Safe consensus control for stochastic networked systems

doi:10.1016/j.automatica.2025.112677

Safe consensus control for stochastic networked systems

2025-11-12

0

PRE

AI

DOI：10.1016/j.automatica.2025.112677

Original

Original request for help

Save

Abstract

En 中文

A key challenge in stochastic networked systems (SNSs) is providing probabilistic guarantees for both consensus and safety through designing a safe consensus controller. This paper presents a novel safe consensus control approach to address this issue. For controller design of SNSs, we establish a safe consensus criterion using an exponential control barrier function (ECBF), achieving reach-avoid-stay thereby ensuring safe consensus in probability. By this criterion, we provide an exact lower bound probability of achieving safe consensus. Based on the established ECBF constraints, we formulate the safe consensus control problem as a convex semi-definite programming (SDP) problem, where an implicit controller can be efficiently computed online within an algorithmic framework. Even if the widely used online quadratic programming (QP) framework fails to compute a controller for general SNSs (where controller induces noise), our proposed SDP algorithm remains applicable. Additionally, we present an anti-consensus criterion for collision avoidance between nodes by an inverse exponential control barrier function (IECBF). We provide a probabilistic guarantee for anti-consensus and formulate its resulting problem as an SDP to compute the controller. Finally, two numerical examples demonstrate the effectiveness of the proposed approach, where Monte Carlo simulations validate the repeatability of the proposed approach under varying initial conditions and stochastic noise levels.

Keywords:

Stochastic networked systems

Safe consensus control

Control barrier functions

Semi-definite programming

Safe consensus control for stochastic networked systems

Abstract

Journal

Organization