Search Results

LiDAR sensors are common in automated driving due to their high accuracy. However, LiDAR processing algorithm development suffers from lack of diverse training data, partly due to sensors’ high cost and rapid development cycles. Public datasets (e.g. KITTI) offer poor coverage of edge cases, whereas these samples are essential for safer self-driving. We address the unmet need for abundant, high-quality LiDAR data with the development of a synthetic LiDAR point cloud generation tool and validate this tool’s performance using the KITTI-trained PIXOR object detection model. The tool uses a single camera raycasting process and filtering techniques to generate segmented and annotated class specific datasets.

Understanding cylinder-kit tribology is pivotal to durability, emission management, reduced oil consumption, and efficiency of the internal combustion engine. This work addresses the understanding of the fundamental aspects of oil transport and combustion gas flow in the cylinder kit, using simulation tools and high-performance computing. A dynamic three-dimensional multi-phase, multi-component modeling methodology is demonstrated to study cylinder-kit assembly tribology during the four-stroke cycle of a piston engine. The percentage of oil and gas transported through different regions of the piston ring pack is predicted, and the mechanisms behind this transport are analyzed. The velocity field shows substantial circumferential flow in the piston ring pack, leading to blowback into the combustion chamber during the expansion stroke.

Deep learning based approaches for object detection are heavily dependent on the nature of data used for training, especially for vehicles driving in cluttered urban environments. Consequently, the performance of Convolutional Neural Network (CNN) architectures designed and trained using data captured under clear weather and favorable conditions, could degrade rather significantly when tested under cloudy and rainy conditions. This naturally becomes a major safety issue for emerging autonomous vehicle platforms relying on CNN based object detection methods. Furthermore, despite a noticeable progress in the development of advanced visual deraining algorithms, they still have inherent limitations for improving the performance of state-of-the-art object detection. In this paper, we address this problem area by make the following contributions.

Three dimensional numerical simulation of the transient turbulent jet and ignition processes of a premixed methane-air mixture of a turbulent jet ignition (TJI) system is performed using Converge computational software. The prechamber initiated combustion enhancement technique that is utilized in a TJI system enables low temperature combustion by increasing the flame propagation rate and therefore decreasing the burn duration. Two important components of the TJI system are the prechamber where the spark plug and injectors are located and the nozzle which connects the prechamber to the main chamber. In order to model the turbulent jet of the TJI system, RANS k-ε and LES turbulent models and the SAGE chemistry solver with a reduced mechanism for methane are used.

Turbulent jet ignition is a pre-chamber ignition enhancement method that produces a distributed ignition source through the use of a chemically active turbulent jet which can replace the spark plug in a conventional spark ignition engine. In this paper combustion visualization and characterization was performed for the combustion of a premixed propane/air mixture initiated by a pre-chamber turbulent jet ignition system with no auxiliary fuel injection, in a rapid compression machine. Three different single orifice nozzles with orifice diameters of 1.5 mm, 2 mm, and 3 mm were tested for the turbulent jet igniter pre-chamber over a range of air to fuel ratios. The performance of the turbulent jet ignition system based on nozzle orifice diameter was characterized by considering both the 0-10 % and the 10-90 % burn durations of the pressure rise due to combustion.

A three-dimensional piston ring model has been developed using finite element method with eight-node hexahedral elements. The model predicts the piston ring conformability with the cylinder wall as well as the separation gap between the interfaces if existing in the radial direction. In addition to the radial interaction between the ring front face and the cylinder wall, the model also predicts the contact between the ring and groove sides in the axial direction. This means, the ring axial lift, ring twist, contact forces with the groove sides along the circumferential direction are all calculated simultaneously with the radial conformability prediction. The ring/groove side contact can be found for scraper ring at static condition, which is widely used as the second compression ring in a ring pack. Thermal load is believed having significant influence on the ring pack performance.

Fully three-dimensional computational fluid dynamic simulations with detailed chemistry of a single-orifice turbulent jet ignition device installed in a rapid compression machine are presented. The simulations were performed using the computational fluid dynamics software CONVERGE and its RANS turbulence models. Simulations of propane fueled combustion are compared to data collected in the optically accessible rapid compression machine that the model's geometry is based on to establish the validity and limitations of the simulations and to compare the behavior of the different air-fuel ratios that are used in the simulations.

A multicomponent droplet evaporation model which discretizes the one-dimensional mass and temperature profiles inside a droplet with a finite volume method has been developed and implemented into a large-eddy simulation (LES) model for spray simulations. The LES and multicomponent models were used along with the KH-RT secondary droplet breakup model to simulate realistic fuel sprays in a closed vessel. The effect of various spray and ambient gas parameters on the liquid penetration length of different single component and multicomponent fuels was investigated. The numerical results indicate that the spray penetration length decreases non-linearly with increasing gas temperature or pressure and is less sensitive to changes in ambient gas conditions at higher temperatures or pressures. The spray models and LES were found to predict the experimental results for n-hexadecane and two multicomponent surrogate diesel fuels reasonably well.

Large Eddy Simulations of atomization and evaporation of liquid fuel sprays in diesel engine conditions are performed with stochastic breakup and non-equilibrium droplet heat and mass transfer models. The size and number density of the droplets generated by the breakup model are assumed to be governed by a Fokker-Planck equation, describing the evolution of the PDF of droplet radii. The fragmentation intensity spectrum is considered to be Gaussian and the scale of Lagrangian relative velocity fluctuations is included in the breakup frequency calculations. The aerodynamic interactions of droplets in the dense part of the spray are modeled by correcting the relative velocity of droplets in the wake of other droplets. The stochastic breakup model is employed together with the wake interaction model for simulations of non-evaporating and evaporating sprays in various gas temperature and pressure conditions.

Air-to-fuel (A/F) ratio is the mass ratio of the air-to-fuel mixture trapped inside a cylinder before combustion begins, and it affects engine emissions, fuel economy, and other performances. Using an A/F ratio and dual-fuel ratio control oriented engine model, a multi-input-multi-output (MIMO) sliding mode control scheme is used to simultaneously control the mass flow rate of both port fuel injection (PFI) and direct injection (DI) systems. The control target is to regulate the A/F ratio at a desired level (e.g., at stoichiometric) and fuel ratio (ratio of PFI fueling vs. total fueling) to any desired level between zero and one. A MIMO sliding mode controller was designed with guaranteed stability to drive the system A/F and fuel ratios to the desired target under various air flow disturbances.

This paper focuses on the numerical investigation of the mixing and combustion of ethanol and gasoline in a single-cylinder 3-valve direct-injection spark-ignition engine. The numerical simulations are conducted with the KIVA code with global reaction models. However, an ignition delay model mitigates some of the deficiencies of the global one-step reaction model and is implemented via a two-dimensional look-up table, which was created using available detailed kinetics models. Simulations demonstrate the problems faced by ethanol operated engines and indicate that some of the strategies used for emission control and downsizing of gasoline engines can be employed for enhancing the combustion efficiency of ethanol operated engines.

Human intraspecific variation is a complex problem, but may be better understood by using computational models in tandem with knowledge about the genetic bases of phenotypic traits. These results can be used in a multitude of settings. To move closer to this goal, biologically-realistic mappings between genotype and phenotype are constructed using genetic algorithm and neural network-like models. These models allow for gene-gene and gene-environment interactions to be characterized in the resulting phenotype. Two types of model are introduced: a simple, two-layer model, and a more complex model. The final section will focus on trends of growth and development in relation to relationship to modeling anthropometric traits and other morphological phenomena.

Electro-pneumatic valve actuators are used to eliminate the cam shaft of a traditional internal combustion engine. They are used to control the opening timing, duration, and lift of both intake and exhaust valves. A physics based nonlinear mathematical model called the level one model was built using Newton's law, mass conservation and thermodynamic principles. A control oriented model, the level two model, was created by partially linearizing the level one model for model reference parameter identification. This model reduces computational throughput and enables real-time implementation. A model reference adaptive control system was used to identify the nonlinear parameters that were needed for generating a feedforward control signal. The closed-loop valve lift tracking, valve opening and closing timing control strategies were proposed.

Internet worms have shown the capability to compromise millions of network hosts in a matter of seconds, thereby precluding human countermeasures. A worm over a vehicular ad hoc network (VANET) can, in addition to the well-known threats, pose a whole new class of traffic-related threats (ranging from congestion to large-scale accidents). To combat these automated adversaries, security patches can be distributed by good worms. An accurate VANET-based worm propagation model is essential to protect from malicious worms and to efficiently utilize good worms for distribution of security patches. This paper derives an approximate closed-form mathematical model of worm propagation over VANETs. Simulation results assert that the proposed model captures the VANET worm propagation dynamics with outstanding accuracy.

Cavitation induced cylinder liner erosion can be a significant durability problem in high power density diesel engines. It is typically discovered in the field, thus causing costly redesigns. The application of a predictive simulation to analyze the liner cavitation process upfront could identify the problem early on and enable significant savings. Hence, this work investigates the ability of the computational fluid dynamics (CFD) multiphase flow simulation tool to handle vibration induced cavitation. A flow of liquid through a U-shaped duct is analyzed, where a middle segment of the duct is set to vibrate in a manner resembling vibration of the cooling jacket walls in an internal combustion engine. Velocity, pressure and vapor concentration fields are tracked for two cases, distinguished by different frequencies of duct wall vibration.

In an effort to create computer models to promote rapid, cost-effective prototyping while easing design changes, more information about how people interact with seats is needed. Predicting the occupant location, their geometry, and motion within a vehicle leads to a better determination of safety restraint location, controls reach, and visibility - factors that affect the overall operation of the vehicle. Based on the Michigan State University JOHN model, which provides a biomechanical simulation of the torso posture, experiments were conducted to examine the change of postures due to seat and interior package factors. The results can be incorporated into the posture prediction model of the RAMSIS program to give a more detailed prognosis of the spine curvature and refine the model-seat interactions. This paper will address findings of the experimental study with relation to model development.

In order to improve the reliability of fan design and the prediction of underhood engine cooling based on CFD, Valeo Motors and Actuators and Michigan State University have teamed up to develop a comprehensive experimental and numerical database. The initial focus has been on the simulations of the isolated fan environment in two different test facilities. To understand the discrepancies observed in the comparisons of integral performances, the first detailed hot wire measurements on the MSU test facility have been collected. The data are split into mean velocity components and RMS fluctuations. The former are successfully compared to three detailed turbulent numerical simulations of the corresponding facilities.

Injuries to the lower extremities are common during car accidents because the lower extremity is typically the first point of contact between the occupant and the car interior. While injuries to the knee, ankle and hip are usually not life threatening, they can represent a large societal burden through treatment costs, lost work days and a reduced quality of life. The aim of the current study was to specifically study injuries associated with the knee and to propose a methodology which could be used to prevent future knee injuries. To understand the scope of this problem, a study was designed to identify injury trends in car crashes for the years 1979-1995. The NASS (National Accident Sampling System) showed that 10% of all injuries were to the knee, second only to head and neck injuries. Most knee injuries resulted from knee-to-instrument panel contact. Subfracture injuries were most common (contusions, abrasions, lacerations) followed by gross fracture injuries.

Since the 1960's, automotive seats have been designed and evaluated with tools and procedures described in the SAE Recommended Practice J826. The SAE J826 design template and testing manikin each have a torso with a flat lower back shape and with a single joint at the H-point. The JOHN models provide a more anatomically detailed representation of human shape and movement. The articulations of the JOHN torso (pelvic, lumbar, and thoracic) segments are coupled so that their relative positions are determined by a single parameter related to spinal curvature. This paper describes the development and use of the JOHN biomechanical models for seating design.

Lower extremity (knee) injury prediction resulting from impact trauma is currently based on a bone fracture criterion derived from experiments on predominantly aged cadavers. Subsequent experimental and theoretical studies indicate that more aged, pathological specimens require higher, not lower, loads to initiate bone fracture. This suggests that a bone fracture criterion based solely on aged specimens may not be representative of the current driving population. In the current study, we sought to determine if cadaver age, physical size, sex, baseline joint pathology, or patellar geometry correlated with fracture load. An analysis was made of data from previous impact experiments conducted on fifteen isolated cadaver knees using a consistent impact protocol. The protocol consisted of sequentially increasing the impact energy with a rigid interface until gross fracture. Gross bone fractures occurred at loads of 6.9±2.0 kN (range 3.2 to 10.6 kN) using this protocol.