Beshah Ayalew | Clemson University (original) (raw)

Papers by Beshah Ayalew

IEEE Transactions on Vehicular Technology, Oct 1, 2019

An autonomous V2V communication mode (also known as side-link mode 4), which facilitates V2V comm... more An autonomous V2V communication mode (also known as side-link mode 4), which facilitates V2V communication in out of eNB coverage areas, has recently been introduced into the Long term evolution (LTE) standard. Recent research has studied the performance of this LTE-V2V autonomous mode for a highway use case. However, performance analysis for a highway use case cannot be easily applied to an intersection use case as it may contain non-line-of-sight (NLOS) communication links. In this paper, we analyze and evaluate the safety message broadcasting performance of LTE-V2V autonomous mode in an urban intersection scenario. Considering practical path loss models, we present the impact of NLOS communication link on the overall message dissemination performance. Through the analytical and simulation results, we show that the overall message dissemination performance degrades drastically with increasing vehicle density and increasing distance of the transmitting vehicle from the intersection. To improve the performance, we propose a vehicle-assisted relaying scheme in which the relaying vehicle is selected in an autonomous manner. We also present two resource allocation strategies for the relaying vehicle. For low to medium vehicle density along the street, we observe significant improvement in message dissemination through relaying compared to the scheme without relaying.

SAE International journal of passenger cars, Apr 5, 2016

This paper discusses a steering rack force estimation scheme using test-rig generated models. In ... more This paper discusses a steering rack force estimation scheme using test-rig generated models. In addition to friction identification, a model of the electric power steering system is identified by the use of the instrumented test-rig. It turns out that the friction in the steering system is highly load-dependent, asymmetric with respect to speed, and shows no Stribeck effects. A LuGre model is adopted and fitted to approximate the measured dynamic friction. Consequently, this model is used in a friction compensator which is combined with a linear disturbance observer to estimate the steering rack force. The proposed estimation scheme is analyzed via evaluated system simulations and experiments on the steering system test-rig. Finally, considering the fact that the friction level varies with each steering device manufactured and installed, the paper discusses algorithms for friction level adaptation.

Journal of Manufacturing Science and Engineering-transactions of The Asme, Oct 1, 2015

This paper deals with the systematic optimization method for multiple input variables (laser irra... more This paper deals with the systematic optimization method for multiple input variables (laser irradiation power and scanning speed) in a class of laser-aided powder deposition (LAPD) processes. These processes are normally described by a coupled system of nonlinear partial differential equations (PDEs). To begin with, a desired solid–liquid (S/L) interface geometry is first approximated from a few practical process target parameters that define the desired process properties. Then, the control problem is formulated as one of seeking the optimal combination of process inputs that achieves close tracking of the desired S/L interface in quasi-steady state. The paper details the derivation of the adjoint-based solution for this PDE-constrained multivariable control input optimization problem. The effectiveness of the proposed method is illustrated via a case study on a laser cladding process.

International Journal of Mechanical Sciences, Oct 1, 2015

Abstract A thick ring on a unilateral elastic foundation can be used to model important applicati... more Abstract A thick ring on a unilateral elastic foundation can be used to model important applications such as non-pneumatic tires or bushing bearings. This paper presents a reduced-order compensation scheme for computing the static deformation response of a thick ring supported by a unilateral elastic foundation to an arbitrary applied force. The ring considered is an orthotropic and extensible ring that can be treated as a Timoshenko beam. The elastic foundation is a two-parameter foundation with a linear torsional stiffness but a unilateral radial stiffness whose value vanishes when compressed or tensioned. The paper first derives the deformation response for the linear foundation case for which Fourier expansion techniques can be applied to obtain an analytical solution. Then, the nonlinear unilateral foundation problem is solved via an iterative compensation scheme that identifies regions with vanishing radial stiffness and applies a compensation force to the linear foundation model to counteract the excessive foundation forces that would not be there with the unilateral foundation. This scheme avoids the need for solving the complex set of nonlinear differential equations and gives a computationally efficient tool for rapidly analyzing and designing such systems. Representative results are compared with Finite Element Analysis (FEA) results to illustrate the validity of the proposed approach.

This paper presents a systematic control inputs optimization method for melting interface trackin... more This paper presents a systematic control inputs optimization method for melting interface tracking in laser-aided powder deposition (LAPD) processes. Using a proposed interface approximation and a coordinate system moving with the laser source, and adopting the enhanced thermal conductivity method, the process model is first reduced to a set of coupled partial differential equations (PDEs) in fixed spatial domains. Then, the control problem of achieving process target properties is formulated as one of optimizing the control inputs to track a prescribed melting interface which is approximated from required process target parameters. This proposed optimization scheme is solved by the adjoint-based gradient method for which an algorithm is provided. A weighting scheme is also proposed to overcome feasibility issues with poor interface specifications and still achieve improved tracking of target parameters. The proposed scheme is illustrated through a simulation-based case study on a laser cladding process.

In this paper, a scheme for estimation and control of moisture content in infrared (IR) drying of... more In this paper, a scheme for estimation and control of moisture content in infrared (IR) drying of paints/coatings is proposed. To deal with the infinite-dimensional nature of the process model associated with the moisture diffusion in the coating film, POD-Galerkin method is first applied for model reduction. Then, an unscented Kalman filter (UKF) is devised for distributed moisture content estimation and nonlinear model predictive control (NMPC) system is designed for tracking a desired average moisture content profile with optimized energy needs. To enhance the control performance in the presence of potential modeling uncertainties, a robust design is also included in the proposed NMPC scheme. The effectiveness of this approach is demonstrated via simulated applications to IR drying of automotive waterborne coatings.Copyright © 2015 by ASME

In this paper a Model Predictive Control (MPC) strategy is utilized to model a professional drive... more In this paper a Model Predictive Control (MPC) strategy is utilized to model a professional driver negotiating a set driving circuit in minimum time. MPC is inherently suboptimal because not all future information is incorporated into each horizon of the control scheme. Motivated by how professional drivers learn race circuits in order to best exploit its features, we will alleviate some of the suboptimality inherent to MPC by optimizing the local cost function of each MPC horizon. This will allows objectives over a local segment to be properly adjusted such that the global goal of minimizing maneuvering time over a full maneuver is approximated. This problem is solved utilizing a cascaded optimization structure with the inner loop recursively solving the MPC problem around the track and the outer loop optimizing the weights of the local MPC horizons. It will be shown that by varying weights at key locations on a particular maneuver, performance gains can be realized compared to a traditional time optimal MPC strategy.

2020 IEEE Conference on Control Technology and Applications (CCTA), Aug 1, 2020

A commonly proposed method for improving traffic flow on freeways is speed harmonization. The eff... more A commonly proposed method for improving traffic flow on freeways is speed harmonization. The effectiveness of current speed harmonization approaches, such as variable speed limits, is extremely reliant on human driver compliance. Connected and automated vehicles (CAVs) are expected to come to market within this decade, offering the opportunity to eliminate or reduce the reliance on human compliance. However, extending current roadside infrastructure-based approaches of assigning centrally computed harmonization speeds to individual vehicles is, costly. An alternative approach is to have individual vehicles estimate the traffic state on-board and make distributed decisions to achieve the harmonization goal autonomously. In this work, we present a distributed algorithm for estimating the current average speed of traffic. We couple this with a distributed 2D maneuver planning approach. Then, we study the impact on traffic efficiency in terms of energy consumption and travel time at varying CAV penetration rates.

Iet Intelligent Transport Systems, Dec 1, 2020

Engineering Applications of Artificial Intelligence

SAE International Journal of Commercial Vehicles

Certain advanced driver assistance systems (ADAS) have the potential to boost energy efficiency i... more Certain advanced driver assistance systems (ADAS) have the potential to boost energy efficiency in real-world scenarios. This article details a radar-based driver assistance scheme designed to minimize fuel consumption for a commercial vehicle by predictively optimizing braking and driving torque inputs while accommodating the driver’s demand. The workings of the proposed scheme are then assessed with a novel integration of the driver assistance functionality in randomized traffic microsimulation. Although standardized test procedures are intended to mimic urban and highway speed profiles for the purposes of evaluating fuel economy and emissions, they do not explicitly consider the interactions present in real-world driving between the ego vehicle equipped with ADAS and other vehicles in traffic. This article presents one approach to address the drawback of standardized test procedures for evaluating the fuel economy benefits of ADAS technologies. This approach is demonstrated by us...

2020 IEEE Conference on Control Technology and Applications (CCTA)

Predictive or look-ahead strategies that attempt to incorporate upcoming environmental (e.g. road... more Predictive or look-ahead strategies that attempt to incorporate upcoming environmental (e.g. road topography, visibility) and traffic state (e.g. preceding vehicles, speed limits, traffic signal state/schedule) information for the purposes of optimizing vehicular energy consumption have been attracting a lot of attention in the past decade. In this work, we propose and evaluate a radar-based predictive kinetic energy management (PKEM) scheme that is applicable as an add-on driver assistance module for a large electric vehicle. This paper details our approach to modeling each subsystem of the framework including the interacting multiple model radar filter, the model predictive controller, and the powertrain. We found that there are clear energy saving benefits for the PKEM scheme with minimal compromises on travel time. These benefits can be in the order of 10–12% over the baseline driver-only case in urban environments and are negligible on the highway cycle. Results included demonstrate the potential trade-offs and accommodations of driver desired inputs.

University of Texas at Austin, 2019

In this paper the nonlinear model predictive control (NMPC) of UV-induced curing of composite mat... more In this paper the nonlinear model predictive control (NMPC) of UV-induced curing of composite material for manufacturing of thick parts is proposed. The process involves layer-by-layer curing of thin composite laminates to form thick part. The model for NMPC switches when a new layer is added to the existing layer. The layer addition times are determined externally. The offline optimal control is used to determine the optimal time and temperature profile which will give uniform cure distribution of a thick composite material. Once the temperature trajectory and optimal time sequences are found, the NMPC is implemented for online control. The objective is to determine theoretical optimal behavior (assuming the process measurement is available) which will be used for online switching NMPC for tracking the reference temperature.