Control improved for vehicle platoons with better stability guarantees
Residual Learning-Based Control of Vehicle Platoons with $\ell_2$ Stability Guarantees via Recurrent Equilibrium Networks
Robotics
Summary
Keeping cars lined up safely on the road can be tricky when each car acts differently and faces unexpected events. The authors improved a basic controller by adding a special kind of neural network that learns from past driving data to fix errors. This neural network is designed to keep the whole line of cars stable and reduce spacing mistakes. Their tests showed this method works better than the original controller. This approach helps vehicles follow each other more closely and smoothly.
What this means in practice
- •For automotive control engineers: Enhance vehicle platoon controllers to reduce errors and ensure stable car following despite disturbances.
- •For industrial automation teams: Improve coordinated multi-robot or vehicle systems where precise, robust spacing and velocity control is required.
Authors
Brian Delgado, Anh-Tu Nguyen, Hamid Taghavifar
Abstract
This paper proposes a residual learning-based control framework for heterogeneous vehicle platoons subject to parametric uncertainty and external disturbances. A nominal controller designed via Linear Matrix Inequalities (LMIs), along with disturbance-observer compensation, is enhanced by a Recurrent Equilibrium Network (REN) trained offline using stored trajectories and nominal-model prediction errors. The REN is constrained to satisfy a prescribed $\ell_2$-gain bound, enabling sufficient small-gain conditions for local closed-loop stability and disturbance string stability. Experiments demonstrate reduced spacing and velocity errors relative to the nominal controller.