Lanlan Su

Yongkang Su, Sei Zhen Khong, Lanlan Su

This paper investigates output consensus in heterogeneous dynamical networks within a plug-and-play framework. The networks are interconnected through nonlinear diffusive couplings and operate in the presence of measurement and communication noise. Focusing on systems that are input feedforward passive (IFP), we propose a passivity-compensation approach that exploits the surplus passivity of coupling links to locally offset shortages of passivity at the nodes. This mechanism enables subnetworks to be interconnected without requiring global reanalysis, thereby preserving modularity. Specifically, we derive locally verifiable interface conditions, expressed in terms of passivity indices and coupling gains, to guarantee that consensus properties of individual subnetworks are preserved when forming larger networks.

Yongkang Su, Joaquin Carrasco, Iñaki Esnaola et al.

This letter investigates the problem of output synchronisation in heterogeneous dynamical networks with nonlinear diffusive couplings in the presence of disturbances on the coupling links. By exploiting relative dissipativity properties between adjacent agents, distributed conditions are established to guarantee output synchronisation. Specifically, these conditions can be verified using only local information associated with neighbouring agents and coupling links. As an illustration, a heterogeneous network of Goodwin oscillators is considered, where the relative dissipativity properties between neighbouring oscillators are characterised and used to analyse synchronisation.

Amornyos Horprasert, Esa Apriaskar, Xingyu Liu et al.

One of the key challenges that Reinforcement Learning (RL) faces is its limited capability to adapt to a change of data distribution caused by uncertainties. This challenge arises especially in RL systems using deep neural networks as decision makers or policies, which are prone to overfitting after prolonged training on fixed environments. To address this challenge, this paper proposes Gaussian Process Diffusion Policy (GPDP), a new algorithm that integrates diffusion models and Gaussian Process Regression (GPR) to represent the policy. GPR guides diffusion models to generate actions that maximize learned Q-function, resembling the policy improvement in RL. Furthermore, the kernel-based nature of GPR enhances the policy's exploration efficiency under distribution shifts at test time, increasing the chance of discovering new behaviors and mitigating overfitting. Simulation results on the Walker2d benchmark show that our approach outperforms state-of-the-art algorithms under distribution shift condition by achieving around 67.74% to 123.18% improvement in the RL's objective function while maintaining comparable performance under normal conditions.