Display:

Results

Viewing 1 to 30 of 692
2015-04-14
Technical Paper
2015-01-0216
Ping-Min Hsu, Ming Hung Li, Kuo-Ching Chang
This paper studies noise filtering in an autonomous emergency braking (AEB) system with a sensor fusion between a millimeter wave (MMW) radar and a camera. The AEB system may automatically brake cars by mistake due to the sudden appearance of noise signal if it does not involve any noise filtering mechanism for the object sensor. This may cause some accidents. To avoid such accidents, we propose the filters for three kinds of noise—named as twice harmonic noise, ground noise, and specular refection noise—which are commonly suffered while applying MMW radars. The former is caused by the reflection of a radar wave between a target object and the MMW radar. Based on the concern that the MMW radar evaluates the sensing distance based on the time interval between the transmitting and receiving of the radar wave, one of the sensing distances would be twice as longer as one of others due to the wave reflection. The object featuring this characteristic is further filtered.
2015-04-14
Technical Paper
2015-01-1399
Dee Kivett, Victor Gallas Cervo, Aparna Mantha, John Smith
A common result of aging is a decline in peripheral vision. This study provides a preliminary feasibility analysis of an improved method for alerting older drivers of oncoming traffic in blind-spots. Luminescence with an intuitive color-scheme is used as the primary stimulus to permeate a wider field of useful vision than that of existing technology in use today. This method was developed based on concepts of affordance-based design through its adaptation to address specific cognitive and visual acuity challenges of the elderly. The study involved evaluation of alert recognition times among drivers ranging in age from 16 to 82 and was performed in a driving simulator. The result is an improved, intuitive technique for hazard alert that shows significant improvement over existing technology for all age groups, not just the elderly. The results highlight the significance of optimization of alert placement within the useful field of view of elderly drivers.
2015-04-14
Technical Paper
2015-01-0215
Reena Kumari Behera, Smita Nair, Vinay Vaidya
Redundancy plays a key role in increasing the computation time in case of most vision based systems. In vision based applications, the images captured from the camera are processed pixel by pixel in order to get the desired information. There is plenty of redundant data in most of the images. Removal of this unwanted data would help in increasing the processing time considerably. This paper presents a simple yet novel approach to remove unwanted data from a given image. The proposed work focuses on clutter removal from outdoor scenes. More specifically, this approach would be more applicable to Advanced Driver Assistance Systems (ADAS). An outdoor scene captured consists of two main parts a) ground region consisting of the road area and other lane markings especially white or yellow in color. b) The background region consisting of various structures, trees, sky etc. The definition of unwanted regions depends on the application.
2015-04-14
Technical Paper
2015-01-0213
Vinuchackravarthy Senthamilarasu, Anusha Baskaran, Krishnan Kutty
In the research field of automotive systems, Advanced Driver Assistance Systems (ADAS) are gaining of paramount importance. As significance for such systems increase, the challenges associated with it also increases. These challenges can arise due to technology, human factors, or due to nature (haze, fog etc.) In terms of visibility for the drivers as well as in vision based ADAS, haze formation in the atmosphere poses the challenging problem. In this paper, the proposed method addresses a novel technique of enhancing the quality in terms of visibility and visual perception of the haze affected images. Using HSV color space and the haze model, the haze affected images are recuperated. The proposed procedure involves retaining of hue (H) and scaling of saturation (S) value of each pixel between the haze input and de-hazed output images. In addition, a simple method for manipulating the ‘V’ space to de-haze the input image is also proposed.
2015-04-14
Technical Paper
2015-01-0256
Changbo Fu, Paul (Tim) Freeman, John R. Wagner
Abstract: **Driver modeling is essential to both vehicle design and control unit development. It can improve the understanding of human driving behavior and decrease the cost and risk of vehicle system verification and validation. In this paper, three driver models were implemented to simulate the behavior of drivers subject to a run-off-road recovery event. Target path planning, pursuit behavior, compensate behavior, physical limitations, and neuromuscular modeling were taken into consideration in the feedforward/feedback driver model. A transfer function driver model and a cost function based driver model from a popular vehicle simulation software were also simulated and a comparison of these three models was made. The feedforward/feedback driver model exhibited the best balance of performance with smallest overshoot (0.226m), medium settling time (1.20s) and recovery time (4.30s).
2015-04-14
Technical Paper
2015-01-0259
Tyler Zellmer, Julio Rodriguez, John R. Wagner, Kim Alexander, Philip Pidgeon
Abstract According to the National Highway Traffic Safety Administration (NHTSA), motor collisions account for nearly 2.4 million injuries and 37 thousand fatalities each year in the United States. A great deal of research has been done in the area of vehicular safety, but very little has been completed to ensure licensed drivers are properly trained. Given the inherent risks in driving itself, the test for licensure should be uniform and consistent. To address this issue, an inexpensive, portable data acquisition and analysis system has been developed for the evaluation of driver performance. A study was performed to evaluate the system, and each participant was given a normalized driver rating. The average driver rating was µ=55.6, with a standard deviation of σ=12.3. All but 3 drivers fell into the so-called "Target Zone", defined by a Driver Rating of µ± 1σ.
2015-04-14
Technical Paper
2015-01-1706
Sreegururaj Jayachander, Krishna Raj Nair M K
Melatonin, otherwise popularly known as the "sleep hormone" is known to govern the human circadian rhythms. Current studies indicate that the generation of melatonin is impacted by the ambient light. The natural sleep inducing behavior during night and in darkness, is also due to the same phenomenon. Studies have shown that light of particular wavelengths in the visible spectrum affect the amount of melatonin secreted by the human body. Blue light in the wavelengths of around 468 nm is known to inhibit the melatonin secretion, the most. This branch of science known as photo-biology is in its nascent stage and is a matter of research pursued by endocrinologists and other lighting researchers. Photo-biology has several potential applications in the automotive industry, the principal one being driver drowsiness prevention.
2015-04-14
Technical Paper
2015-01-0258
Venkatesh Kareti, Priti Ranadive, Vinay Vaidya
Various Advanced Driver Assists Systems (ADAS) are being used today to increase safety of drivers. These systems viz. Forward Collision Warning (FCW), Lane Departure Warning (LDW), Pedestrian Detection (PD), are all based on inputs captured using a front mounted camera. It would be useful to combine all these applications together and process the same input for different application purpose. Additionally, multicore processors are now easily available and can be used for integrating multiple ADAS applications. This would lead to reduced cost and maintenance of ADAS systems with the same performance benefits. Since current ADAS applications are sequential and/or use single core processors there is a need to parallelize these applications so that multiple cores can be utilized optimally. In this paper, we discuss our experiments and results while attempting to integrate two such ADAS applications on a multicore embedded platform.
2015-04-14
Technical Paper
2015-01-0310
R Danymol, Krishnan Kutty
Camera sensors that are made of silicon photodiodes, used in ordinary digital cameras are sensitive to visible as well as NIR wavelength. However, since the human vision is sensitive only in the visible region, a hot mirror/infrared blocking filters are used in cameras. Certain complimentary attributes of NIR data are, therefore, lost in this process. RGB and NIR images are captured in entirely two different spectra/wavelength, thereby retaining different information. In this paper, an attempt is made to estimate an NIR image from a given optical image. This was undertaken using the compressed sensing framework. The NIR data estimation is formulated as an image recovery problem in compressed sensing. The NIR data is considered as missing pixel information and its approximation is done during the image recovery phase. Thus for a given optical image, with NIR data being considered as missing information, the recovered NIR data gives the corresponding NIR image.
2015-04-14
Technical Paper
2015-01-0309
Mayurika Chatterjee, Atchyuta Rao, Chaitanya Rajguru
Parking assist systems have become very common in current vehicles. The purpose of such a system is to assist the driver to park the vehicle without collision. The sensors serve as eyes of the driver during parking maneuver by sensing any obstacle in the path. The parking sensors, typically ultrasonic sensors, are mounted on front and rear of vehicle to assist the driver to park the vehicle. Thus, such a system can cover only the front and rear portion of the vehicle and is unable to cover the side portions of the vehicle. This paper proposes a novel method to monitor the perimeter of a vehicle while parking using minimum sensors placed at strategic locations. A local map of the parking area is generated using data from sensors which helps in identifying static obstacles. The map is constantly updated in real time during parking. The algorithm ascertains that the entire perimeter of the vehicle is protected from impending collisions in real time.
2015-04-14
Technical Paper
2015-01-0312
Jiji Gangadharan, Shanmugaraj Mani, Krishnan Kutty
Advanced driver assistance systems have become an inevitable part of most of the modern cars. Their use is mandated by regulations in some cases; and in other cases where vehicle owners have become more safety conscious. Vision / camera based ADAS systems are widely in use today. However, it is to be noted that the performance of these systems depends on the quality of the image/video captured by the camera. Low illumination is one of the most important factors which degrade the image quality. In order to improve the system performance under low light condition, it is required to first enhance the input images/frames. In this paper, an image enhancement algorithm is proposed that would automatically enhance images to a near ideal condition. This is accomplished by mapping features taken from images acquired under ideal illumination conditions on to the target low illumination images/frames.
2015-04-14
Technical Paper
2015-01-0311
Reecha Yadav, Vinuchackravarthy Senthamilarasu, Krishnan Kutty, Vinay Vaidya, Sunita Ugale
In view of the continuous efforts by the automotive fraternity, for achieving traffic safety, detecting pedestrians from image/video has become an extensively researched topic in recent times. The task of detecting pedestrians in the urban traffic scene is complicated by the considerations involving pedestrian figure size, articulation, fast dynamics, background clutter, etc. A number of methods using different sensor technologies have been proposed in the past for the problem of pedestrian detection. To limit the scope, this paper reviews the techniques involved in day-time detection of pedestrians, with emphasis on the methods making use of a monocular visible-spectrum sensor. The paper achieves its objective by discussing the basic framework involved in detecting a pedestrian, while elaborating the requisites and the existing methodologies for implementing each stage of the basic framework.
2015-04-14
Technical Paper
2015-01-0265
Martin Krammer, Philip Stirgwolt, Helmut Martin
A single malfunction in a product can initiate a costly product recall. With ever increasing reliance on electronic and electric components contributing to the functional safety of next generation vehicles, a malfunction may cause such inconvenience too. Automotive E/E systems perform highly networked functions, including dense system interactions, and with a large number of product variants and features, complexity increases beyond human susceptibility. This is considered to be the main source for systematic faults and does not only affect newly developed systems, like complex driver assistance systems. Also well-known systems often bear a non-negligible safety threat. Safety critical automotive product development starts with a comprehensive and well-coordinated concept phase.
2015-04-14
Technical Paper
2015-01-0277
Seth Placke, John Thomas, Dajiang Suo
Automobiles are becoming ever more complex as advanced safety features are integrated into the vehicle platform. As the pace of integration and complexity of new feature rises, it is becoming increasingly difficult for system engineers to assess the impact of new additions on vehicle safety and performance. In response to this challenge, a new approach for analyzing multiple control systems using the Systems Theoretic Process Analysis (STPA) framework has been developed. The new approach meets the growing need of system engineers to analyze integrated control systems, that may or may not have been developed in a coordinated manner, and assess them for safety and performance. The new approach identifies unsafe combinations of control actions, from one or more control systems, that could lead to an accident. For example, independent controllers for Auto Hold, Engine Idle Stop, and Adaptive Cruise Control may interfere with each other in certain situations.
2015-04-14
Technical Paper
2015-01-1464
Qiang Chen, Miao Lin, Bing Dai, Jiguang Chen
The objective of this work was to describe typical accident scenarios for pedestrian accidents in China. The accident analysis aims to develop test procedures for assessing Autonomous Emergency Braking (AEB) systems. Beyond that, this study was also with the goal of estimating the effectiveness of potential reduction of fatally and severely injured pedestrians by AEB systems.Based on statistics, more than 25% of traffic fatalities were pedestrians in China. Autonomous Emergency Braking (AEB) systems are already penetrating the vehicle market and are designed to offer protection against the occurrence and severity of collisions. However there is a need to evaluate the systems and their effectiveness. Test methods for such active safety systems are being developed and will be implemented in NCAP tests in the near future, e.g. 2016 in Euro-NCAP, and most probably in 2018 C-NCAP tests (still in consulting phase).
2015-04-14
Journal Article
2015-01-1595
Kristoffer Lundahl, Chih Feng Lee, Erik Frisk, Lars Nielsen
Rollover has for long been a major safety concern for trucks, and will be even more so as automated driving is envisaged to becoming a key element of future mobility. A natural way to address rollover is to extend the capabilities of current active-safety systems with a system that intervenes by steering or braking actuation when there is a risk of rollover. Assessing and forecasting the rollover is usually performed using rollover indices, which can be calculated either from lateral acceleration, lateral load transfer, or roll energy. Since these indices are evaluated based on different physical observations it is unclear how they can be compared and how well they reflect rollover events in different situations. In this paper we investigate the implication of the above mentioned rollover indices, in different critical maneuvers, for a heavy 8x4 twin-steer truck.
2015-04-14
Technical Paper
2015-01-0316
Kazuto Yokoyama, Masahiro Iezawa, Yohei Akashi, Toshihide Satake, Yukiyasu Akemi, Satoru Inoue, Ryotaro Suzuki
Parking assist systems which relieve burden of drivers have been put into practice in the world. Mitsubishi Electric has also been developing several technologies to achieve the system. Peripheral environment around the car in a parking lot is detected with Mitsubishi Electric’s ultrasonic sensors, and an algorithm makes a map of the environment to determine whether or not the car can park at an available space. On the basis of the created map, a smooth and efficient path to drive and park the car is generated with an optimization technique. While the car is moving to the available space, the position and attitude of the vehicle is estimated from wheel speed sensors and yaw rate one. Steering is automatically controlled with a Mitsubishi Electric’s electric power steering system. In particular, this paper focuses on a speed control which can be applied to an automatic driving control including an automatic parking system.
2015-04-14
Technical Paper
2015-01-1404
Arda Kurt, Güchan Özbilgin, Keith A. Redmill, Rini Sherony, Ümit Özgüner
In this paper, a series of design, development, and implementation details for testing and evaluation of Lane Departure Warning and Prevention systems are being discussed. The approach taken to generate a set of repeatable and relevant test scenarios and to formulate the test procedures to ensure the fidelity of the collected data includes initial statistical analysis of applicable statistics; growth and probabilistic pruning of a test matrix; simulation studies to support procedure design; and vehicle instrumentation for data collection. The success of this comprehensive approach strongly suggests that the steps illustrated in this paper can serve as guidelines towards a more general class of vehicular safety and advanced driver assistance systems evaluation.
2015-04-14
Technical Paper
2015-01-1401
Pawel Skruch, Rafal Dlugosz, Krzysztof Kogut, Pawel Markiewicz, Dominik Sasin, Maciej Różewicz
Active Safety (AS) and Advanced Driver Assistance Systems (ADAS) can be considered nowadays as a distributed embedded software system where independent microprocessor systems, called Electronic Control Units (ECUs), communicate together using different communication protocols. Typical AS or ADAS functionality is then realized by several ECUs communicating each other. AS and ADAS systems interact with other ECUs in a vehicle, usually, via communication networks and with vehicle’s surroundings via camera, radar, or laser sensors. Quality assurance and safety standards combined with increasing complexity and reliability demands related to vision sensing, radar sensing and data fusion, often together with a short time to market makes the development of such systems challenging. As the number of important for the system road scenarios grows, mathematical modelling and computer simulation become important engineering tasks to assure required quality and compliance to safety standards.
2015-04-14
Technical Paper
2015-01-1410
Shotaro Odate, Kazuhiro Daido, Yosuke Mizutani
According to the National Automotive Sampling System (NASS) Crashworthiness Data System (CDS), which is a North American automobile accident database, collision events referred to as multiple-collision accidents, in which multiple collisions occur during travel, account for 55% of all accidents. In addition, multiple-collision accidents in which collision events following the first collision event are more severe than the first event account for 20% of all accidents. In a first collision, the system had simultaneously operated to restrain and protect the vehicle occupant. If the multiple-collision accidents occurs, because the system for restraining and protecting vehicle occupants will have already deployed, the performance of the system can be limited from subsequent collisions.
2015-04-14
Technical Paper
2015-01-1412
Xuan Zhou, Walter Niewoehner
The presentation and paper will explain a new algorithm-based approach to fasten future advanced driver assistance systems (ADAS) with simultaneously increasing the reliability. In the past the TTC (time to collision) was used to calculate and forecast traffic situations leading to a collision with an obstacle moving in front of the own vehicle in the same direction or coming towards it. Situations with the trajectories of the involved parties not being in-line (e.g. crossing, lane changing manoeuvres), parties changing the direction of movement, or parties changing the speed are more complex to calculate. The new method developed bases on an algorithm using the data from the area under sensor surveillance (e.g. by radar) to calculate a so-called collision tube. The collision tube describes the relative trajectories of the own vehicle and those of other traffic participants or obstacles.
2015-04-14
Technical Paper
2015-01-1411
Caroline Crump, David Cades, Robert Rauschenberger, Emily Hildebrand, Jeremy Schwark, Brandon Barakat, Douglas Young
Advanced Driver Assistive System technologies are currently available in many passenger vehicles that provide safety benefits and will ultimately lead to autonomous, “self-driving” vehicles. One technology that has the potential for having substantial safety benefits is the forward collision warning and mitigation (FCWM) system, which is designed to (1) warn drivers of imminent front-end collisions, (2) potentiate driver braking responses, and (3) have the ability to apply the vehicle’s brake autonomously to slow, or, in some cases, stop the vehicle prior to a forward collision. Although the proliferation of such technologies can, in many ways, mitigate the necessity of a timely braking response by a driver in an emergency situation, how this system affects a driver’s overall ability to safely, efficiently, and comfortably operate a motor vehicle remains unclear.
2015-04-14
Technical Paper
2015-01-1413
Louis Tijerina, Michael Blommer, Reates Curry, Radhakrishnan Swaminathan, Dev Kochhar, Walter Talamonti
Objective: Investigate statistical effects of repeated measures design in FCW (warning vs. no-warning ) evaluation Background: Repeated measures designs are often used in FCW testing despite concerns that 1st exposure creates expectancy effects which may dilute or bias outcomes Method: 32 participants were divided into groups of 8 for an AA, BB, AB, BA design (A= no warning; B=FCW). They drove in a high-fidelity, motion-base simulator with a visual distraction task. After some 25 minutes of driving a simulated nighttime rural highway, a high-intensity forward collision threat arose during the distraction task. Response time was analyzed. Results: There was evidence of differential carryover and significant Period 1 vs. 2 effects which dilute the magnitude of difference between FCW and no warning relative to 1st exposure only. Also there was a trend toward slower response with no-warning after FCW exposure as first exposure than after no-warning as first exposure.
2015-04-14
Technical Paper
2015-01-1211
Zhuoping Yu, Caitao Jian, Songyun Xu, Lu Xiong
This paper is to research the dynamic response of active power source (APS) of electro hydraulic brake (EHB) system, which is a new design system. The dynamic response of APS could impact the effectiveness of brake system, the reason is that the slow-response of APS will cause the slow speed of building pressure and increasing the braking distance. So researching the dynamic response of active power source is very important for brake system. First of all, the part is about the components and working principle of EHB system and some parameters of active power source. EHB systemconsists of active power source (APS), pedal feel emulator (PFE), electro control unit (ECU) and hydraulic control unit (HCU). APS includes motor, master cylinder and retarding mechanism consisting of warm, gear and gear rack. Secondly, this paper proposes a restricted distribution control strategy——the control strategy of restricted distribution (CSRD).
2015-04-14
Technical Paper
2015-01-0214
Ramya Deshpande, Krishnan Kutty, Shanmugaraj Mani
In modern cars, the Advanced Driver Assistance Systems (ADAS) is cardinal point for safety and regulation. The proposed method detects visual saliency regions in a given image. Multiple ADAS systems require many sensors and multicore processors for fast processing of data in real time; which leads to the increase in cost. In order to balance the cost and safety, the system should process only required information and neglect the rest. Human visual system perceives only important content in the scene while leaving rest of portions unprocessed. The studies on human psycho visual system hypothesize similar behavior in human perception. The proposed method aims to model the similar behavior in computer vision with the concept of visual saliency. Saliency in still images is computed by color, frequency and positional difference. A region is salient, if its color or pattern is unique. The color difference between the regions in Lab Space highlights the visual difference.
2015-04-14
Journal Article
2015-01-0313
Ugo Rosolia, Francesco Braghin, Andrew Alleyne, Edoardo Sabbioni
This paper presents a nonlinear control approach to achieve good performances in vehicle path following and collision avoidance when the vehicle is driving under cruise highway conditions. Nonlinear model predictive control (NLMPC) is adopted to achieve online trajectory control based on a simplified vehicle model. GMRES/Continuation algorithm is used to solve the online optimization problem. Simulation and experimental results show that the proposed controller is capable of tracking the desired path as well as avoiding the obstacles.
2015-04-14
Technical Paper
2015-01-0493
Ying Wang, Ye Wang, You Qu, Sumin Zhang, Weiwen Deng
Abstract Vision-based Advanced Driver Assistance Systems has achieved rapid growth in recent years. Since vehicle field testing under various driving scenarios can be costly, tedious, unrepeatable, and often dangerous, simulation has thus become an effective means that reduces or partially replaces the conventional field testing in the early development stage. However, most of the commercial tools are lack of elaborate lens/sensor models for the vehicle mounted cameras. This paper presents the system-based camera modeling method integrated virtual environment for vision-based ADAS design, development and testing. We present how to simulate two types of cameras with virtual 3D models and graphic render: Pinhole camera and Fisheye camera. We also give out an application named Envelope based on pinhole camera model which refers to the coverage of Field-of-Views (FOVs) of one or more cameras projected to a specific plane.
2015-04-14
Journal Article
2015-01-1519
Robert Suender, Günther Prokop, Thomas Roscher
Starting from the USA in 2008 and followed by the European Union commencing in 2012, legal requirements concerning “Tire Pressure Monitoring Systems” (TPMS) for passenger cars and light trucks will be introduced in China as well and therefore in the third of the three largest automobile markets worldwide. Changes of pressure dependent physical tire properties such as dynamic roll radius and a certain tire Eigen mode, which are included in the ABS-wheel speed signals, indicates pressure loss in an indirect manner. Systems with corresponding working principles are called “indirect Tire Pressure Monitoring System” (iTPMS). Since the tire is a structural element with varying characteristics according to the design parameters, the roll radius and frequency behavior due to pressure loss is variable as well. As a consequence, tires have to be evaluated regarding there compatibility to iTPMS during the vehicle development process.
2015-04-14
Journal Article
2015-01-1403
Yi lu Murphey, Dev S. Kochhar, Paul Watta, Xipeng Wang, Tianyu Wang
A host of new technologies, features and functions are continuously being added to vehicles to make the driving task and journey safe, pleasant, relaxing, enjoyable, and even exciting for the driver. An encompassing framework for research has been to understand and push further the need for ‘driver wellness’, the definition for which is still elusive. Suffice to say that ‘wellness’ is reflected in feeling good before, during and after the drive. Objective measures, primarily driver physiology, reflect wellness, but in an as yet not fully understood way. Murphey and Kochhar [1, 2] developed a Transportable Instrument Package (TIP) for in-vehicle on-the road driving data recording, and used machine learning and neural networks to explore the underlying relationships. In this paper we report on research that shows how in-vehicle, on-the-road driver physiological measures can be used to predict the driver’s intention to change lanes, even before such a lane-change is initiated.
2015-04-14
Journal Article
2015-01-1408
Kristofer D. Kusano, Hampton C. Gabler
Intersection crashes are a frequent and dangerous crash mode in the U.S. Emerging Intersection Advanced Driver Assistance Systems (I-ADAS) aim to assist the driver to mitigate the consequences of vehicle-to-vehicle crashes at intersections. In support of the design and evaluation of such intersection assistance systems, characterization of the road, environment, and drivers associated with intersection crashes is necessary. The objective of this study was to characterize intersection crashes using nationally representative crash databases that contained all severity, serious injury, and fatal crashes. This study utilized four national crash databases: the National Automotive Sampling System, General Estimates System (NASS/GES); the NASS Crashworthiness Data System (CDS); and the Fatality Analysis Reporting System (FARS) and the National Motor Vehicle Crash Causation Survey (NMVCCS).
Viewing 1 to 30 of 692

Filter