General Multi-agent Systems Research Articles

Dynamic Event-Triggered-Based Finite-Time Distributed Tracking Control of Networked Multi-UAV Systems

The distributed tracking of multiple unmanned aerial vehicles (UAVs) is a hotspot due to its broad applications in various fields, while continuous communication among UAVs is often impractical, especially in time-sensitive tasks or environments with limited bandwidth. With this in mind, this paper presents a finite-time leader-following distributed tracking control scheme for general multi-agent systems, with a particular emphasis on its application for networked UAVs. Theoretically, a dynamic event-triggered mechanism is proposed, which features a novel finite-time stable dynamic variable within its triggering rule, ensuring that neither controller updates nor trigger detection requires continuous communication. This event-triggered finite-time controller facilitates efficient network resource management and timely mission response in UAV cooperation, enhancing adaptability to onboard wireless communication networks and time-sensitive tasks. The method allows for the customization of parameters in the internal dynamic variables to adjust the convergence rate and event-triggering frequency of the system, while also preventing Zeno behavior. Moreover, a Lyapunov-based analysis is conducted to theoretically verify the finite-time stability of the closed-loop system and its applicability in directed communication networks. Finally, some numerical simulations are performed to validate the effectiveness of the proposed distributed control scheme for networked multi-UAV systems.

Distributed formation containment control for multi-agent systems via dynamic event-triggering communication mechanism

This paper focuses on resolving the distributed formation containment control issue in general multi-agent systems (MASs) possessing multiple leader agents subject to limited communication resources. The unnecessary data transmissions among agents in MASs are reduced by utilizing a unique dynamic event-triggering communication mechanism (DETCM), which can build communication time sequences for each agent automatically. Based on the sampled states of agents at event-triggering instants, a novel distributed formation containment control protocol is proposed. By mathematical analysis, the formation containment control problem is turned into the problem of stability in a reduced-order system. The global formation containment of the MASs is finally achieved through the application of Lyapunov-Krasovskii functionals and general inequalities. Moreover, a valid example is also given to support the validity and effectiveness of the findings.

An introductory review of swarm technology for spacecraft on‐orbit servicing

AbstractThis review paper presents a comprehensive evaluation and forward‐looking perspective on the underexplored topic of servicing target objects using spacecraft swarms. Such targets can be known or unknown, cooperative or uncooperative, and pose significant challenges in modern space operations due to their inherent complexity and unpredictability. Successfully servicing space objects is vital for active debris removal and broader on‐orbit servicing tasks such as satellite maintenance, repair, refueling, orbital assembly, and construction. Significant effort has been invested in the literature to explore the servicing of targets using a single spacecraft. Given its advantages and benefits, this paper expands the discussion to encompass a swarm approach to the problem. This review covers various single‐spacecraft approaches and presents a critical examination of the existing, although limited, body of work dedicated to servicing orbital objects using multiple spacecraft. The focus is also broadened to include some influential studies concerning the characterization, capture, and manipulation of physical objects by general multiagent systems, a subject with significant parallels to the core interest of this manuscript. Furthermore, this article also delves into the realm of simultaneous localization and mapping, highlighting its application within close‐proximity operations in space, especially when dealing with unknown uncooperative targets. Special attention is paid to the benefits that this field can receive from distributed multiagent architectures. Finally, an exploration of the promising field of swarm robotics is presented, with an emphasis on its potential to revolutionize the servicing of orbital target objects. Concurrently, a survey of general research directly engaging swarms in the orbital context is conducted. This review aims to bridge the knowledge gap and stimulate further research in the underexplored domain of servicing space targets with spacecraft swarms.

Leader–Follower Tracking for General Multiagent Systems With Unknown Leader Input and Limited Actuation

Leader–Follower Tracking for General Multiagent Systems With Unknown Leader Input and Limited Actuation

Fixed-time time-varying output formation–containment control of heterogeneous general multi-agent systems

This work investigates fixed-time time-varying output formation–containment control of heterogeneous general multi-agent systems, which comprises one virtual leader, multiple leaders, and followers. First, fixed-time nonlinear control law is constructed, such that leaders can achieve time-varying output formation in a finite time. Meanwhile, leaders can track the virtual leader’s output trajectory. Next, two complete distributed adaptive fixed-time observers are designed aiming to recover a convex hull. This is key to control issue in this paper. Further, nonlinear control law is constructed for followers, such that followers can move into the convex hull formed by multiple leaders. Finally, two examples are provided to verify the feasibility of the theoretical results.

Fixed-time consensus for stochastic multi-agent systems with discontinuous dynamics via quantized control

In this study, the fixed-time consensus (FDTC) for stochastic multi-agent systems (MASs) with discontinuous inherent dynamics is investigated via quantized control. Firstly, an improved lemma for fixed-time (FDT) stability is derived and several more precise estimations for settling time (SLT) are gained by using certain special functions. Secondly, a more general MAS containing discontinuous inherent dynamics and stochastic perturbations is considered, which is closer to practical life. Thirdly, to overcome the limitation of communication, two kinds of quantized control protocols are designed. Besides, in the light of the graph theory, non-smooth analysis, fixed-time (FDT) stability and stochastic analysis theory, some sufficient conditions are put forward to achieve FDTC of MASs. Finally, the validity of the derived theoretical results is testified by two numerical examples.

Learning to Schedule Joint Radar-Communication With Deep Multi-Agent Reinforcement Learning

Radar detection and communication are two essential sub-tasks for the operation of next-generation autonomous vehicles (AVs). The forthcoming proliferation of faster 5G networks utilizing mmWave has raised concerns on interference with automotive radar sensors, which has led to a body of research on Joint Radar-Communication (JRC). This paper considers the problem of time-sharing for JRC, with the additional simultaneous objective of minimizing the average age of information (AoI) transmitted by a JRC-equipped AV. We first formulate the problem as a Markov Decision Process (MDP). We then propose a more general multi-agent system, with an appropriate medium access control (MAC) protocol, which is formulated as a partially observed Markov game (POMG). To solve the POMG, we propose a multi-agent extension of the Proximal Policy Optimization (PPO) algorithm, along with algorithmic features to enhance learning from raw observations. Simulations are run with a range of environmental parameters to mimic variations in real-world operation. The results show that the chosen deep reinforcement learning methods allow the agents to obtain strong performance with minimal a priori knowledge about the environment.

Newton Optimization on Helmholtz Decomposition for Continuous Games

Many learning problems involve multiple agents optimizing different interactive functions. In these problems, the standard policy gradient algorithms fail due to the non-stationarity of the setting and the different interests of each agent. In fact, algorithms must take into account the complex dynamics of these systems to guarantee rapid convergence towards a (local) Nash equilibrium. In this paper, we propose NOHD (Newton Optimization on Helmholtz Decomposition), a Newton-like algorithm for multi-agent learning problems based on the decomposition of the dynamics of the system in its irrotational (Potential) and solenoidal (Hamiltonian) component. This method ensures quadratic convergence in purely irrotational systems and pure solenoidal systems. Furthermore, we show that NOHD is attracted to stable fixed points in general multi-agent systems and repelled by strict saddle ones. Finally, we empirically compare the NOHD's performance with that of state-of-the-art algorithms on some bimatrix games and continuous Gridworlds environment.

Practical scaled consensus for nonlinear multiagent systems with input time delay via a new distributed integral-type event-triggered scheme

This paper studies scaled-based practical consensus issue for multiagent systems with input time delay by a fully continuous communication-free integral-type event-triggered scheme. By choosing the proper scales, scaled consensus can be induced to synchronization, bipartite consensus or cluster consensus. By defining a continuous communication-free measurement error for the integral-type event-triggered mechanism, the new integral-type event-triggered condition is proposed which can not only reduce the energy consumption but also prolong the interevent time. Then, with time domain analysis method, the distributed integral-type event-triggered control problem for nonlinear general multiagent systems involving input time delay is investigated and then the second-order counterpart, with a calculated upper-bound for time-delay. Moreover, it is concluded that with such event-triggered protocols, practical scaled consensus can be achieved without the exhibition of Zeno behavior. At last, simulations are shown to support the results.

A distributed adaptive scheme for multiagent systems

AbstractIn traditional adaptive control, the certainty equivalence principle suggests a two‐step design scheme. A controller is first designed for the ideal situation assuming the uncertain parameter was known, and it renders a Lyapunov function. Then, the uncertain parameter in the controller is replaced by its estimation that is updated by an adaptive law along the gradient of Lyapunov function. This principle does not generally work for a multiagent system as an adaptive law based on the gradient of (centrally constructed) Lyapunov function cannot be implemented in a distributed fashion, except for limited situations. In this paper, we propose a novel distributed adaptive scheme, not relying on gradient of Lyapunov function, for general multiagent systems. In this scheme, asymptotic consensus of a second‐order uncertain multiagent system is achieved in a network of directed graph.

Robust containment control of heterogeneous non‐linear multi‐agent systems via power series approach

In this study, the robust containment control problem of general non-linear multi-agent systems (MASs) subject to structural uncertainties is studied. The leaders are also described by general non-linear dynamics satisfying a locally quasi-Lipschitz condition and a distributed non-linear observer is designed to estimate the leaders' states for the followers. The solvability of the regulator equations associated with the non-linear dynamics of agents is normally essential for solving the containment control problem of heterogeneous MASs, but the closed-form solution of many non-linear regulator equations may not be obtained. Towards this end, the power series approach is adopted to decompose the regulator equations into a series of solvable linear equations. Based on the solution of the linear equations as the feedforward information, the distributed robust containment control scheme based on state feedback and output feedback control is proposed. The p -copy internal model is employed to compensate the parameter uncertainties of the follower agents. A numerical example is studied to demonstrate the effectiveness and efficiency of the proposed control law.

Consensus seeking of multi-agent systems with intermittent communication: a persistent-hold control strategy

ABSTRACT To deal with the consensus problem of general first-order multi-agent systems with intermittent communication, a persistent-hold consensus algorithm is designed by introducing the hold control after the usual intermittent control. Under the fixed topology, consensus convergence conditions are obtained based on matrix theory and graph theory for the agents under our proposed persistent-hold strategy, and the results show that our proposed strategy has less conservative converging conditions than the usual intermittent control. Besides, consensus convergence conditions are also obtained for the agents under the switching topologies. Numerical simulations show the correctness of theoretical results.

Formation control with disturbance rejection for a class of Lipschitz nonlinear systems

In this paper, we consider the leader-follower formation control problem for general multi-agent systems with Lipschitz nonlinearity and unknown disturbances. To deal with the disturbances, a disturbance observer-based control strategy is developed for each follower. Then, a time-varying formation protocol is proposed based on the relative state of the neighbouring agents and sufficient conditions for global stability of the formation control are identified using Lyapunov method in the time domain. The proposed strategy and analysis guarantee that all signals in the closed-loop dynamics are uniformly ultimately bounded and the formation tracking errors converge to an arbitrarily small residual set. Finally, the validity of the proposed controller is demonstrated through a numerical example.

Group controllability of discrete-time multi-agent systems

Controllability is a fundamental issue concerning control of multi-agent networks and a very important research topic in the modeling, analysis and coordination control of multi-agent systems. Group controllability problem is a further extension of the controllability problem of the general multi-agent systems, which mainly studies the cooperation and control of multi-agent systems with multiple sub-groups or multiple intelligence clusters. Comparing with the controllability of the general multi-agent systems, the group controllability is not only to consider the information interaction among the groups, but also to consider the information interaction between different groups, which makes the system reflect the effect of a whole and also the internal structure of the sub-groups. This paper addresses the group controllability problems of discrete-time multi-agent systems with time-delay, in which both switching topology and fixed topology are considered. This paper also proposes the general definition of the group controllability, as well as establishes group controllability criteria from the algebraic and graphical perspectives. Numerical examples and simulations are proposed to illustrate the theoretical results.

Noise tolerance leader-following of high-order nonlinear dynamical multi-agent systems with switching topology and communication delay

This paper investigates the leader-following tracking consensus problem for high-order nonlinear dynamical multi-agent systems with switching topology and communication delay under noisy environments. In order to reflect a more realistic situation, we introduce a general multi-agent systems model and also further investigate its robust consensus under noisy environments, the topology of the network is assumed to be in a finite set of arbitrarily stochastic switching, the communication delay is also considered in the tracking control protocols. The mean square consensus sufficient conditions of multi-agent systems are explored via the common stochastic Lyapunov functional stability theory, in other words, can be solved by linear matrix inequality schemes. The mean square consensus condition is derived to provide a rigorous condition for leader-following of high-order nonlinear dynamical multi-agent systems with considerable scale. In particular, we prove that the proposed algorithm is robust against the bounded communication delay in noisy environments. On the other hand, when it involves many multi-agent systems, a more conservative but effective consensus protocol is also raised. The consensus protocols only require low-dimensional matrices, which are independent of the network size. In addition, the consensus criteria of two cases without communication delay or noisy environment are also proposed. A simple optimization program is also developed to determine the maximum allowable communication delay. Finally, in order to demonstrate the effectiveness and feasibility of the consensus protocol obtained in this paper, the numerical examples are given.

Observer based consensus for nonlinear multi-agent systems with communication failures

In this paper, consensus problem of general nonlinear multi-agent systems with communication failures is studied based on a distributed consensus algorithm. Under the assumption that the communication graphs are strongly connected, we derive some sufficient consensus criterion for both cases fixed and switched topologies using the relative measured output states of each agent. Moreover, the case that there is a leader in the communication graph is also explicitly investigated. Finally, simulation examples are presented to show the effectiveness of the obtained results.

Multi-agent simulation on multiple GPUs

Multi-agent simulation is widely used in many areas including biology, economic, political, and environmental science to study complex systems. Unfortunately, it is computationally expensive. In this paper, we shall explore the implementation of a general multi-agent simulation system on a system with multiple GPUs acting as accelerators. In particular, we have ported the popular Java multi-agent simulation framework MASON to a nVidia CUDA-based multi-GPU setting. We evaluated our implementation over different numbers and types of nVidia GPUs. For our evaluation, we ported three models in the original version of MASON. On the well-known Boids model, we achieved a speedup of 187×. Using a fictional model, we showed that speedup of up to 468× is possible. In the paper, we shall also describe the detailed internals of our system, and the various issues we encountered and how they were solved.

Observer-based distributed consensus for general nonlinear multi-agent systems with interval control inputs

In this paper, observer-based distributed consensus for general nonlinear multi-agent systems with interval control inputs under strongly connected balanced topology is encountered when the relative states of agents are unavailable or undesirable. Theoretical analysis method is further extended to the case of general nonlinear multi-agent systems under switching setting. Moreover, tracking problem on the leader–follower scenario is also explicitly investigated under a mutual assumption that the communication graph, which represents the interaction among agents, contains a directed spanning tree with the leader as its root. It is shown that the consensus for underlying considered multi-agent systems can be desirable as long as the data missing rate does not exceed a certain threshold. Finally, simulation examples are presented to effectively corroborate the analytical findings.

Containment control of general multi-agent systems with directed random switching topology

This paper devotes itself to the containment tracking problems of general time-varying high-order multi-agent systems (MAS). In this paper, the directed graph (digraph) topologies of the MAS are random Markovian switching, and the dynamics are not merely confined to be linear. The control protocol that only based on the neighbor׳s states is designed. The containment tracking results are obtained in the asymptotic unbiased mean square sense and with bounded error in mean square sense for general linear case and nonlinear case respectively. Simulations are also provided to demonstrate the effectiveness of our results.

Multi-Agent System for Business Applications

In this paper we describe a multi-agent system which implements distributed computational system. The software architecture identifies distributed locations as abstractions of distributed computational system and various software agents as abstractions of different autonomous functional components. The interactions among agents within a system are the most critical for the application. Most common and one of the elementary types of e-commerce applications is consumer buying behavior. We propose a model of a general consumer buying behaviour application using general multi-agent system to examine its expressiveness. By identifying specific roles of software agents and designing new specific interaction patterns for them we implement the commercial model.