Showing papers by "Swaroop Darbha published in 2015"

Output Regulation of Nonlinear Systems With Application to Roll-to-Roll Manufacturing Systems

Abstract: This paper deals with the problem of synthesizing feedforward control to aid the regulation of output of a nonlinear system in the presence of partially known exogenous inputs. The problem appears in many engineering applications including Roll-to-Roll (R2R) manufacturing systems. Currently known methods for this problem either require the solution of a constrained partial differential equation or the preview information of the signal to be tracked. The novelty of this paper lies in synthesizing feedforward control as the solution of a system of differential-algebraic equations, which is considerably less complex and suitable for practical implementation. In this paper, we consider the problem of regulating the output while rejecting the disturbances and apply it to R2R manufacturing systems. The problem of tracking reference signals can also be handled with the suggested technique. We assume that the disturbance signal is the output of a known exogenous system with unknown initial conditions. A parameter identification scheme to estimate the unknown initial conditions is developed. The proposed technique is successfully applied to control of web tension in a large R2R machine which mimics most of the features of industrial R2R machines and contains real-time hardware and software that is used in industrial practice. Extensive experiments were conducted to evaluate the proposed scheme for web tension control under various experimental conditions, including different web speeds and materials. We will present and discuss the representative experimental results with the proposed technique and provide a comparison with an industrial PI control scheme to highlight the benefits of using the proposed scheme.

Computation of Lower Bounds for a Multiple Depot, Multiple Vehicle Routing Problem With Motion Constraints

Abstract: This paper considers the problem of planning paths for a collection of identical vehicles visiting a given set of targets, such that the total lengths of their paths are minimum. Each vehicle starts at a specified location (called a depot) and it is required that each target to be on the path of at least one vehicle. The path of every vehicle must satisfy the motion constraints of every vehicle. In this paper, we develop a method to compute lower bound to the minimum total path lengths by relaxing some of the constraints and posing it as a standard multiple traveling salesmen problem (MTSP). A lower bound is often important to ascertain suboptimality bounds for heuristics and for developing stopping criterion for algorithms computing an optimal solution. Simulation results are presented to show that the proposed method can be used to improve the lower bounds of instances with four vehicles and 40 targets by approximately 39%.

On maximizing algebraic connectivity of networks for various engineering applications

Abstract: We discuss a simplified version of an open problem in system realization theory which has several important practical applications in complex networks. Particularly, we focus on algebraic connectivity (λ 2 ) of the Laplacian of the network as an objective for maximization. We show that the maximum value of forced response of a linear mechanical (spring-mass) system can be minimized when the λ 2 of the corresponding stiffness matrix is maximized. In the case of motion control problems related to vehicle localization with noisy measurements, we show that the λ 2 plays a vital role to control their positions towards a desired formation. In UAV adhoc infrastructure networks, we show that the λ 2 of the information flow graph governs the stability of the rigid formation with respect to disturbance attenuation. In the context of UAV adhoc communication networks, we also provide a physical interpretation for the maximization of λ 2 via the closely related Cheeger constant or the isoperimetric number. We further discuss a Fiedler vector based mixed-integer linear programing formulation for the problem of maximizing λ 2 and obtain optimal solutions and upper bounds based on cutting plane methods. Computational results corroborate the performance of proposed algorithms.

Synthesizing Robust Communication Networks for Unmanned Aerial Vehicles With Resource Constraints

Abstract: In this article, we address the problem of synthesizing communication networks for unmanned aerial vehicles (UAVs) in the presence of resource constraints. UAVs can be deployed as backbone nodes in ad hoc networks that can be central to civilian and military applications. The cost of operation of the network depends on the resources that are used such as the total power consumption associated with the network and the number of communication links in the network. The objective of the problem is to synthesize a communication network that maximizes connectivity subject to the cost of operation being within the specified budget for the resources. It is known that algebraic connectivity is a measure of robust connectivity and hence, it is chosen as an objective for optimization. We pose the network synthesis problem as a mixed-integer semidefinite program (MISDP): (1) provide an algorithm for computing optimal solutions using cutting plane methods; (2) develop lower bounds by posing the problem as a binary semidefinite program; and (3) construct feasible solutions using heuristics and estimate their quality. The network synthesis problem is a nondeterministic polynomial--time (NP)-hard problem. We provide some computational results to corroborate the performance of the proposed algorithms.

Web tension regulation with partially known periodic disturbances in roll-to-roll manufacturing systems

Abstract: This paper deals with the problem of synthesizing feedforward control to aid the regulation of web tension in the presence of partially known exogenous inputs. The problem appears in many engineering applications including Roll-to-Roll (R2R) manufacturing systems where the governing equation for tension is nonlinear. Currently known methods for the nonlinear output regulation problem either require the solution of a constrained partial differential equation or the preview information of the signal to be tracked. In this paper, we consider the problem of regulating web tension while rejecting periodic disturbances and use a novel approach to synthesize feedforward control as the solution of a system of differential-algebraic equations, which is considerably less complex and suitable for practical implementation. We assume that the disturbance signal is the output of a known exogenous system with unknown initial conditions. A parameter identification scheme to estimate the unknown initial conditions is developed. The proposed technique is successfully applied to web tension regulation in a large R2R machine which contains real-time hardware and software that is used in industrial practice. Extensive experiments were conducted to evaluate the proposed scheme under various experimental conditions, including different web speeds and materials. We will discuss a representative sample of the results with the proposed nonlinear tension regulator and provide a comparison with a well-tuned industrial PI control scheme to highlight the benefits of using the proposed scheme.

Estimation of Location and Orientation From Range Measurements

Abstract: Localization is an important required task for enabling vehicle autonomy. Localization entails the determination of the position of the center of mass and orientation of a vehicle from the available measurements. In this paper, we focus on localization by using the range measurements available to a vehicle from the communication of its multiple onboard receivers with roadside beacons. The model proposed for measurement is as follows: the true distance between a receiver and a beacon is at most equal to a predetermined function of the range measurement. The proposed procedure for localization is as follows: Based on the range measurements specific to a receiver from the beacons, a finite LP (linear programming) is proposed to estimate the location of the receiver. The estimate is essentially the Chebychev center of the set of possible locations of the receiver. In the second step, the location estimates of the vehicle are corrected using rigid body motion constraints and the orientation of the rigid body is thus determined. Two numerical examples provided at the end corroborate the procedures developed in this paper.© 2015 ASME