Search Results

The course will enable the learner to apply the fundamental principles behind safety of machine learning to a wide range of applications. The course guides learners through an appropriate selection of methods and tools tailored to the learner’s specific projects. With the acquired knowledge the learner will be able to shape the development and assessment of ML-based safety-related functions enabling their teams to leverage the power of advanced ML techniques without undermining safety.

When vehicles share certain information wirelessly via Dedicated Short Range Communications (DSRC), they enable a new layer of electronic vehicle safety that, when needed, can generate warnings to drivers and even initiate automatic preventive actions. Vehicle location and velocity provided by Global Navigation Systems (GNSS), including GPS, are key in allowing vehicle path estimation. GNSS is effective in accurately determining a vehicle's location coordinates in most driving environments, but its performance suffers from obstructions in dense urban environments. To combat this, augmentations to GNSS are being contemplated and tested. This testing has been typically done using a reference GNSS system complimented by expensive military-grade inertial sensors, which can still fail to provide adequate reference performance in certain environments.

We report here results from first year of the BMW MINI E deployment. BMW deployed 450 MINI E?s to North America. Nearly 50% were leased by households in Los Angeles and the New York area. PH&EV Center researchers surveyed MINI E drivers throughout their year with the vehicles, focusing on the experiences of 50 households who volunteered for more detailed interviews. We report here their experiences with driving electric vehicles, adaptions to daily range limitations, and using electricity as a fuel. Presenter Thomas Turrentine, Univ. of California-Davis

ECOtality North America, in partnership with the Idaho National Laboratory (INL), Nissan North America, General Motors, and over 40 government, electric utility, and private organizations, has launched a large-scale demonstration of electric vehicle charging infrastructure. This demonstration, called The EV Project, will deploy more than 15,000 level 2 and DC fast chargers in private residence, commercial, and public locations in seven market areas in Arizona, California, Oregon, Tennessee, Texas, Washington state, and Washington, D.C. The EV Project will also include a total of 5,700 Nissan Leaf battery electric vehicles and 2,600 Chevrolet Volt extended range electric vehicles, operated by consumers and fleets in each of the market areas. This demonstration, which is funded by the U.S. Department of Energy�s (DOE) Vehicle Technologies Program, represents the largest ever deployment of electric vehicles and charging infrastructure.

In Aeronautic industry, when we launch a new industrialization for an aircraft sub assembly we always have the same questions in mind for drilling operations, especially when focusing on lean manufacturing. How can we avoid dismantling and deburring parts after drilling operation? Can a drilling centre perform all the tasks needed to deliver a hole ready to install final fastener? How can we decrease down-time of the drilling centre? Can a drilling centre be integrated in a pulse assembly line? How can we improve environmental efficiency of a drilling centre? It is based on these main drivers that AIRBUS has developed, with SPIE and SOS, a new generation of drilling centre dedicated for hard materials such as titanium, and high thicknesses. The first application was for the assembly of the primary structure of A350 engine pylons. The main solution that was implemented meeting several objectives was the development of orbital drilling technology in hard metal stacks.

ISO 26262 is the actual standard for Functional Safety of automotive E/E (Electric/Electronic) systems. One of the challenges in the application of the standard is the distribution of safety related activities among the participants in the supply chain. In this paper, the concept of a Safety Element out of Context (SEooC) development will be analyzed showing its current problematic aspects and difficulties in implementing such an approach in a concrete typical automotive development flow with different participants (e.g. from OEM, tier 1 to semiconductor supplier) in the supply chain. The discussed aspects focus on the functional safety requirements of generic hardware and software development across the supply chain where the final integration of the developed element is not known at design time and therefore an assumption based mechanism shall be used.

Abstract A number of studies have shown that driving an unfamiliar vehicle has the potential to introduce additional risk, especially for novice drivers. However, such studies have generally used statistical methods based on analyzing crash and near-crash data from a range of driver groups, and therefore the evaluation has the potential to be subjective and limited. For a more objective perspective, this study suggests that it would be worthwhile to consider vehicle dynamic signals obtained from the Controller Area Network (CAN-Bus) and smartphones. This study, therefore, is focused on the effect of driver experience and vehicle familiarity for issues in driver modeling and distraction. Here, a group of 20 drivers participated in our experiment, with 13 of them having participated again after a one-year time lapse in order for analysis of their change in driving performance.

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 In today’s competitive automobile market, driver comfort is at utmost importance and the bar is being raised continuously. Gear Shifting is a crucial customer touch point. Any issue or inconvenience caused while shifting gear can result into customer dissatisfaction and will impact the brand image. While there are continual efforts being taken by most of the car manufactures, “Double Bump” in gearshift has remained as a pain area and impact severely on the shift feel. This is more prominent in North-South (N-S) transmissions. In this paper ‘Double Bump’ is a focus area and a mathematical / analytical approach is demonstrated by analyzing ‘impacting parameters’ and establishing their co-relation with double bump. Additionally, the results are also verified with a simulation model.

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.

1. On page 111, the authors have described a method to assess driver distraction. In this method, participants maintained a white square size on a forward display by using a game gas pedal of like in car-following situation. The size of the white square is determined by calculating the distance to a virtual lead vehicle. The formulas to correct are used to explain variation of acceleration of the virtual lead vehicle. The authors inadvertently incorporated old formulas they had used previously. In the experiments discussed in the article, the corrected formulas were used. Therefore, there is no change in the results. The following from the article:

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.

Abstract A key element in an ergonomically designed driver’s seat in a car is the correct identification of driver seating position and posture accommodation. Current practice by the automotive Original Equipment Manufacturer (OEM) is to utilize the Society of Automotive Engineering (SAE) J1517 standard practice as a reference. However, it was found that utilizing such guidelines, which were developed based on the American population, did not fit well with the anthropometry and stature of the Malaysian population. This research seeks to address this issue by comparing the SAE J1517 Model against Malaysian preferred driving position. A total of 62 respondents were involved for the driver seating position and accommodation study in the vehicle driver’s seat buck mockup survey and measurements. The results have shown that the Malaysian drivers prefer to sit forward as compared to the SAE J1517 Model and have shorter posture joint angle.

Abstract The advancement towards development of autonomy follows either the bottom-up approach of gradually improving and expanding existing Advanced Driver Assist Systems (ADAS) technology where the driver is present in the control loop or the top-down approach of directly developing autonomous vehicle hardware and software using alternative approaches without the driver present in the control loop. Most ADAS systems today fall under the classification of SAE Level 1 which is also referred to as the driver assistance level. The progression from SAE Level 1 to SAE Level 2 or partial automation involves the critical task of merging automated lateral control and automated longitudinal control such that the tasks of steering and acceleration/deceleration are not required to be handled by the driver under certain conditions [1].

Abstract Future SAE Level 4 and Level 5 autonomous vehicles (AV) will require novel applications of localization, perception, control, and artificial intelligence technology in order to offer innovative and disruptive solutions to current mobility problems. This article concentrates on low-speed autonomous shuttles that are transitioning from being tested in limited traffic, dedicated routes to being deployed as SAE Level 4 automated driving vehicles in urban environments like college campuses and outdoor shopping centers within smart cities. The Ohio State University has designated a small segment in an underserved area of the campus as an initial AV pilot test route for the deployment of low-speed autonomous shuttles. This article presents initial results of ongoing work on developing solutions to the localization and perception challenges of this planned pilot deployment.

Abstract To evaluate that automated vehicle is as safe as a human driver, a following question is studied: how does an automated vehicle react under extreme conditions close to collision? In order to understand the collision avoidance capability of an automated vehicle, we should analyze not only such post-extreme condition behavior but also pre-extreme condition behavior. We present a theory to analyze the collision avoidance capability of automated driving technologies. We also formulate a collision avoidance equation on the theory. The equation has two types of solutions: response driving plans and preparation driving plans. The response driving plans are supported by response strategy on which the vehicle reacts after detection of a hazard and they are highly efficient in terms of travel time.

Abstract Management of expressway networks has been mainly focused on defect management without looking at the correlations with accidental risks. This causes unsustainability in expressway infrastructure maintenance since such defects may not be a contributing factor toward public safety. Thus it is necessary to incorporate accidental events for decision-making in infrastructure management. This study has developed a novel approach to machine learning (ML) that incorporates actual primary data from the last 10 years of single-vehicle accidents (SVA) by collisions with motorway facilities, or so-called single-vehicle collisions with fixed objects. The ML is firstly aimed at identifying the influential factors of SVA in relation to finding effective countermeasures for accidents by integrating the correlation analysis, multiple regression analysis, and ML techniques. The study reveals that wet pavement conditions have a significant effect on SVA.

This paper focuses on full body dynamical analysis of car ingress/egress motion. It aims at proposing an experimental protocol adapted for analysing joint loads using inverse dynamics. Two preliminary studies were first performed in order to 1/ define the main driver/car interactions so as to allow measuring the contact forces at all possible contact zones and 2/ identify the design parameters that mainly influence the discomfort. In order to verify the feasibility of the protocol, a laboratory study was carried out, during which two subjects tested two car configurations. The experimental equipment was composed of a variable car mock-up, an optoelectronic motion tracking system, two 6D-force plates installed on the ground next to the doorframe and on the car floor, a 6D-Force sensor between the steering wheel and the steering column, and two pressure maps on the seat. Motions were reconstructed from measured surface markers trajectories using inverse kinematics.

A driving simulator capable of duplicating the critical sensations incurred during a spin, or when a driver is engaged in drift cornering, was constructed by Mitsubishi Heavy Industries, Ltd., and Hiromichi Nozaki of Kogakuin University. Specifically, the simulator allows independent movement along three degrees of freedom and is capable of exhibiting extreme yaw and lateral acceleration behaviors. Utilizing this simulator, the control characteristics of drift cornering have become better understood. For example, after a J-turn behavior experiment involving yaw angle velocity at the moment when the drivers attention transitions to resuming straight ahead driving, it is now understood that there are major changes in driver behavior in circumstances when simulator motions are turned off, when only lateral acceleration motion is applied, when only yaw motion is applied, and when combined motions (yaw + lateral acceleration) are applied.

The ride dynamics of typical North-American soil compactors were investigated via analytical and experimental methods. A 12-degrees-of-freedom in-plane ride dynamic model of a single-drum compactor was formulated through integrations of the models of various components such as driver seat, cabin, roller drum and drum isolators, chassis and the tires. The analytical model was formulated for the transit mode of operation at a constant forward speed on undeformable surfaces with the roller vibrator off. Field measurements were conducted to characterize the ride vibration environments during the transit mode of operation. The measured data revealed significant magnitudes of whole-body vibration of the operator-station along the vertical, lateral, pitch and roll-axes. The model results revealed reasonably good agreements with ranges of the measured vibration data.