Optimal Control of Connected Automated Vehicles with Event/Self-Triggered Control Barrier Functions

TL;DR

This paper introduces event-triggered and self-triggered control barrier function approaches for connected automated vehicles, ensuring safety constraints are always feasible and reducing communication needs.

Contribution

It develops novel event/self-triggered schemes that guarantee feasibility of control problems and eliminate the need for fixed sampling intervals in CAV safety control.

Findings

Both approaches significantly reduce infeasibility cases.

Communication among CAVs is decreased without performance loss.

Simulation results validate the effectiveness of the methods.

Abstract

We address the problem of controlling Connected and Automated Vehicles (CAVs) in conflict areas of a traffic network subject to hard safety constraints. It has been shown that such problems can be solved through a combination of tractable optimal control problem formulations and the use of Control Barrier Functions (CBFs) that guarantee the satisfaction of all constraints. These solutions can be reduced to a sequence of Quadratic Programs (QPs) which are efficiently solved on-line over discrete time steps. However, the feasibility of each such QP cannot be guaranteed over every time step. To overcome this limitation, we develop both an event-triggered approach and a self-triggered approach such that the next QP is triggered by properly defined events. We show that both approaches, each in a different way, eliminate infeasible cases due to time-driven inter-sampling effects, thus also eliminating the need for selecting the size of time steps. Simulation examples are included to compare the two new schemes and to illustrate how overall infeasibilities can be significantly reduced while at the same time reducing the need for communication among CAVs without compromising performance.