Search Results

Forward collision warning is one of the most challenging concerns in the safety of autonomous vehicles. A cooperation between many sensors such as LIDAR, Radar and camera helps to enhance the safety. Apart from the importance of having a reliable object detector, the safety system should have requisite capabilities to make reasonable decisions in the moment. In this work, we concentrate on detecting front vehicles of autonomous cars using a monocular camera, beyond only a detection method. In fact, we devise a solution based on a cooperation between a deep object detector and a reinforcement learning method to provide forward collision warning signals. The proposed method models the relation between acceleration, distance and collision point using the area of the bounding box related to the front vehicle. An agent of learning automata as a reinforcement learning method interacts with the environment to learn how to behave in eclectic hazardous situations.

Government accident statistics show that approximately 35% of all car accident victims suffer an injury to the head and face. Such injuries are common during frontal, side, and rollover accidents as the head may impact the steering wheel, side pillars, windshield, or roof. Further, non-threatening injuries (i.e abrasions) may be suffered due to contact with the deployed airbag, or, in the case of an out-of-position occupant, a deploying airbag. While the forces and accelerations measured internal to the head are known to correlate with serious head injury (i.e. concussion, skull fracture, diffuse axonal injury), it is currently not possible to record how the loads are distributed over the head and face with the current ATD. Ultimately, such data could eventually be used to provide improved resolution as to the probability of superficial, soft tissue damage since past cadaver studies show that the distribution of contact pressures are related to such injuries.

With many advantages of GDI technology, one major disadvantage is high HC emissions. The primary goal of this study is to determine the optimum values of engine parameters that would result in maximum power output from a GDI engine, with complete combustion, minimum hydrocarbon (HC) emissions, and minimum specific fuel consumption. A two-dimensional engine geometry with a piston-bowl was selected for faster engine CFD simulations. The first part involves a study of the affect of engine parameters on performance and HC emissions. The parameters considered were, equivalence ratio (mass of injected fuel), injection timing, ignition timing, engine RPM, spray cone angle, and velocity of fuel injection. The second part of the study involves determining the optimum values of fuel mass injected, injection timing, and ignition timing in order to maximize power output while limiting the amount of fuel left unburned after the end of the expansion process.

The fatal automobile accidents can be attributed to fatigued and distracted driving by drivers. Driver Monitoring Systems alert the distracted drivers by raising alarms. Most of the image based driver drowsiness detection systems face the challenge of failure proof performance in real time applications. Failure in face detection and other important part (eyes, nose and mouth) detections in real time cause the system to skip detections of blinking and yawning in few frames. In this paper, a real time robust and failure proof driver drowsiness detection system is proposed. The proposed system deploys a set of detection systems to detect face, blinking and yawning sequentially. A robust Multi-Task Convolutional Neural Network (MTCNN) with the capability of face alignment is used for face detection. This system attained 97% recall in the real time driving dataset collected. The detected face is passed on to ensemble of regression trees to detect the 68 facial landmarks.

As automated driving becomes more common, simulation of vehicle dynamics and control scenarios are increasingly important for investigating motion control approaches. In this work, a study of the differences between the Pure Pursuit and Stanley autonomous vehicle controllers, based on vehicle dynamics responses, is presented. Both are geometric controllers that use only immediate vehicle states, along with waypoint data, to control a vehicle’s future direction as it proceeds from point to point, and both are among the most popular lateral controllers in use today. The MATLAB Automated Driving Toolbox is employed to implement and virtually test the Pure Pursuit and Stanley lateral controllers in different driving scenarios. These include low intensity scenarios such as city driving, and emergency maneuvers such as the moose test.

A hardware and software architecture suitable for a safety-critical steer-by-wire systems is presented. The architecture supports three major failure modes and features several safety protocols and mechanisms. Failures due to component failures, software errors, and human errors are handled by the architecture and safety protocols. A test implementation using replicated communication channels, controllers, sensors, and actuators has been performed. The test implementation uses the CAN protocol, Motorola S12 microcontrollers, and Microchip MCP250XX components with a steering wheel and road wheel simulator. The focus of the paper is on the application level, using system engineering principles which incorporate a holistic approach to achieve safety at various levels.

Analysis of the National Automotive Sampling System (NASS) data reveals that vehicle rollover accidents account for a relatively a small number of accidents, but the associated frequency of serious injury is high compared to frontal or side impact. These data demonstrate the apparently elevated probability of head and neck injury during rollover, with head injury occurring more frequently, injured 4.5 times more frequently than the neck when considering all injuries. Automotive industry researchers have performed numerous rollover tests with instrumented ATD's and have predicted an elevated probability of neck injury with little chance of head injury. This contradicts field data (NASS-CDS) which suggests a high frequency of head injury with little chance of neck injury. This difference may be explained in part, through the different volumes of data presented in the literature.

Bond graph modeling is a powerful technique to study the complex interactions occurring between various components in a system. A few investigations were carried out to study vehicle dynamics using Bondgraphs, but are limited to 2 degree of freedom systems [1,2&3]. In this work, a 4-DOF-vehicle model was developed using bond graphs. A frequency response analysis was also carried out to study the natural frequencies. This model was later validated using Lagrangian principles. The results correlated well for a typical passenger car using the manufacturer supplied information available in the public domain.

The objective of this study was to determine risk of injury to the driver during a frontal impact in a Formula SAE vehicle. Formula SAE is a collegiate student design competition where every year universities worldwide build and compete with open-wheel formula-style race cars. Formula SAE 2006 rules stipulate the use of an impact attenuator to absorb energy in the event of a frontal impact. These rules mandated an average deceleration not to exceed 20-g from a speed of 7.0 m/s (23 ft/s), but do not specify a specific time or pulse shape of the deceleration. The pulse shapes tested in this study included an early high-g, constant-g, and late high-g pulse. The tests were performed using the deceleration sled at the Kettering University Crash Safety Center. Using industry standard practices, this study examined the driver's risk of injury with regard to neck and femur loads, head and chest accelerations, as well as kinematic analysis using high speed video.

This paper deals with a DC motor driven Electric Power Assisted Steering system integrating a manual rack and pinion steering system. The DC motor is controlled by an electronic controller and connected to the steering system through a set of gears. The DC motor and the steering were modeled using principles of bond graphs and combined with Controller to form the system model. The state equations describing the system were developed using the bond graphs. The performance of the system was simulated using Matlab/Simulink. The electronic controller develops an output voltage to control the DC motor, considering vehicle speed and Steering Column torque inputs. This paper used a regression-weighted function to describe the operation of the controller. Transient and steady state performance characteristics of the system were evaluated including its damping characteristics.

Numerous studies have documented that lower extremity injury is second only to the head and face in automotive accidents. Such injuries are common because the lower extremity is typically the first point of contact between the occupant and the car interior. Of all lower extremity injuries, the knee is the most common site of trauma. This typically results from high speed contact with the instrument panel which can produce fracture and subfracture (contusions, lacerations, abrasions) level injuries. Current Federal safety guidelines use a bone fracture criterion which is based solely on a peak load. The criterion states that loads exceeding 10 kN will likely result in gross bone fracture. However, cadaver experiments have shown that increased contact area (via padding) over the knee can significantly increase the amount of load that can be tolerated before fracture or subfracture injury.

As advanced vehicle controls and autonomy become mainstream in the automotive industry, the need to employ traditional mathematical models and control strategies arises for the purpose of simulating autonomous vehicle handling maneuvers. This study focuses on lane change maneuvers for autonomous vehicles driving at low speeds. The lane change methodology uses PID (Proportional-Integral-Derivative) controller to command the steering wheel angle, based on the yaw motion and lateral displacement of the vehicle. The controller was developed and tested on a bicycle model of an electric vehicle (a Chevrolet Bolt 2017), with the implementation done in MATLAB/Simulink. This simple mathematical model was chosen in order to limit computational demands, while still being capable of simulating a smooth lane change maneuver under the direction of the car’s mission planning module at modest levels of lateral acceleration.

Steer-by-wire is a system where there are no mechanical connections between the steering wheel and the tires. With the inception of electric and hybrid cars, steer-by-wire is becoming more common. A steer-by-wire car opens many opportunities for additional feedback on the steering wheel. Providing haptic feedback through the steering wheel will add additional depth and capabilities to make the driving experience safer. In this paper we investigated the effects of force feedback on the steering wheel in order to detect and/or avoid blind spot collisions. Two types of force feedback are examined using a driving simulator: a rumble and a counter steering force. A rumble on the steering wheel can avoid blind-spot accidents by providing feedback to drivers about vehicles in their blind spots. Providing counter steering force feedback can help in the reduction in blind-spot accidents. The results show that adding counter steering force feedback did reduce blind-spot related collisions.

In September 1893, little could 23-year-old mechanic J. Frank Duryea dream of the changes that would be brought about by his creation -- a frail gasoline buggy that made its debut on the streets of Springfield, Massachusetts. Charles E. and J. Frank Duryea, two brothers from rural Illinois, were the founders of the American automobile industry. The Duryea Motor Wagon company was the first company organized in the United States for the manufacture of automobiles. The attention-getting, older brother Charles demanded - and to date has received - the principal credit for these pioneering accomplishments. A bitter family feud between the brothers, which was even carried on by their families after their deaths, further muddied the question about the individual brothers' contributions. However, in Carriages Without Horses: J. Frank Duryea and the Birth of the American Automobile Industry, historian and author Richard P. Scharchburg proves that the quiet, self-effacing younger brother J.

The “Range-of Motion of the Cervical Spine of Children” study is a collaboration between Kettering University and McLaren Regional Medical Center in Flint, Michigan to quantify and establish benchmarks of “normal” range of motion (ROM) in children. The results will be analyzed to determine mean and standard deviation of degrees of rotation and used to improve the occupant protection in motor vehicles, sports equipment and benefits of physical therapy. The data will be invaluable in the development of computational models to analyze processes involving children in motion.

FMVSS 223 and 224 set standards for “Rear Impact Protection” for trailers and semi-trailers with a gross weight rating greater than 10000 pounds. A limited amount of experimental data is available for evaluating the different performance attributes of rear impact guards. The crash tests are usually limited to fixed parameters such as impact speed, guard height, strength and energy absorption, etc. There also seems to be some misunderstanding of the interdependence of guard strength and energy absorption, and their combined effect on the guard's ability to limit underride while keeping occupant acceleration forces in a safe range. In this paper, we validated the Finite Element (FE) model of an existing rear impact guard against actual FMVSS 223 tests. We also modified a previously evaluated FE model of a 1990 Ford Taurus by updating its hood geometry and material properties.

Use of the top tether attachment in three commonly available anchor points provides added restraint of child restraint systems (CRS). Three tether attachment positions were used; floor, behind the head rest (parcel deck) and at the ceiling. The three anchor points are comparable in efficacy while no tether allows increased travel of the anthropomorphic test device (ATD) head. Two series of six tests were conducted at a max speed of 20 mph and peak deceleration of 16 G's using a deceleration sled test apparatus. The first series of tests was conducted at a 90 degree impact angle. On average there is 9% less head travel when using the tether attachment compared to not using the tether attachment, all other conditions begin equal. The second series of tests was conducted at a 73 degree impact angle, there is 15% less head travel when using the tether attachment compared to not using the tether attachment, all other conditions begin equal.

Rollover crashes are responsible for a significant proportion of traffic fatalities each year, while they represent a relatively small proportion of all motor vehicle collisions. The purpose of this study was to focus on rollover events from an occupant's perspective to understand what type of industry test method, ATD, computer based model, and injury assessment measures are required to provide occupant protection during rollovers. Specific injuries most commonly experienced in rollovers along with the associated injury sources were obtained by review of 1998-2000 NASS-CDS records. These data suggest that models capable of predicting the likelihood of brain injuries, specifically subarachnoid and subdural hemorrhage, are desirable. Ideally, the model should also be capable of predicting the likelihood of rib fractures, lung contusions and shoulder (clavicular and scapular) fractures, and facet, pedicle, and vertebral body fractures in the cervical spine.

Since the popularization of the Electric Vehicle (EV) there has been a large movement of consumers, governments, and the automotive industry due to its environmentally friendly characteristics. Unlike an IC engine, the batteries use multitudes of rare earth minerals and complex manufacturing processes which in some cases have been shown to produce as many emissions as an ICE vehicle over its entire lifespan. Another unnoticed important environmental concern has been the final recycling and disposal of the power train after its use. Unlike an ICE engine, which can be melted down or re-used, recycling batteries are much more difficult. In most cases the recycling process and the byproducts produced can be very harmful to the environment. This paper aims to be a complete cradle-to-grave analysis of all emissions produced in the life of an EV battery.

Data-driven modeling can help improve understanding of the governing equations for systems that are challenging to model. In the current work, the Sparse Identification of Nonlinear Dynamical systems (SINDy) is used to predict the dynamic behavior of dynamic problems for NVH applications. To show the merit of the approach, the paper demonstrates how the equations of motions for linear and nonlinear multi-degree of freedom systems can be obtained. First, the SINDy method is utilized to capture the dynamic behavior of linear systems. Second, the accuracy of the SINDy algorithm is investigated with nonlinear dynamic systems. SINDy can output differential equations that correspond to the data. This method can be used to find equations for dynamical systems that have not yet been discovered or to study current systems to compare with our current understanding of the dynamical system.