Lateral Control of Holonomic Platoons via Spatial Vehicle-to-Vehicle Learning

This paper proposes a lateral controller for holonomic vehicle platoons that integrates both feedforward and feedback control strategies. A spatial axis is defined to assess the lateral tracking performance across the platoon characterized by a novel definition of lateral string stability, and a holonomic vehicle model is developed. An unconstrained model predictive controller is then employed for trajectory tracking, and a novel feedforward vehicle-to-vehicle (V2V) learning-based controller is further incorporated to enhance transient tracking performance through V2V interaction. In this feedforward-feedback controller, a sufficient condition for lateral string stability is derived through contraction mapping analysis, and its practical implementation is validated via simulations involving omnidirectional wheeled vehicles.