Observer-based control for vehicle platoons under uncertainties and measurement noise

The robust platoon control problem is investigated for connected and automated vehicles (CAVs) subject to parametric uncertainties, exogenous disturbances, and measurement noise. To address these combined challenges, the nonlinear longitudinal dynamics is first reformulated into an exact linear representation using the inverse dynamics control approach, while the parametric uncertainties and external disturbances are jointly modeled as a lumped disturbance for unified estimation. A unified observer structure is then developed by integrating a generalized extended state observer (GESO) with a Kalman filter (KF). By combining the estimation capability of the GESO and the noise-attenuation property of the KF, the proposed scheme provides complementary advantages over existing observer-based methods. Furthermore, a composite control law is contrived by incorporating a feedforward compensation term based on the lumped disturbance estimation and a state-estimate-based feedback control component. Rigorous analysis is conducted to attest that the resulting platoon system satisfies weak input-to-state string stability under parametric uncertainties, external disturbances, and measurement noise. Finally, simulation studies are given to evaluate the effectiveness and superior robustness of the contrived control method. This work contributes an integrated observer–controller framework that unifies lumped disturbance estimation, noise-resilient state observation, and composite feedforward–feedback control, thereby offering a promising solution for robust and string-stable CAV platoon control in real-world environments.