Matin Jafarian

Formation control using binary information

In this paper, we study the problem of formation keeping of a network of strictly passive systems when very coarse information is exchanged. We assume that neighboring agents only know whether their relative position is larger or smaller than the prescribed one. This assumption results in very simple control laws that direct the agents closer or away from each other and take values in finite sets. We show that the task of formation keeping while tracking a desired velocity and rejecting matched disturbances is still achievable under the very coarse information scenario. In contrast with other results of practical convergence with coarse or quantized information, here the control task is achieved exactly.Distributed formation keeping control is a motion coordination problem which aims at achieving a desired geometrical shape for the positions of a group of agents (e.g. robots). In problems of formation control, an important component is the flow of information among the agents. Although the usual assumption in the literature is the exchange of perfect information among the agents, the latter might be a restrictive requirement due to real-world constraints. To cope with this restriction, quantized information and control have been proposed and studied in the literature. In particular, there has been a growing interest in binary quantizers and controllers owing to the recent developments in cyber-physical systems. This thesis is mainly focused on the problem of distributed position-based formation keeping of a group of continuous-time dynamic agents using binary controllers. The binary information and control models a sensing scenario in which each agent detects whether or not the components of its current distance vector from a neighbor are above or below the prescribed distance and applies a force (in which each component takes a binary value) to reduce or increase the actual distance. In this context, we consider different classes of dynamical agents, including strict output passive systems, unicycles, and nonholonomic wheeled carts. For the control design and analysis, we use tools from discontinuous dynamical systems, passivity, hybrid dynamical systems, graph theory and internal-model-based approach.

Exact formation control with very coarse information

This paper investigates a formation control problem for agents modeled as double integrators when very coarse information is exchanged. We assume that neighboring agents only know whether their relative position is larger or smaller than the prescribed one. The use of this binary information results in very simple control inputs that direct the agents closer or away from each other and take values in finite sets. We also show that the task of keeping a formation and tracking a reference velocity which is only known to the formations leader is still achievable under the very coarse information scenario that we consider. In contrast with the other results of practical convergence with coarse or quantized information, here the control task is achieved exactly.

Formation control of a multi-agent system subject to Coulomb friction

This paper considers the formation control problem for a network of point masses which are subject to Coulomb friction. A dynamical model including the planar discontinuous friction force is presented in the port-Hamiltonian framework. Moreover, continuous and discontinuous controllers are designed in order to achieve a desired prescribed formation. The main results are derived using tools from nonsmooth Lyapunov analysis. It is shown that the continuous static feedback controller fails to achieve the exact formation, while the discontinuous controller achieves the desired task exactly. Numerical simulations are provided to illustrate the effectiveness of the approach.

Performance Comparison of a Planar Bipedal Robot with Rigid and Compliant Legs

The purpose of this work is to study the effect of placing passive storage elements (springs) along the robot legs on its performance. We first present the model of a planar passive dynamic walker with compliant ground contact model, then replace its rigid legs with compliant legs. Simulation results validate the effectiveness of compliant legs in improving the performance (robustness and velocity) of the robot.

A price-based approach for voltage regulation and power loss minimization in power distribution networks

We present a price-based approach to deal with the challenges of the electrical power distribution systems with renewable generations. In specific, we address the power loss minimization and voltage regulation taking into account the actual grid capacity. Analogously, the cost function is reformulated to represent a social welfare maximization problem. Distributed optimization formulations, which bridge the physical power grid to the market-based approach, are presented and analyzed.

Ternary and hybrid controllers for the rendezvous of unicycles

This paper presents the rendezvous (consensus) of the orientations and average positions of a group of unicycles. We assume that the decentralized controllers designed for consensus of the average positions take only values in the set {-1; 0; +1}. In addition, we introduce a hybrid controller to reach the consensus of the orientations. Finite-time practical consensus of the average positions is proven despite the simple ternary control laws together with asymptotic consensus of the orientations.

Coordination with binary controllers Formation control and disturbance rejection

In this paper, we study the problem of formation keeping of a network of strictly passive systems when very coarse information is exchanged. We assume that neighboring agents only know whether their relative position is larger or smaller than the prescribed one. This assumption results in very simple control laws that direct the agents closer or away from each other and take values in finite sets. We show that the task of formation keeping while tracking a desired velocity and rejecting matched disturbances is still achievable under the very coarse information scenario. In contrast with other results of practical convergence with coarse or quantized information, here the control task is achieved exactly.Distributed formation keeping control is a motion coordination problem which aims at achieving a desired geometrical shape for the positions of a group of agents (e.g. robots). In problems of formation control, an important component is the flow of information among the agents. Although the usual assumption in the literature is the exchange of perfect information among the agents, the latter might be a restrictive requirement due to real-world constraints. To cope with this restriction, quantized information and control have been proposed and studied in the literature. In particular, there has been a growing interest in binary quantizers and controllers owing to the recent developments in cyber-physical systems. This thesis is mainly focused on the problem of distributed position-based formation keeping of a group of continuous-time dynamic agents using binary controllers. The binary information and control models a sensing scenario in which each agent detects whether or not the components of its current distance vector from a neighbor are above or below the prescribed distance and applies a force (in which each component takes a binary value) to reduce or increase the actual distance. In this context, we consider different classes of dynamical agents, including strict output passive systems, unicycles, and nonholonomic wheeled carts. For the control design and analysis, we use tools from discontinuous dynamical systems, passivity, hybrid dynamical systems, graph theory and internal-model-based approach.

Coordination with binary controllers

In this paper, we study the problem of formation keeping of a network of strictly passive systems when very coarse information is exchanged. We assume that neighboring agents only know whether their relative position is larger or smaller than the prescribed one. This assumption results in very simple control laws that direct the agents closer or away from each other and take values in finite sets. We show that the task of formation keeping while tracking a desired velocity and rejecting matched disturbances is still achievable under the very coarse information scenario. In contrast with other results of practical convergence with coarse or quantized information, here the control task is achieved exactly.Distributed formation keeping control is a motion coordination problem which aims at achieving a desired geometrical shape for the positions of a group of agents (e.g. robots). In problems of formation control, an important component is the flow of information among the agents. Although the usual assumption in the literature is the exchange of perfect information among the agents, the latter might be a restrictive requirement due to real-world constraints. To cope with this restriction, quantized information and control have been proposed and studied in the literature. In particular, there has been a growing interest in binary quantizers and controllers owing to the recent developments in cyber-physical systems. This thesis is mainly focused on the problem of distributed position-based formation keeping of a group of continuous-time dynamic agents using binary controllers. The binary information and control models a sensing scenario in which each agent detects whether or not the components of its current distance vector from a neighbor are above or below the prescribed distance and applies a force (in which each component takes a binary value) to reduce or increase the actual distance. In this context, we consider different classes of dynamical agents, including strict output passive systems, unicycles, and nonholonomic wheeled carts. For the control design and analysis, we use tools from discontinuous dynamical systems, passivity, hybrid dynamical systems, graph theory and internal-model-based approach.