Problem Of Multi-agent Systems Research Articles

Nearly optimal fixed time sliding mode controller for leader–follower consensus problem with partially unknown nonlinear agents

This paper presents a conceptual framework for addressing the consensus problem in multi-agent systems with unknown nonlinear dynamics using reinforcement learning (RL) based nearly optimal sliding mode controller (SMC). The agents’ dynamics are assumed to have uncertainty and mismatched disturbance. An adaptive fixed-time estimator is introduced to estimate uncertain dynamics and disturbances for each agent at a certain time. The paper proposes two control strategies. In the first strategy, a controller is designed, incorporating adaptive SMC and an optimal controller. Adaption law in SMC estimates the bound of fixed time estimator error before convergence, ultimately achieving asymptotic convergence to the sliding surface and converting the agent’s dynamics to linear. This enables solving a linear consensus problem using an RL-based adaptive optimal controller through an on-policy critic–actor method. The second control strategy enhances the adaptive SMC into a fixed-time controller, reducing the time to reach the sliding surface regardless of initial conditions. Consequently, the convergence time of the consensus error to zero is diminished. This reduction in reaching time results in faster convergence of the consensus error to zero. The effectiveness of both strategies is validated through numerical experiments on two real system models, aligning with the theoretical proofs.

Bipartite containment control of multi-agent systems subject to adversarial inputs based on zero-sum game

In this paper, we investigate bipartite containment control problem of multi-agent systems (MASs) with signed directed graph under adversarial inputs. Firstly, we define the bipartite containment error and establish the equivalence between the bipartite containment error converging to zero and the achievement of bipartite containment control. Subsequently, we prove that the bounded L2-gain bipartite containment problem under adversarial inputs can be reformulated as a multi-player zero-sum differential graphical game problem and can be solved via the solution to the coupled Hamilton-Jacobi-Isaacs (HJI) equation. To address this, we propose a policy iteration (PI) algorithm and prove its convergence under different updating cases. The proposed algorithm is implemented by neural networks (NNs) and a numerical simulation example is provided to show its effectiveness.

Prescribed-time leader-follower consensus and containment control for nonlinear multi-agent systems with event-triggered control protocols

This paper discusses the prescribed-time leader-follower consensus and containment control problems for a class of nonlinear multi-agent systems with external disturbances. First, an event-triggered control protocol with sign function is proposed, which can ensure that all agents achieve consensus within a prescribed time. Second, as the sliding mode control has the advantage of good robustness to external disturbances, an improved event-triggered control protocol is designed by using the integral sliding mode control method and some sufficient conditions are derived for achieving the prescribed-time leader-follower consensus. Moreover, the prescribed-time containment control problem for multi-agent systems with multiple leaders is investigated by using the event-triggered control strategy. It is shown that Zeno behavior can be excluded in the above consensus issues by choosing parameters appropriately. Finally, several numerical examples are given to demonstrate the effectiveness and reliability of the proposed approaches.

Min–max consensus of multi-agent systems in random networks

This paper focuses on the study of the min–max consensus problem for multi-agent systems in random networks. A min–max consensus algorithm is proposed, which incorporates the weighted average of the agent’s state, the maximum value of its neighbors’ states, and the minimum value of its neighbors’ states. The interaction network is composed of a directed random network where edges exist with probability and are independent of each other. Then, the convergence of the min–max consensus algorithm is investigated utilizing max-plus algebra, min-plus algebra, and stochastic analysis theory. Sufficient and necessary conditions are given to ensure that the multi-agent systems achieve min–max consensus almost surely and in mean-square, respectively. Finally, some numerical simulations are conducted to confirm the effectiveness of the proposed consensus algorithm.

Multi-group consensus of multi-agent systems subject to semi-Markov jump topologies against hybrid cyber-attacks

Multi-agent systems are widely used in practice. The existing results on the consensus of multi-agent systems mainly focus on the complete consensus, which can not deal with the situation of the system in the face of multiple tasks. In this paper, the multi-group consensus problem of multi-agent systems with semi-Markov jump topologies against hybrid cyber-attacks is studied. The purpose is to ensure that the multi-agent systems can achieve different consensus goals when they encounter hybrid cyber-attacks, and to solve the problem of time-varying communication topology between agents modeled by the semi-Markov process. Additionally, two types of cyber-attacks are taken into account in a single framework through two independent sets of Bernoulli sequences. Subsequently, a necessary condition is established to guide the system toward achieving the multi-group consensus objective. Lastly, a series of illustrative examples are presented to underscore the validity of the designed controller.

Adaptive expected-time tracking consensus control for multiagent systems with external disturbances and time delays

This paper studies the expected-time tracking control problem for multiagent systems with disturbances and time delays. The expected-time control scheme can ensure that control objectives can be completed within the desired time. In the other words, the output signal can track the reference signal in an expected time. Based on a special error and the adaptive technology, the expected-time tracking consensus control scheme is achieved. Moreover, by using the neural networks technique, the unknown nonlinear functions which often exist in the dynamic of the system are handled. According to the Lyapunov stability theorem and the Lyapunov–Krasovskii function, it is proved that the consensus errors are semi-globally uniformly ultimately bounded. At last, the feasibility of the control method is verified by a simulation example.

Distributed event-triggered collaborative control for multiagent systems against DoS attacks

This paper deals with the event-triggered collaborative control problem of multiagent systems (MASs) under denial-of-service (DoS) attacks. For large-scale MASs, it is difficult to implement fully distributed collaborative control by tuning the coupling strengths of all edges due to the limitation of computing resources in practical applications. To overcome this difficulty, an edge-pinning-based adaptive strategy for coupling strengths is established, under which only the coupling strengths of partial edges are tuned instead of all ones. Under this strategy, the edge-pinning-based resilient distributed event-triggered synchronous and asynchronous adaptive laws are designed, both of which ensure the implementation of security consensus against DoS attacks and the elimination of Zeno behavior. It should be pointed out that the designed two resilient consensus strategies are fully distributed. Additionally, under these event-triggered strategies, the updating of control protocols and the monitoring of triggering functions do not require continuous communications, thus reducing the communication burden among agents. Finally, simulations conducted on a network composed by six aircrafts are provided to illustrate the theoretical results.

Dynamic Multi-Target Self-Organization Hunting Control of Multi-Agent Systems

In this paper, we present a novel coordinated method tailored to address the dynamic multi-target hunting control problem in multi-agent systems, offering significant practical value. Our approach encompasses several key components: initially, we introduce a task allocation model that integrates a fuzzy inference system with a particle swarm optimization algorithm. This hybrid model efficiently allocates hunting tasks for scattered evading targets, effectively transforming the dynamic multi-target hunting problem into multiple dynamic single-target-hunting problems. This transformation enhances the speed and efficacy of task allocation. Subsequently, we propose an attraction/repulsive model grounded in potential field theory. This model facilitates the coordinated hunting of each target by organizing agents into subgroups. Relying solely on relative position and velocity information between agents and targets, our model simplifies computation, while maintaining effectiveness. Furthermore, the coordination of hunting activities for each target is achieved through a series of agent subgroups, guided by our proposed motion model. This systematic approach ensures a cohesive and efficient hunting strategy. Finally, we validate the effectiveness and feasibility of our proposed method through simulation results. These results provide empirical evidence of the method’s efficacy and potential applicability in real-world scenarios.

Robust Distributed Containment Control with Adaptive Performance and Collision Avoidance for Multi-Agent Systems

This paper deals with the containment control problem for multi-agent systems. The objective is to develop a distributed control scheme that leads a sub-group of the agents, called followers, within the convex hull that is formed by the leaders, which operate autonomously. Towards this direction, we propose a twofold approach comprising the following: (i) a cyber layer, where the agents establish, through the communication network, a consensus on a reference trajectory that converges exponentially fast within the convex hull of the leaders and (ii) a physical layer, where each agent tracks the aforementioned trajectory while avoiding collisions with other members of the multi-agent team. The main contributions of this work lie in the robustness of the proposed framework in both the trajectory estimation and the tracking control tasks, as well as the guaranteed collision avoidance, despite the presence of dynamic leaders and bounded but unstructured disturbances. A simulation study of a multi-agent system composed of five followers and four leaders demonstrates the applicability of the proposed scheme and verifies its robustness against both external disturbances that act on the follower model and the dynamic motion of the leaders. A comparison with a related work is also included to outline the strong properties of the proposed approach.

Guaranteed-performance consensus for multi-agent systems with Lipschitz nonlinearity

This paper mainly addresses the guaranteed-performance consensus problem of multi-agent systems (MASs), under the energy constraints and Lipschitz nonlinear dynamics. First, to reach consensus under energy constraints, the consensus protocol and the quadratic performance function are proposed. Afterward, by using the properties of nonsingular transformation matrix and Lipschitz’s nonlinear representation, the consensus function conditions are expressed. Moreover, considering the guaranteed-performance cost, this paper designs the quadratic performance function derived from the Riccati inequality and Lyapunov–Krasovskii theorem. Eventually, numerical simulations are implemented to validate the theoretical results.

Reward function design method to achieve system-level objectives in ambulance diversion problem

This study addresses a reward function design problem in multi-agent systems, using ambulance diversion (AD) problem in emergency medical services (EMS) as an application. While Emergency Departments (EDs) may declare AD to alleviate overcrowding, excessive use of this strategy by individual EDs can have negative consequences, such as limited access and delayed care for emergency patients. To eliminate negative consequences caused by individual decisions, we propose a reward function design method to align individual agents’ decisions with socially optimal strategies. This method ensures that an equilibrium status is achieved under socially optimal operation by removing incentives for deviations from socially optimal behavior while maximizing system welfare without extra investment. The proposed method offers an exact solution for designing reward function of a stochastic game model with finite state and action spaces. We conduct numerical experiments for a 2-ED system and demonstrate that the designed reward function properly guides the EDs toward a socially optimal AD strategy.

Event-triggered predictive control for cooperation-competition multi-agent systems under DoS attacks

This paper concerns the bipartite consensus problem of multi-agent systems(MASs) with competitive- cooperative network topology under denial-of-service (DoS) attacks. Firstly, this work extensively analyzes the competitive phenomena that may exist in the information interchange of agents in contrast to the single cooperative behavior between agents. Based on this, some necessary conditions are provided for the system to attain the bipartite consensus. In addition, the event-triggered mechanism (ETM) effectively lowers unnecessary information sharing between agents and eliminates Zeno behavior. Furthermore, the predictive method provides the system with exceptional resistance against common energy-limited DoS attacks and the ability to compensate for information loss caused by DoS attacks. Finally, a numerical simulation proves that the proposed approach is feasible.

Containment control for second-order multi-agent systems with time-varying delays via variable-augmented-based free-weighting matrices

This paper investigates the containment control problem for multi-agent systems (MASs), where the time-varying communication delays are considered. The aim of this paper is to introduce new methods for the containment control research of MASs with time-varying delays. To obtain better results on maximum admissible upper bounds (MAUBs) of the delays to ensure the containment control, the variable-augmented-based free-weighting matrices (VABFWMs) method is introduced to avoid the higher-order terms of time-varying delays in estimating the derivative of the Lyapunov-Krasovskii functionals (LKFs). So our method no longer requires the quadratic function negative-determination lemma. In addition, by utilizing the proposed method, we design a containment controller to ensure that all followers are included into the convex hull formed by leaders. Finally, a numerical example is presented to illustrate the feasibility and superiority of our method.

A Distributed Model Predictive Control Strategy for Constrained Multi-Agent Systems: The Uncertain Target Capturing Scenario

In this paper, a distributed control architecture is presented for addressing the target capturing problem of a multi-agent system whose dynamics is described by double integrator models subject to bounded disturbance effects. Starting from novel kinematic models used as reference trajectories, the aim consists in driving the multi-agent system within the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">containment</i> region without entering the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">distancing</i> one and, whenever necessary, there remaining confined. The underlying control problem has been tackled by means of model predictive control arguments. In particular, two different distributed strategies have been developed, and adequately switched to each other, in order to guarantee constraint satisfaction within the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">capturing</i> region despite any disturbance realization. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This paper proposes a methodological solution for dealing with the capturing problem for multi-agent systems operating in uncertain environments where the target is defined by the ring resulting from the distancing and containment ellipsoids. Differently from the existing literature, the proposed approach combines into a unique framework properties coming from potential fields theory and distributed model predictive control philosophy. It is interesting to put in light that the underlying control strategy allows the users to face surveillance and rescue operations, to cite a few, in computational affordable way due to the required memory resources because most of computations can be straightforwardly moved in the off-line phase.

Energy-Efficient Distributed Formation Control of Sampled-Data Multiagent Systems With Packet Losses.

This article investigates an energy-efficient formation control problem for multiagent systems with sampled data and random packet losses. A Bernoulli stochastic variable is used to describe packet losses, which is defined relative to the transmission energy consumed by antennas. A distributed sampled-data formation control law is designed and some analytic sufficient conditions on the formation control are derived. It is revealed that the sampling period, control parameters, network topology, as well as the transmission energy used by sensors impose inherent limitations in achieving the formation. Also, a method for searching the minimum allowable transmission energy is presented. Finally, numerical simulations are shown for illustration and verification.

Coalition Formation for Task Allocation Using Multiple Distance Metrics (Student Abstract)

Simultaneous Coalition Structure Generation and Assignment (SCSGA) is an important research problem in multi-agent systems. Given n agents and m tasks, the aim of SCSGA is to form m disjoint coalitions of n agents such that between the coalitions and tasks there is a one-to-one mapping, which ensures each coalition is capable of accomplishing the assigned task. SCSGA with Multi-dimensional Features (SCSGA-MF) extends the problem by introducing a d-dimensional vector for each agent and task. We propose a heuristic algorithm called Multiple Distance Metric (MDM) approach to solve SCSGA-MF. Experimental results confirm that MDM produces near optimal solutions, while being feasible for large-scale inputs within a reasonable time frame.

An Estimation of Distribution Based Algorithm for Continuous Distributed Constraint Optimization Problems

Continuous Distributed Constraint Optimization Problem(C-DCOP) is a constraint processing framework for continuous variables problems in multi-agent systems. There is a constraint cost function between two mutually restrictive agents in C-DCOP. The goal of the C-DCOP solving algorithm is to keep the sum of constraint cost functions in an extreme state. In a C-DCOP, each function is defined by a set of continuous variables. At present, some C-DCOP solving algorithms have been proposed, but there are some common problems such as the limitation of constraints cost function form, easy to fall into local optimum, and lack of anytime attribute. Aiming at these thorny problems, we propose a parallel optimization algorithm named Estimation of Distribution Based Algorithm for Continuous Distributed Constraint Optimization Problems (EDA-CD). In EDA-CD, each solution is regarded as an individual, and the distribution of agent value is jointly described by all outstanding individuals. Firstly, all agents cooperate to hold a distributed population. Secondly, each agent calculates the mean and variance of its variables to build probability models in parallel. Finally, the agent evaluates the fitness of samples and updates the probability model through cooperative communication on Breadth First Search (BFS) pseudo-tree. We theoretically prove that EDA-CD is an anytime algorithm. The extensive experimental results on four types of benchmark problems show that the proposed EDA-CD outperforms the state-of-the-art C-DCOP algorithms and has about 20% improvement in solution quality.

Periodic event-triggered bipartite containment control for nonlinear multi-agent systems with input delay

ABSTRACT This paper studies the periodic event-triggered bipartite containment control problem of multi-agent systems with unmeasurable states and input delays. To deal with the system input delay problem, a Pade approximation technique is introduced. By utilising the sampled output of the system, a fuzzy state observer is constructed to estimate the unknown state. A periodic event-triggered mechanism is designed, which can effectively reduce the usage of communication resources. Compared with the traditional event-triggered control manner, a periodic event-triggered control manner only monitors the preset threshold condition at the sampling instant to determine whether to update the current control signal, which can automatically guarantee the minimum lower bound of the execution interval, avoiding the occurrence of Zeno behaviour. Stability analysis verifies that all signals in the closed-loop system are bounded, and the bipartite tracking error can converge to a small neighbourhood of the origin. Finally, a practical example further verifies the effectiveness of the designed control scheme.

Consensus of multi-agent systems with randomly occurring nonlinearities via uncertain pinning control under switching topologies

This paper is focused on the consensus problem of multi-agent systems via uncertain pinning control under switching topologies. The stochastic disturbances and randomly occurring nonlinearities are proposed to describe more realistic systems. The communication topology is modeled by a directed graph and it is divided into two cases, the consensus problem is discussed in these two cases. In addition, there exist some uncertain pinning connections between the followers and leader due to switching topologies, the distributed control protocol is designed to satisfy the follower asymptotically converge to the leader. By constructing suitable multiple Lyapunov functions and utilizing tools of M-matrix theory, some sufficient consensus criteria are deduced to reach this goal. Finally, two examples are given to verify the correctness of the proposed method.

Online Planning for Autonomous Mobile Robots with Different Objectives in Warehouse Commissioning Task

Recently, multi-agent systems have become widespread as essential technologies for various practical problems. An essential problem in multi-agent systems is collaborative automating picking and delivery operations in warehouses. The warehouse commissioning task involves finding specified items in a warehouse and moving them to a specified location using robots. This task is defined as a spatial task-allocation problem (SPATAP) based on a Markov decision process (MDP). It is considered a decentralized multi-agent system rather than a system that manages and optimizes agents in a central manner. Existing research on SPATAP involving modeling the environment as a MDP and applying Monte Carlo tree searches has shown that this approach is efficient. However, there has not been sufficient research into scenarios in which all agents are provided a common plan despite the fact that their actions are decided independently. Thus, previous studies have not considered cooperative robot behaviors with different goals, and the problem where each robot has different goals has not been studied extensively. In terms of the cooperative element, the item exchange approach has not been considered effectively in previous studies. Therefore, in this paper, we focus on the problem of each robot being assigned a different task to optimize the percentage of picking and delivering items in time in social situations . We propose an action-planning method based on the Monte Carlo tree search and an item-exchange method between agents. We also generate a simulator to evaluate the proposed methods. The results of simulations demonstrate that the achievement rate is improved in small- and medium-sized warehouses. However, the achievement rate did not improve in large warehouses because the average distance from the depot to the items increased.