracy and adaptability than a standalone GA. Furthermore, the
stability controller prevents rollover in high-speed maneuvers and reduces peak
state indicators by over 41.2% in low-speed scenarios. Robustness tests also
confirm its effectiveness under low-adhesion road conditions and heavy payloads.
Compared with a conventional fixed-parameter MPC, the proposed adaptive
architecture improves key stability indicators by 19% to 25%, effectively
enhancing the dynamic safety of semi-trailers.
