arXiv subjects

Magnus Egerstedt

Publications and source records attributed to Magnus Egerstedt.

88 records · Page 5Linked to original sources

arXiv2016

Graph Distances and Controllability of Networks

In this technical note, we study the controllability of diffusively coupled networks from a graph theoretic perspective. We consider leader-follower networks, where the external control inputs are injected to only some of the agents, namely the leaders. Our main result relates the controllability of such systems to the graph distances between the agents. More specifically, we present a graph topological lower bound on the rank of the controllability matrix. This lower bound is tight, and it is applicable to systems with arbitrary network topologies, coupling weights, and number of leaders. An algorithm for computing the lower bound is also provided. Furthermore, as a prominent application, we present how the proposed bound can be utilized to select a minimal set of leaders for achieving controllability, even when the coupling weights are unknown.

eess.SY↗

arXiv2016

Cloud-Enabled Differentially Private Multi-Agent Optimization with Constraints

We present an optimization framework for solving multi-agent nonlinear programs subject to inequality constraints while keeping the agents' state trajectories private. Each agent has an objective function depending only upon its own state and the agents are collectively subject to global constraints. The agents do not directly communicate with each other but instead route messages through a trusted cloud computer. The cloud computer adds noise to data being sent to the agents in accordance with the framework of differential privacy in order to keep each agent's state trajectory private from all other agents and any eavesdroppers. This private problem can be viewed as a stochastic variational inequality and is solved using a projection-based method for solving variational inequalities that resembles a noisy primal-dual gradient algorithm. Convergence of the optimization algorithm in the presence of noise is proven and a quantifiable trade-off between privacy and convergence is extracted from this proof. Simulation results are provided that demonstrate numerical convergence for both $ε$-differential privacy and $(ε, δ)$-differential privacy.

math.OC↗

arXiv2016

Safe, Remote-Access Swarm Robotics Research on the Robotarium

This paper describes the development of the Robotarium -- a remotely accessible, multi-robot research facility. The impetus behind the Robotarium is that multi-robot testbeds constitute an integral and essential part of the multi-agent research cycle, yet they are expensive, complex, and time-consuming to develop, operate, and maintain. These resource constraints, in turn, limit access for large groups of researchers and students, which is what the Robotarium is remedying by providing users with remote access to a state-of-the-art multi-robot test facility. This paper details the design and operation of the Robotarium as well as connects these to the particular considerations one must take when making complex hardware remotely accessible. In particular, safety must be built in already at the design phase without overly constraining which coordinated control programs the users can upload and execute, which calls for minimally invasive safety routines with provable performance guarantees.

cs.RO↗

arXiv2015

Inter-Robot Interactions in Multi-Robot Systems Using Braids

This paper describes a framework for multi-robot coordination and motion planning with emphasis on inter-agent interactions. We focus on the characterization of inter-agent interactions with sufficient level of abstraction so as to allow for the enforcement of desired interaction patterns in a provably safe (i.e., collision-free) manner, e.g., for achieving rich movement patterns in a shared space, or to exchange sensor information. We propose to specify interaction patterns through elements of the so-called braid group. This allows us to not focus on a particular pattern per se, but rather on the problem of being able to execute a whole class of patterns. The result from such a construction is a hybrid system driven by symbolic inputs that must be mapped onto actual paths that both realize the desired interaction levels and remain safe in the sense that collisions are avoided.

cs.MA↗

arXiv2015

A Game-theoretic Formulation of the Homogeneous Self-Reconfiguration Problem

In this paper we formulate the homogeneous two- and three-dimensional self-reconfiguration problem over discrete grids as a constrained potential game. We develop a game-theoretic learning algorithm based on the Metropolis-Hastings algorithm that solves the self-reconfiguration problem in a globally optimal fashion. Both a centralized and a fully distributed algorithm are presented and we show that the only stochastically stable state is the potential function maximizer, i.e. the desired target configuration. These algorithms compute transition probabilities in such a way that even though each agent acts in a self-interested way, the overall collective goal of self-reconfiguration is achieved. Simulation results confirm the feasibility of our approach and show convergence to desired target configurations.

cs.MA↗

arXiv2015

Cloud-Based Centralized/Decentralized Multi-Agent Optimization with Communication Delays

We present and analyze a computational hybrid architecture for performing multi-agent optimization. The optimization problems under consideration have convex objective and constraint functions with mild smoothness conditions imposed on them. For such problems, we provide a primal-dual algorithm implemented in the hybrid architecture, which consists of a decentralized network of agents into which centralized information is occasionally injected, and we establish its convergence properties. To accomplish this, a central cloud computer aggregates global information, carries out computations of the dual variables based on this information, and then distributes the updated dual variables to the agents. The agents update their (primal) state variables and also communicate among themselves with each agent sharing and receiving state information with some number of its neighbors. Throughout, communications with the cloud are not assumed to be synchronous or instantaneous, and communication delays are explicitly accounted for in the modeling and analysis of the system. Experimental results are presented to support the theoretical developments made.

math.OC↗

arXiv2015

Communication-Free Distributed Coverage for Networked Systems

In this paper, we present a communication-free algorithm for distributed coverage of an arbitrary network by a group of mobile agents with local sensing capabilities. The network is represented as a graph, and the agents are arbitrarily deployed on some nodes of the graph. Any node of the graph is covered if it is within the sensing range of at least one agent. The agents are mobile devices that aim to explore the graph and to optimize their locations in a decentralized fashion by relying only on their sensory inputs. We formulate this problem in a game theoretic setting and propose a communication-free learning algorithm for maximizing the coverage.

eess.SY↗

arXiv2015

Formation of Robust Multi-Agent Networks Through Self-Organizing Random Regular Graphs

Multi-agent networks are often modeled as interaction graphs, where the nodes represent the agents and the edges denote some direct interactions. The robustness of a multi-agent network to perturbations such as failures, noise, or malicious attacks largely depends on the corresponding graph. In many applications, networks are desired to have well-connected interaction graphs with relatively small number of links. One family of such graphs is the random regular graphs. In this paper, we present a decentralized scheme for transforming any connected interaction graph with a possibly non-integer average degree of k into a connected random m-regular graph for some m in [k, k + 2]. Accordingly, the agents improve the robustness of the network with a minimal change in the overall sparsity by optimizing the graph connectivity through the proposed local operations.

cs.MA↗

arXiv2014

Deploying robots with two sensors in $K_{1,6}$-free graphs

Let $G$ be a graph of minimum degree at least two with no induced subgraph isomorphic to $K_{1,6}$. We prove that if $G$ is not isomorphic to one of eight exceptional graphs, then it is possible to assign two-element subsets of $\{1,2,3,4,5\}$ to the vertices of $G$ in such a way that for every $i\in\{1,2,3,4,5\}$ and every vertex $v\in V(G)$ the label $i$ is assigned to $v$ or one of its neighbors. It follows that $G$ has fractional domatic number at least $5/2$. This is motivated by a problem in robotics and generalizes a result of Fujita, Yamashita and Kameda who proved that the same conclusion holds for all $3$-regular graphs.

math.CO↗

arXiv2014

Cloud-Based Optimization: A Quasi-Decentralized Approach to Multi-Agent Coordination

New architectures and algorithms are needed to reflect the mixture of local and global information that is available as multi-agent systems connect over the cloud. We present a novel architecture for multi-agent coordination where the cloud is assumed to be able to gather information from all agents, perform centralized computations, and disseminate the results in an intermittent manner. This architecture is used to solve a multi-agent optimization problem in which each agent has a local objective function unknown to the other agents and in which the agents are collectively subject to global inequality constraints. Leveraging the cloud, a dual problem is formulated and solved by finding a saddle point of the associated Lagrangian.

math.OC↗

arXiv2014

Multi-Robot Control Using Time-Varying Density Functions

This paper presents an approach to externally influencing a team of robots by means of time-varying density functions. These density functions represent rough references for where the robots should be located. To this end, a continuous-time algorithm is proposed that moves the robots so as to provide optimal coverage given the density functions as they evolve over time. The developed algorithm represents an extension to previous coverage algorithms in that time-varying densities are explicitly taken into account in a provable manner. A distributed approximation to this algorithm is moreover proposed whereby the robots only need to access information from adjacent robots. Simulations and robotic experiments show that the proposed algorithms do indeed exhibit the desired behaviors in practice as well as in theory.

math.OC↗

arXiv2014

Continuous-time Proportional-Integral Distributed Optimization for Networked Systems

In this paper we explore the relationship between dual decomposition and the consensus-based method for distributed optimization. The relationship is developed by examining the similarities between the two approaches and their relationship to gradient-based constrained optimization. By formulating each algorithm in continuous-time, it is seen that both approaches use a gradient method for optimization with one using a proportional control term and the other using an integral control term to drive the system to the constraint set. Therefore, a significant contribution of this paper is to combine these methods to develop a continuous-time proportional-integral distributed optimization method. Furthermore, we establish convergence using Lyapunov stability techniques and utilizing properties from the network structure of the multi-agent system.

eess.SY↗

arXiv2013

From Global Linear Computations to Local Interaction Rules

A network of locally interacting agents can be thought of as performing a distributed computation. But not all computations can be faithfully distributed. This paper investigates which global, linear transformations can be computed using local rules, i.e., rules which rely solely on information from adjacent nodes in a network. The main result states that a linear transformation is computable in finite time using local rules if and only if the transformation has positive determinant. An optimal control problem is solved for finding the local interaction rules, and simulations are performed to elucidate how optimal solutions can be obtained.

math.OC↗

arXiv2013

Distributed Control of Multiagent Systems under Unknown Persistent Disturbances

This paper focuses on the consensus and formation problems of multiagent systems under unknown persistent disturbances. Specifically, we propose a novel method that combines an existing consensus (or formation) algorithm with a new controller. The new controller has an integral action that produces a control input based on an error signal locally projected onto the column space of the graph Laplacian. This action allows agents to achieve a consensus or a predetermined formation objective under constant or time-varying disturbances. This study has analyzed the stability properties of this architecture and provides several illustrative examples to demonstrate the proposed approach.

math.OC↗

arXiv2012

A Tight Lower Bound on the Controllability of Networks with Multiple Leaders

In this paper we study the controllability of networked systems with static network topologies using tools from algebraic graph theory. Each agent in the network acts in a decentralized fashion by updating its state in accordance with a nearest-neighbor averaging rule, known as the consensus dynamics. In order to control the system, external control inputs are injected into the so called leader nodes, and the influence is propagated throughout the network. Our main result is a tight topological lower bound on the rank of the controllability matrix for such systems with arbitrary network topologies and possibly multiple leaders.

eess.SY↗

arXiv2011

Algorithm for Optimal Mode Scheduling in Switched Systems

This paper considers the problem of computing the schedule of modes in a switched dynamical system, that minimizes a cost functional defined on the trajectory of the system's continuous state variable. A recent approach to such optimal control problems consists of algorithms that alternate between computing the optimal switching times between modes in a given sequence, and updating the mode-sequence by inserting to it a finite number of new modes. These algorithms have an inherent inefficiency due to their sparse update of the mode-sequences, while spending most of the computing times on optimizing with respect to the switching times for a given mode-sequence. This paper proposes an algorithm that operates directly in the schedule space without resorting to the timing optimization problem. It is based on the Armijo step size along certain Gateaux derivatives of the performance functional, thereby avoiding some of the computational difficulties associated with discrete scheduling parameters. Its convergence to local minima as well as its rate of convergence are proved, and a simulation example on a nonlinear system exhibits quite a fast convergence.

eess.SY↗