Search Results

Abstract Lane changes are stressful maneuvers for drivers, particularly during high-speed traffic flows. However, modeling driver’s lane-changing decision and implementation process is challenging due to the complexity and uncertainty of driving behaviors. To address this issue, this article presents a personalized Lane-Changing Model (LCM) for Advanced Driver Assistance System (ADAS) based on deep learning method. The LCM contains three major computational components. Firstly, with abundant inputs of Root Residual Network (Root-ResNet), LCM is able to exploit more local information from the front view video data. Secondly, the LCM has an ability of learning the global spatial-temporal information via Temporal Modeling Blocks (TMBs). Finally, a two-layer Long Short-Term Memory (LSTM) network is used to learn video contextual features combined with lane boundary based distance features in lane change events.

Abstract Obstacle avoidance is an important function in self-driving vehicle control. When the vehicle move from any arbitrary start positions to any target positions in environment, a proper path must avoid both static obstacles and moving obstacles of arbitrary shape. There are many possible scenarios, manually tackling all possible cases will likely yield a too simplistic policy. In this paper reinforcement learning is applied to the problem to form effective strategies. There are two major challenges that make self-driving vehicle different from other robotic tasks. Firstly, in order to control the vehicle precisely, the action space must be continuous which can’t be dealt with by traditional Q-learning. Secondly, self-driving vehicle must satisfy various constraints including vehicle dynamics constraints and traffic rules constraints. Three contributions are made in this paper.

Abstract This article presents a fault diagnosis approach for roller bearing by applying the autocorrelation approach to filtered vibration measured signal. An optimal Morlet wavelet filter is applied to eliminate the frequency associated with interferential vibrations; the raw measured signal is filtered with a band-pass filter based on a Morlet wavelet function whose parameters are optimized based on maximum Kurtosis. Autocorrelation enhancement is applied to the filtered signal to further reduce the residual in-band noise and highlight the periodic impulsive feature. The proposed technique is used to analyze the experimental measured signal of investigated vehicle gearbox. An artificial fault is introduced in vehicle gearbox bearing an orthogonal placed groove on the inner race with the initial width of 0.6 mm approximately. The faulted bearing is a roller bearing located on the gearbox input shaft - on the clutch side.

Abstract Heat generation characteristics of lithium ion batteries are vital for both the optimization of the battery cells and thermal management system design of battery packs. Compared with other factors, internal resistance has great influence on the thermal behavior of Li-ion batteries. Focus on a 3 Ah pouch type battery cell with the NCM/C material system, this paper quantitatively evaluates the battery heat generation behavior using an Extended Volume-Accelerating Rate Calorimeter in combination with a battery cycler. Also, internal resistances of the battery cell are measured using both the hybrid pulse power characteristic (HPPC) and electro-chemical impedance spectroscopy (EIS) methods. Experimental results show that the overall internal resistance obtained by the EIS method is close to the ohmic resistance measured by the HPPC method. Heat generation power of the battery cell is small during discharge processes lower than 0.5 C-rate.

Abstract Battery Drain problems can occur in the vehicle due to improper network management between electronic control units (ECUs). Aim of this paper is to identify the factors that cause transmission and cease of transmission of a network management message of an ECU along with its application messages that controls the sleep/wake-up performance of other ECUs in the network. Strategy used here is, based on the root cause analysis of problems found in Display unit in vehicle environment, the functional CAN signals impacting sleep/wake-up behavior is re-mapped along with the state flow transition of AUTOSAR NM Algorithm. A re-defined test case design and simulation for vehicle model is created. Especially it focuses on validating the impact of functional CAN signals on DUT’s sleep/wake-up performance.

Abstract The high-frequency noise issue is one of the most significant noise, vibration, and harshness problems, particularly in battery electric vehicles (BEVs). The sound package treatment is one of the most important approaches toward solving this problem. Owing to the limitations imposed by manufacturing error, assembly error, and the operating conditions, there is often a big difference between the actual values and the design values of the sound package components. Therefore, the sound package parameters include greater uncertainties. In this article, an uncertainty analysis method for BEV interior noise was developed based on an interval model to investigate the effect of sound package uncertainty on the interior noise of a BEV. An interval perturbation method was formulated to compute the uncertainty of the BEV’s interior noise.

Abstract Disc pad physical properties are believed to be important in controlling brake friction, wear and squeal. Thus these properties are carefully measured during and after manufacturing for quality assurance. For a given formulation, disc pad porosity is reported to affect friction, wear and squeal. This investigation was undertaken to find out how porosity changes affect pad natural frequencies, dynamic modulus, hardness and compressibility for a low-copper formulation and a copper-free formulation, both without underlayer, without scorching and without noise shims. Pad natural frequencies, modulus and hardness all continuously decrease with increasing porosity. When pad compressibility is measured by compressing several times as recommended and practiced, the pad surface hardness is found to increase while pad natural frequencies and modulus remain essentially unchanged.

Abstract A key problem of the construction of fully electric aircraft is the limited energy density of battery packs. It is generally accepted that this can only be overcome via new, denser battery chemistry together with a further increase in the efficiency of power utilization. One appealing approach for achieving the latter is using laser-assisted filler-based joining technologies, which offers unprecedented flexibility for achieving battery cell connections with the least possible electrical loss. This contribution presents our results on the effect of various experimental and process parameters on the electrical and mechanical properties of the laser-formed bond.

Abstract Improving the energy performance of batteries can increase the reliability of electric aircraft. To achieve this goal, battery management systems (BMS) are required to keep the temperature within the battery pack and cells below the safety limits and make the temperature distribution as even as possible. Batteries have a limited service life as a result of unwanted chemical reactions, physical changes that cause the loss of active materials in the structure, and internal resistance increase during the charging and discharging cycle of the battery. These changes usually affect the electrical performance of batteries. Battery life can be increased only by reducing or preventing unwanted chemical reactions. Lithium-ion (Li-ion) batteries are a suitable option due to their high specific energy and energy density advantages. In this study, the necessity of heat management is emphasized. The discharge tests of the Li-ion battery provided 94.6 Wh under 10C and 90.9 Wh under 1C.

Abstract The range of an aircraft is determined by the amount of energy that its batteries can store. Today, larger batteries are used to increase the range of electric vehicles, although energy efficiency decreases as the weight of the vehicles increases. Among the elements, lithium (Li) is the lightest and has the highest electrochemical potential. Therefore, the use of Li-ion batteries is recommended for hybrid aircraft. In addition, Li-ion batteries are the most common type of battery that is used in portable electronic devices such as smartphones, tablets, and laptops. However, Li-ion batteries may explode due to temperature. Therefore, the thermal analysis of Li-ion batteries was investigated both experimentally and numerically. Li-ion batteries were connected in series (the number is 9). Noboru’s theory of heat generation was discussed in the estimation of energy data.

Abstract Dynamic stresses exist in parts of a catalyst muffler caused by the vibration of a moving vehicle, and it is important to clarify and predict the vibration response properties for preventing fatigue failures. Assuming a vibration isolating installation in the vehicle frame, the vibration transmissibility and local dynamic stress of the catalyst muffler were examined through a vibration machine. Based on the measured data and by systematically taking vibration theories into consideration, a new prediction method of the vibration modes and parameters was proposed that takes account of vibration isolating and damping. A lumped vibration model with the six-element and one mass point was set up, and the vibration response parameters were analyzed accurately from equations of motion. In the vibration test, resonance peaks from the hanging bracket, rubber bush, and muffler parts were confirmed in three excitation drives, and local stress peaks were coordinate with them as well.

Abstract In this article, a unique design for Hydro-Pneumatic Energy Harvesting Suspension HPEHS system is introduced. The design includes a hydraulic rectifier to maintain one-way flow direction in order to obtain maximum power generation from the vertical oscillation of the suspension system and achieve handling and comfort car drive. A mathematical model is presented to study the system dynamics and non-linear effects for HPEHS system. A simulation model is created by using Advanced Modeling Environment Simulations software (AMEsim) to analyze system performance. Furthermore, a co-simulation platform model is developed using Matlab-Simulink and AMEsim to optimize the PID controller parameters of the external variable load resistor applied on the generator by using Particle Swarm Optimization (PSO).

Abstract Recently, electric heavy-duty tractor-trailers (EHDTTs) have assumed significance as they present an immediate solution to decarbonize the transportation sector. Hence, to illustrate the economic viability of electrifying the freight industry, a detailed numerical model to estimate the battery capacity for an EHDTT is proposed for a route between Washington, DC, to Knoxville, TN. This model incorporates the effects of the terrain, climate, vehicular forces, auxiliary loads, and payload in order to select the appropriate motor and optimize the battery capacity. Additionally, current and near-future battery chemistries are simulated in the model. Along with equations describing vehicular forces based on Newton’s second law of motion, the model utilizes the Hausmann and Depcik correlation to estimate the losses caused by the capacity offset of the batteries. Here, a Newton-Raphson iterative scheme determines the minimum battery capacity for the required state of charge.

Abstract Vehicle rolling resistance and weight are two of the factors that affect fuel economy. The vehicle tire rolling resistance has a more significant influence than aerodynamics drags on fuel economy at lower vehicle speeds, particularly true for medium- and heavy-duty trucks. Less vehicle weight reduces inertia loads, uphill grade resistance, and rolling resistance. The influence of weight on the fuel economy can be considerable particularly in light- to medium-duty truck classes because the weight makes up a larger portion of gross vehicle weight. This article presents an empirical investigation and a numerical analysis of the influences of rolling resistance and weight on the fuel economy of medium-duty trucks. The experimental tests include various tires and payloads applied on a total of 21vehicle configurations over three road profiles. These tests are used to assess the sensitivity of rolling resistance and weight to the vehicle fuel economy.

Abstract Vehicle cabin air quality depends on various parameters such as number of passengers, fan speed, and vehicle speed. In addition to controlling the temperature inside the vehicle, HVAC control system has evolved to improve cabin air quality as well. However, there is no standard test method to ensure reliable and repeatable comparison among different cars. The current study defined Cabin Air Quality Index (CAQI) and proposed a test method to determine CAQI. CAQIparticles showed dependence on the choice of metrics among particle number (PN), particle surface area (PS), and particle mass (PM). CAQIparticles is less than 1 while CAQICO2 is larger than 1. The proposed test method is promising but needs further improvement for smaller coefficient of variations (COVs).

Abstract This study investigates methods to precisely control a coolant pump in an internal combustion engine. The goal of this research is to minimize power consumption while still meeting optimal performance, reliability and durability requirements for an engine at all engine-operating conditions. This investigation achieves reduced fuel consumption, reduced emissions, and improved powertrain performance. Secondary impacts include cleaner air for the earth, reduced operating costs for the owner, and compliance with US regulatory requirements. The study utilizes mathematical modeling of the cooling system using heat transfer, pump laws, and boiling analysis to set limits to the cooling system and predict performance changes.

Abstract The supersession of metallic alloys with lightweight, high-strength composites is popular in the aircraft industry. However, aviation electronic enclosures for large format batteries and high power conversion electronics are still primarily made of aluminum alloys. These aluminum enclosures have attractive properties regrading structural integrity for the heavy internal parts, electromagnetic interference (EMI) suppression, electrical bonding for the internal cells, and/or electronics and failure containment. This paper details a lightweight carbon fiber composite chassis developed at Meggitt Sensing Systems (MSS) Securaplane, with a copper metallic mesh co-cured onto the internal surfaces resulting in a 50% reduction in weight when compared to its aluminum counterpart. In addition to significant weight reduction, it provides equal or improved performance with respect to EMI, structural and flammability performance.

Abstract The use of Abrasive Water Jet (AWJ) cutting technology can improve the edge stretchability in sheet metal forming. The advances in technology have allowed significant increases in working speeds and pressures, reducing the AWJ operation cost. The main objective of this work was to determine the effect of selected AWJ cutting parameters on the Hole Expansion Ratio (HER) for a DP800 (Dual-Phase) Advanced High-Strength Steel (AHSS) with s0 = 1.2 mm by using a fractional factorial design of experiments for the Hole Expansion Tests (HET). Additionally, the surface roughness and residual stresses were measured on the holes looking for a possible relation between them and the measured HER. A deep drawing quality steel DC06 with s0 = 1.0 mm was used for reference. The fracture occurrence was captured by high-speed cameras and by Acoustic Emissions (AE) in order to compare both methods.

Abstract This paper presents a machine learning application of the force/torque sensor in a human-robot collaborative manufacturing scenario. The purpose is to simplify the programming for physical interactions between the human operators and industrial robots in a hybrid manufacturing cell which combines several robotic applications, such as parts manipulation, assembly, sealing and painting, etc. A multiclass classifier using Light Gradient Boosting Machine (LightGBM) is first introduced in a robotic application for discriminating five different contact states w.r.t. the force/torque data. A systematic approach to train machine-learning based classifiers is presented, thus opens a door for enabling LightGBM with robotic data process. The total task time is reduced largely because force transitions can be detected on-the-fly. Experiments on an ABB force sensor and an industrial robot demonstrate the feasibility of the proposed method.

Abstract In an automated machining process, monitoring the conditions of the tool is essential for deciding to replace or repair the tool without any manual intervention. Intelligent models built with sensor information and machine learning techniques are predicting the condition of the tool with good accuracy. In this study, statistical models are developed to identify the conditions of the abrasive grinding wheel using the Acoustic Emission (AE) signature acquired during the surface grinding operation. Abrasive grinding wheel conditions are identified using the abrasive wheel wear plot established by conducting experiments. The piezoelectric sensor is used to capture the AE from the grinding process, and statistical features of the abrasive wheel conditions are extracted in time and wavelet domains of the signature. Machine learning algorithms, namely, Classification and Regression Trees (CART) and Support Vector Classifiers (SVC), are used to build statistical models.