This article addresses the problem of distributed state estimation (DSE) for discrete-time interconnected systems, where the observed system is composed of subsystems interconnected through state-to-state and state-to-output couplings. Inspired by the leader-follower consensus method, we propose a distributed observer that enables each subsystem to estimate the entire state of the interconnected system. Under certain structural assumptions, we derive necessary and sufficient conditions for the stability of the estimation error dynamics. We further present a decentralized design of the proposed observer, where the operation and construction of the observer can be completed by each subsystem using its locally available information, including the system's basic configuration, local measurements, and data exchanged with neighboring subsystems. In addition, we demonstrate that our distributed estimation framework can be applied to solve the distributed estimation problem for linear time-invariant (LTI) systems with fixed composition by employing an observability decomposition method. Finally, we illustrate the effectiveness of our scheme by applying it to vehicle platooning.

Distributed cooperative tracking control has emerged as a pivotal research focus in multi-agent systems, particularly for platoon control applications where its decentralized architecture offers significant advantages over centralized approaches. However, the direct exchange of sensitive data between agents raises critical privacy risks, hindering its broader adoption across safety-critical applications. This paper presents a privacy-preserving cooperative tracking framework that rigorously maintains bounded coupling errors, which is a crucial requirement for collision avoidance in vehicular platoons. Departing from conventional methods that compromise privacy through explicit state sharing for error mitigation, our proposed algorithm achieves dual objectives: maintaining prescribed error constraints while preserving agent state confidentiality in directed communication networks with time-varying interaction weights. We establish sufficient conditions for achieving cooperative-tracking consensus with predefined error constraints and characterize the quantitative relationship between the asymptotic convergence rate and control gain parameters. Furthermore, we analyse the privacy-preserving performance against internal and external adversaries, demonstrating that the probability of an adversary inferring states within a finite neighborhood of ground-truth values can be rendered arbitrarily small, even while adversaries retain access to identical communication data streams. This extends classical initial-state privacy to the entire operational timeline under time-varying directed topologies. Numerical examples including an application of cooperative adaptive cruise control demonstrate our proposed algorithm’s efficacy.

Although distributed renewable energy sources (DRESs) provide a sustainable solution to future microgrids (MGs), their fluctuant power outputs can incur frequency instability. The work studies the load frequency control (LFC) for MGs with the integration of wind energy under a hierarchical architecture. At the DRES level, a model predictive control method is employed together with an intensified event-triggered scheme considering multiple historic released signals to improve the computation efficiency. At the MG level, robustness specification is addressed in mean-square asymptotic stability to relieve the fluctuations caused by wind power penetration. Furthermore, the phasor measurement unit (PMU) failure and intermittent transmissions are considered in the control design, leading to the resilient control policy. Besides, this article extends the applicability of conventional small-signal LFC model by adding an uncertain matrix to tolerant the parameter variation due to the shift of the steady-state operating point caused by wind energy integration. The closed-loop performance based on the deployed resilient LFC strategy is verified through hardware-in-the-loop experiments, by which the frequency regulations against PMU failures and intermittent communication at different levels are effectively exhibited.

Neural Radiance Fields (NeRF) have revolutionized 3D computer vision and graphics, facilitating novel view synthesis and influencing sectors like extended reality and e-commerce. However, NeRF's dependence on extensive data collection, including sensitive scene image data, introduces significant privacy risks when users upload this data for model training. To address this concern, we first propose SplitNeRF, a training framework that incorporates split learning (SL) techniques to enable privacy-preserving collaborative model training between clients and servers without sharing local data. Despite its benefits, we identify vulnerabilities in SplitNeRF by developing two attack methods, Surrogate Model Attack and Scene-aided Surrogate Model Attack, which exploit the shared gradient data and a few leaked scene images to reconstruct private scene information. To counter these threats, we introduce S^2NeRF, secure SplitNeRF that integrates effective defense mechanisms. By introducing decaying noise related to the gradient norm into the shared gradient information, S^2NeRF preserves privacy while maintaining a high utility of the NeRF model. Our extensive evaluations across multiple datasets demonstrate the effectiveness of S^2NeRF against privacy breaches, confirming its viability for secure NeRF training in sensitive applications.

In recent years, over-the-air aggregation has been widely considered in large-scale distributed learning, optimization, and sensing. In this paper, we propose an over-the-air federated policy gradient algorithm, where all agents simulta-neously broadcast an analog signal carrying local information to a common wireless channel, and a central controller uses the received aggregated waveform to update the policy parameters. We investigate the effect of noise and channel distortion on the convergence of the proposed algorithm, and establish the complexities of communication and sampling for finding an E-approximate stationary point. Finally, we present some simulation results to show the effectiveness of the algorithm.

The information regarding the driving states of all vehicles is crucial for achieving optimal group performance in a vehicle platoon. This article focuses on the fully distributed driving state estimation problem in open vehicle platoons, which frequently experience arrivals and departures of vehicles. To address this problem, we propose a distributed driving state observer inspired by the leader–follower consensus technique. This observer can reconstruct the global driving state of the platoon, including the positions, velocities, and accelerations of all vehicles. We also derive the necessary and sufficient conditions to ensure the stability of its estimation error dynamics. The proposed observer is highly flexible in platoons with a strongly connected communication network, as it can be constructed and operated using the local knowledge of each vehicle only, without relying on global information of a platoon such as the number of vehicles. We demonstrate the observer's plug-and-play operations in the face of platoon merging and splitting and analyze its estimation stability. Extensive simulation results demonstrate the effectiveness of our theoretical results and the potential of the proposed observer for platoon control.

For a long period, automated vehicles (AVs) or vehicle platoons will coexist with human-driven vehicles (HDVs) in heterogeneous traffic flow, where the cut-in maneuver of human drivers can be frequently expected. In this paper, to understand and simulate the driver decisions on whether to continue the cut-in and when to execute the lane-change during the cut-in process, we propose a two-layer prediction-based decision model by integrating a dynamic prediction module, a continuity decision module, and an execution decision module. To our best knowledge, this is the first study to model the driver decision behavior of the cut-in process. Cut-in experiments are conducted to collect the decision and control data of drivers under one-and two-target-vehicle scenarios, which both include sixty sub-scenarios with different initial velocities, accelerations, or positions of the vehicles. We prove the effectiveness of the proposed model in simulating the driver decision behavior of the cut-in process by comparing the experimental and simulation results under various scenarios over different subjects. Besides, we analyze the effects of some model parameters on the model performance to show their ability to represent different driving styles.

Vehicle platoons (VPs) are groups of vehicles driving together with a short inter-vehicle gap and a harmonized velocity. For a long period, the VPs and human-driven vehicles (HDVs) will coexist in mixed traffic flow, where the cut-in maneuver of the HDVs towards the VPs can be frequently expected. In this paper, to handle such cut-ins, we propose an intention prediction-based control method for the VPs by considering the tradeoff between the platoon integrity and traffic safety. Particularly, the proposed method is designed to prevent as many cut-ins as possible while taking care of the road safety. It consists of a cut-in prediction part, including intention and trajectory prediction algorithms, and a finite state machine (FSM)-based predictive control part, including a high-level FSM and a low-level predictive control. Driver-in-the-loop experiments were conducted in the VP-based driving scenarios to train the intention prediction algorithm and test the proposed method. We show the results detailing the control behavior of the proposed method in a no cut-in test, a mandatory cut-in test, and three discretionary cut-in tests. The results demonstrate that the proposed method can predict the cut-in intention of human drivers in real time. Besides, according to the prediction results, the proposed method can prevent cut-ins for the VPs while taking care of the road safety.

In this paper, we propose a Bi-layer Prediction-based Reduction Branch (BP-RB) framework to speed up the process of finding a high-quality feasible solution for Mixed Integer Programming (MIP) problems. A graph convolutional network (GCN) is employed to predict binary variables' values. After that, a subset of binary variables is fixed to the predicted value by a greedy method conditioned on the predicted probabilities. By exploring the logical consequences, a learning-based problem reduction method is proposed, significantly reducing the variable and constraint sizes. With the reductive sub- MIP problem, the second layer GCN framework is employed to update the prediction for the remaining binary variables' values and to determine the selection of variables which are then used for branching to generate the Branch and Bound (B&B) tree. Numerical examples show that our BP-RB framework speeds up the primal heuristic and finds the feasible solution with high quality.

A vehicle platoon is a group of vehicles driving together with a harmonized speed and a short inter-vehicle gap by using vehicle automation and vehicle-to-vehicle communication. Platoons have to share road with human-driven vehicles (HDVs) and can only be applied in heterogeneous traffic flow for a long period. Driver cut-in behavior (DCB) towards a platoon can be frequently expected in such driving context. In this paper, to understand and simulate such behavior, we propose a platoon-oriented cut-in behavior (POCB) model by fusing a lateral and a longitudinal control model into the queuing network (QN) cognitive architecture. Platoon-oriented cut-in experiments are conducted to collect driver data under cut-in from back and front scenarios, which both include six sub-scenarios with different platoon gaps or initial velocities. We demonstrate the effectiveness of the proposed model in simulating the DCB towards platoons by comparing experimental and simulation results under various driving scenarios across different subjects.