Path optimization for marine vehicles in ocean currents using reinforcement learning

This study proposes a path planning algorithm for marine vehicles based on machine learning. The algorithm considers the dynamic characteristics of the vehicle and disturbance effects in ocean environments. The movements of marine vehicles are influenced by various physical disturbances in ocean environments, such as wind, waves, and currents. In the present study, the effects of ocean currents are the primary consideration. A kinematic model is used to incorporate the nonholonomic motion characteristics of a marine vehicle, and the reinforcement learning algorithm is used for path optimization to generate a feasible path that can be tracked by the vehicle. The proposed approach determines a near-optimal path that connects the start and goal points with a reasonable computational cost when the map and current field data are provided. To verify the optimality and validity of the proposed algorithm, a set of simulations were performed in simulated and actual ocean current conditions, and their results are presented.     

Development of realistic shortest path algorithm considering lane changes

Lane changes occur as many times as turning movements are needed while following a designated path. The cost of a route with many lane changes is likely to be more expensive than that with less lane changes, and unrealistic paths with impractical lane changes should be avoided for drivers' safety. In this regard, a new algorithm is developed in this study to find the realistic shortest path considering lane changing. The proposed algorithm is a modified link‐labeling Dijkstra algorithm considering the effective lane‐changing time that is a parametric function of the prevailing travel speed and traffic density. The parameters were estimated using microscopic traffic simulation data, and the numerical test demonstrated the performance of the proposed algorithm. It was found that the magnitude of the effect of the effective lane‐changing time on determining the realistic shortest path is nontrivial, and the proposed algorithm has capability to exclude links successfully where the required lane changes are practically impossible. Copyright © 2015 John Wiley & Sons, Ltd.     

Estimating vehicle miles traveled (VMT) in urban areas using regression kriging

The recent increase in demand for performance‐driven and outcome‐based transportation planning makes accurate and reliable performance measures essential. Vehicle miles traveled (VMT), the total miles traveled by all vehicles on roadways, has been utilized widely as a proxy for traffic impact assessment, vehicle emissions, gasoline consumption, and crashes. Accordingly, a number of studies estimate VMT using diverse data sources. This study estimates VMT in the urban area of Bucheon, South Korea, by predicting the annual average daily traffic for unmeasured locations using spatial interpolation techniques (i.e., regression kriging and linear regression). The predictive performance of this method is compared with that of the existing Highway Performance Monitoring System (HPMS) method. The results show that regression kriging could provide more accurate VMT estimates than the HPMS method and linear regression, especially with a small sample size. Copyright © 2016 John Wiley & Sons, Ltd.     

Computational study on the effects of volume ratio of DOC/DPF and catalyst loading on the PM and NOx emission control for heavy-duty diesel engines

The use of a diesel particulate filter (DPF) in a diesel aftertreatment system has proven to be an effective and efficient method for removing particulate matter (PM) in order to meet more stringent emission regulations without hurting engine performance. One of the favorable PM regeneration technologies is the NO₂-assisted regeneration method due to the capability of continuous regeneration of PM under a much lower temperature than that of thermal regeneration. In the present study, the thermal behavior of the monolith during regeneration and the conversion efficiency of NO₂ from NO with an integrated exhaust system of a diesel oxidation catalyst (DOC) and DPF have been predicted by one-channel numerical simulation. The simulation results of the DOC, DPF, and integrated DOC-DPF models are compared with experimental data to verify the accuracy of the present model for the integrated DOC and DPF modeling. The effects of catalyst loading inside the DOC and the volume ratio between the DOC and DPF on the pressure drop, the conversion efficiency, and the oxidation rate of PM, have been numerically investigated. The results indicate that the case of the volume ratio of ‘DOC/DPF=1.5’ within the same diameter of both monoliths produced close to the maximum conversion efficiency and oxidation rate of PM. Under the engine operating condition of 175 kW at 2200 rpm, 100% load with a displacement of 8.1, approximately 55 g/ft³ of catalyst (Pt) loading inside the DOC with the active Pt surface of 5.3 m²/g_pt was enough to maximize the conversion efficiency and oxidation rate of PM.     

CVT ratio control with consideration of CVT system loss

A modified CVT ratio map is proposed to obtain the improved fuel economy for a metal belt CVT. Since the CVT system loss, which occupies most of the drivetrain loss, depends on the engine speed, input torque, primary and secondary actuator pressure, a modified CVT ratio map is produced to realize the highest engine-CVT overall efficiency through the consideration of CVT system loss. The modified CVT ratio map is constructed with respect to the demanded vehicle power and present vehicle speed based on the steady state CVT system loss. Using the modified CVT ratio map, performance simulations are carried out using the dynamic models of the CVT powertrain. The simulation results indicate that the modified CVT ratio control provides improved engine-CVT overall system efficiency, and improves the fuel economy of the federal urban driving schedule by 4.9 percent.     

Robust <Emphasis Type="Italic">H</Emphasis><Subscript>∞</Subscript> preview control of an active suspension system with norm-bounded uncertainties

A robust H _∞ preview control is investigated for an active suspension system with look-ahead sensors. The uncertain system is described by a state-space model with linear nominal parts and additional nonlinear time-varying norm-bounded uncertainties. Proof of robust stability and a feedback-type robust H _∞ preview controller are derived by augmenting the dynamics of the original system and previewed road input. As, however, the augmented previewed road input gives the system a much larger dimension than the original system, much more computation time is required for solving of Riccati equations. To resolve this problem, a decomposed robust H _∞ preview controller is proposed. Robust stability and performance variations for system uncertainties are shown using a numerical example of a quarter-car model.     

Corrosion characteristics of coated automotive parts subjected to field and proving ground tests

This project evaluated the corrosion damage over a five-year period of organic coated steels in automotive chassis parts during P/G and field tests using electrochemical impedance spectroscopy (EIS) in a 3.5 wt.% NaCl solution. EIS can provide both quantitative kinetic and mechanistic information about the performance of organic coating/metal systems. The difference in coating performance between the P/G and field test specimens was assessed in relation to the impedance parameters. In particular, it was found that the coating resistance and charge transfer resistance could be used to estimate the difference in corrosion damage between the P/G and field test specimens. The coating performance of the P/G and field test specimens was visually evaluated using SEM. The mechanism of corrosion occurring in the P/G test was similar to that taking place in the field test. The field test environment was about ten times as corrosive as the P/G environment in terms of the electrochemical impedance parameters and surface analysis.     

Effect of injection condition and swirl on D.I. diesel combustion in a transparent engine system

The objective of this work was to investigate the effects of injection conditions and swirl on D.I. diesel combustion using a transparent engine system. The test engine is equipped with a common rail injection system to control injection conditions and to obtain split injection characteristics. A combustion analysis and steady flow test were conducted to measure the heat release rate due to cylinder pressure and the swirl ratio. In addition, spray and diffusion flame images were obtained using a high speed camera. The LII & LIS methods were also used to obtain 2-D soot and droplet distributions. High injection pressure was found to shorten ignition delay, as well as to enhance peak pressure. The results also revealed that the heat release rate in the premixed combustion region was markedly reduced through the use of pilot injection, while the soot distribution and the heat release rate in the diffusion combustion region were increased. The swirl effect was found to shorten ignition delay at certain injection timings, and to enhance the heat release rate in all experimental conditions.     

Investigation of soot formation in Diesel-GTL fuel blends under quiescent conditions

In this study, a visual investigation of sprays and flames is performed, and soot formation in Diesel-GTL fuel blends is studied in a specially designed quiescent constant-volume chamber under various ambient gas temperatures and O₂ concentrations. Similar to the case of soot formation during diesel fuel combustion, the sooting zone during the mixing-controlled combustion of Diesel-GTL blends is located in the leading portion of the jet boundaries. Auto-ignition delay and soot concentration decrease with an increase of GTL content in the fuel blend. Soot also decreases with lower O₂ concentration, higher injection pressure, and lower ambient gas temperature. The lack of soot formation at lower O₂ concentrations and lower temperatures suggests that Diesel-GTL fuel blends can be successfully utilized in low-temperature diesel combustion technologies that are currently being developed. Furthermore, this mixing controlled combustion method with Diesel-GTL blends can be used to modulate various engine operation parameters, and therefore to simultaneously reduce the formation of soot and NOx within a wide range of diesel engine loads.     

Performance analysis according to the combination of energy storage system for fuel cell hybrid vehicle

The need for the unmanned ground combat vehicle (UGCV), which is used for the surveillance, reconnaissance and targeting during extremely dangerous condition on the battlefield, has steadily increased, and the transition from manned ground combat vehicles to unmanned ground combat vehicles is expected to reduce the loss of lives during battle. The UGCV needs many types of capabilities to achieve satisfactory performance. This paper focuses on the modeling and control of the power system of the UGCV, and proposes the fuel cell hybrid system (FCHS) for the power system of the UGCV. The fuel cell hybrid system has many advantages in stealth drive and the system efficiency. In addition, the FCHS is much quieter than the engine generator and generates much less heat. The benefits of the FCHS are advantageous for use in Army operations, which require ‘silent watch’ capability and the ability to operate without showing up on an enemy’s radar screen. The FCHS has a fuel cell and uses an energy storage system (ESS) as a power source. The ESS (e.g., batteries or ultracapacitors) helps the fuel cell supply power to the electric drive system and also recovers energy during deceleration. The ESS makes it possible to improve the efficiency and dynamic characteristic of the power system. In this paper, the FCHS is composed of different combinations of component models. The component sizes are chosen to satisfy performance requirements. In order to determine the power distribution between the fuel cell and the ESS, a power management strategy based on the required power and the SOC (state of charge) of the ESS is proposed. Batteries and ultracapacitor, components of the ESS, have different characteristics. Accordingly, varying the combination of ESS components can change the performance of the power system. The performance of the FCHS with respect to different combinations of ESS is analyzed using simulated results.