In this paper, the distributed Nash equilibrium (NE) searching problem is investigated, where feasible action sets are constrained by nonlinear inequalities and linear equations. Different from most of existing investigations on NE problems, we consider case both cost functions depend actions all players, each player can only have access to information its neighbors. To address problem, a continuous-time gradient-based projected algorithm proposed, leader-following consensus employed for...

In this paper, we study the consensus problem of discrete-time and continuous-time multiagent systems with distance-dependent communication networks, respectively. The weight between any two agents is assumed to be a nonincreasing function their distance. First, consider networks fixed connectivity. case, interaction adjacent always exists but influence could possibly become negligible if distance long enough. We show that can reached under arbitrary initial states decay rate less than given...

This technical note aims to design a decentralized control strategy for multiple double-integrator agents achieve flocking while maintaining an angle-constrained triangulated formation shape in the plane. We novel controller that can steer common velocity meeting some angle constraints; preserve rigidity during agents' motion; and be implemented absence of global coordinate system. Due preservation property, proposed guarantees almost convergence satisfying predefined constraints. Moreover,...

In this paper, the problem of online distributed optimization is investigated, where sum locally dynamic cost functions considered to be strongly pseudoconvex. To address problem, we propose an algorithm based on auxiliary strategy. The involves each agent minimizing its own function subject a common convex set while exchanging local information with others under time-varying directed communication graph sequence. regret employed measure performance algorithm. Under mild conditions graph, it...

In this paper, we consider the flocking problem of multi-agent systems with multiple groups. First, some algorithms using local information are designed to divide agents into any pre-assigned number groups in fixed and switching heterogeneous networks, respectively. Based on algebraic graph theory Barbalat's lemma, convergence criteria established ensure velocity alignment cohesion each subgroup as well collision avoidance between whole group. Second, an algorithm for homogeneous networks is...

This paper introduces the notion of weak rigidity to characterize a framework by pairwise inner products interagent displacements. Compared distance-based rigidity, requires fewer constrained edges in graph determine geometric shape an arbitrarily dimensional space. A necessary and sufficient graphical condition for infinitesimal planar frameworks is derived. As application proposed theory, gradient-based control law nongradient-based are designed group single-integrator modeled agents...

Designing the optimal linear quadratic regulator (LQR) for a large-scale multiagent system is time consuming since it involves solving large-size matrix Riccati equation. The situation further exasperated when design needs to be done in model-free way using schemes such as reinforcement learning (RL). To reduce this computational complexity, we decompose LQR problem into multiple small-size problems. We consider objective function specified over an undirected graph, and cast decomposition...

This article studies angle-based sensor network localization (ASNL) in a plane, which is to determine locations of all sensors network, given partial (called anchors) and angle measurements obtained the local coordinate frame each sensor. First, it shown that framework with nondegenerate bilateration ordering must be fixable, implying can uniquely determined by angles between edges up translations, rotations, reflections, uniform scaling. Then, ASNL proved have unique solution if only...

This article investigates the fuel-optimal guidance problem of end-to-end human-Mars entry, powered-descent, and landing (EDL) mission. It applies a unified modeling scheme develops computationally efficient new optimization algorithm to solve multiphase optimal problem. The EDL is first modeled as control with different dynamics constraints at each phase. Via polynomial approximation discretization techniques, this then reformulated programming By introducing intermediate variables...

Recently introduced distributed zeroth-order optimization (ZOO) algorithms have shown their utility in reinforcement learning (RL). Unfortunately, the gradient estimation process, almost all of them require random samples with same dimension as global variable and/or evaluation cost function, which may induce high variance for large-scale networks. In this paper, we propose a novel algorithm by leveraging network structure inherent objective, allows each agent to estimate its local...

In this note, a distributed subgradient-based algorithm is proposed for continuous-time multi-agent systems to search feasible solution convex inequalities. The involves each agent achieving state constrained by its own inequalities while exchanging local information with other agents under time-varying directed communication graph. With the validity of mild connectivity condition associated graph, it shown that all will reach agreement asymptotically and consensus in set Furthermore, method...

We address model-free distributed stabilization of heterogeneous continuous-time linear multi-agent systems using reinforcement learning (RL). Two algorithms are developed. The first algorithm solves a centralized quadratic regulator (LQR) problem without knowing any initial stabilizing gain in advance. second builds upon the results algorithm, and extends it to with predefined interaction graphs. Rigorous proofs provided show that proposed achieve guaranteed convergence if specific...

This brief investigates distributed coordination control problems under a state-dependent communication graph with several inherent links. In the network, an interaction arises between two agents if either their states differ by less than fixed range or link exists them. By considering that each agent has gain, nonlinear consensus protocol is proposed for single-integrator modeled multi-agent systems. With validity of some initial condition, shown to be connected at any time, which...

This paper investigates the fuel-optimal guidance problem of end-to-end human-Mars entry, powered descent, and landing (EDL) mission. It proposes a computationally efficient modeling scheme new optimization algorithm to solve multi-phase optimal problem. We first model EDL as control with different dynamics constraints at each phase. Via polynomial approximation discretization techniques, this is then reformulated programming By introducing intermediate variables quadratic equality...

This letter proposes two reinforcement learning (RL) algorithms for solving a class of coupled algebraic Riccati equations (CARE) linear stochastic dynamic systems with unknown state and input matrices. The CARE are formulated minimal-cost variance (MCV) control problem that aims to minimize the cost function while keeping its mean at an acceptable range using noisy infinite-horizon full-state feedback quadratic regulator (LQR). We propose RL where matrix can be estimated very first...

Sensor network localization (SNL) is to determine physical coordinates of all sensors in a given global anchors and available measurements among anchors. Two challenges related SNL are find conditions leading uniquely localizable develop effective efficient methods solve problems. This work first proves that infinitesimal rigidity, together with some mild conditions, sufficient for unique localizability considering additional relationships between nonadjacent sensors. On the other hand,...

Existing distributed cooperative multi-agent reinforcement learning (MARL) frameworks usually assume undirected coordination graphs and communication graphs, while estimating a global reward via consensus algorithms for policy evaluation. Such framework may induce expensive costs exhibit poor scalability due to requirement of consensus. In this work, we study MARLs with directed propose RL algorithm where the local evaluations are based on value functions. The function each agent is obtained...

This paper studies angle-based sensor network localization (ASNL) in a plane, which is to determine locations of all sensors network, given partial (called anchors) and angle measurements obtained the local coordinate frame each sensor. Firstly it shown that framework with non-degenerate bilateration ordering must be fixable, implying can uniquely determined by angles between edges up translations, rotations, reflections uniform scaling. Then ASNL proved have unique solution if only grounded...

Individual agents in a multi-agent system (MAS) may have decoupled open-loop dynamics, but cooperative control objective usually results coupled closed-loop dynamics thereby making the design computationally expensive. The computation time becomes even higher when learning strategy such as reinforcement (RL) needs to be applied deal with situation are not known. To resolve this problem, we propose parallel RL scheme for linear quadratic regulator (LQR) continuous-time MAS. idea is exploit...

In this paper, we consider group flocking of multi-agent systems in which agents are dispersed to different subgroups. By using local information, the algorithms proposed solve problem heterogeneous and homogeneous networks, respectively. For each given algorithm, corresponding criterions established ensure both velocity alignment, cohesion subgroup, collision avoidance between any two whole group. The protocol can be modified realize division into pre-assigned number subgroups, while solves...

In this paper, a distributed subgradient-based algorithm is proposed for continuous-time multi-agent systems to search feasible solution convex inequalities. The involves each agent achieving state constrained by its own inequalities while exchanging local information with other agents under time-varying directed communication graph. With the validity of mild connectivity condition associated graph, it shown that all will reach agreement asymptotically and consensus in set Furthermore,...