Search Results

A method for detecting drivers' intentions is essential to facilitate operating mode transitions between driver and driver assistance systems. We propose a driver behavior recognition method using Hidden Markov Models (HMMs) to characterize and detect driving maneuvers and place it in the framework of a cognitive model of human behavior. HMM-based steering behavior models for emergency and normal lane changes as well as for lane keeping were developed using a moving base driving simulator. Analysis of these models after training and recognition tests showed that driver behavior modeling and recognition of different types of lane changes is possible using HMMs.

The challenge of developing a robust, real-time driver gaze classification system is that it has to handle difficult edge cases that arise in real-world driving conditions: extreme lighting variations, eyeglass reflections, sunglasses and other occlusions. We propose a single-camera end-toend framework for classifying driver gaze into a discrete set of regions. This framework includes data collection, semi-automated annotation, offline classifier training, and an online real-time image processing pipeline that classifies the gaze region of the driver. We evaluate an implementation of each component on various subsets of a large onroad dataset. The key insight of our work is that robust driver gaze classification in real-world conditions is best approached by leveraging the power of supervised learning to generalize over the edge cases present in large annotated on-road datasets.

We present research in progress to develop and implement a transportable instrumentation package (TIP) to collect driver data in a vehicle. The overall objective of the project is to investigate the symbiotic relationship between humans and their vehicles. We first describe the state-of-art technologies to build the components of TIP that meet the criteria of ease of installation, minimal interference with driving, and sufficient signals to monitor driver state and condition. This method is a viable alternative to current practice which is to first develop a fully instrumented test vehicle, often at great expense, and use it to collect data from each participant as he/she drives a prescribed route. Another practice, as for example currently being used in the SHRP-2 naturalistic driving study, is to install the appropriate instrumentation for data collection in each individual's vehicle, often requiring several hours.

Accurately predicting real-world vehicle energy consumption is essential for optimizing vehicle designs, enhancing energy efficiency, and developing effective energy management strategies. This paper presents a data-driven approach that utilizes machine learning techniques and a comprehensive dataset of vehicle parameters and environmental factors to create precise energy consumption prediction models. The methodology involves recording real-world vehicle data using data loggers to extract information from the CAN bus systems for ICE and hybrid electric, as well as hydrogen and battery fuel cell vehicles. Data cleaning and cycle-based analysis are employed to process the dataset for accurate energy consumption prediction. This includes cycle detection and analysis using methods from statistics and signal processing, and then pattern recognition based on these metrics.

This paper presents the results of a study of how people interacted with a production voice-command based interface while driving on public roadways. Tasks included phone contact calling, full address destination entry, and point-of-interest (POI) selection. Baseline driving and driving while engaging in multiple-levels of an auditory-vocal cognitive reference task and manual radio tuning were used as comparison points. Measures included self-reported workload, task performance, physiological arousal, glance behavior, and vehicle control for an analysis sample of 48 participants (gender balanced across ages 21-68). Task analysis and glance measures confirm earlier findings that voice-command interfaces do not always allow the driver to keep their hands on the wheel and eyes on the road, as some assume.

This study investigated the effects of three navigation system human-machine interfaces (HMIs) on driver eye-glance behavior, navigational errors, and subjective assessments. Thirty-six drivers drove an unfamiliar 3-segment route in downtown Detroit. HMIs were 2D or 3D (level-of-detail) electronic map display + standard voice prompts, or 3D map-display augmented by photorealistic images + landmark-enhanced voice prompts. Participants drove the same three route segments in order but were assigned a different HMI condition/segment in a 3-period/3-treatment crossover experimental design. Results indicate that drivers’ visual attention using the advanced navigation systems HMIs were within US Department of Transportation recommended visual distraction limits. More turns missed in the first route segment, regardless of HMI, were attributable to greater route complexity and a late-onset voice prompt. Participant’s ratings of HMIs were influenced by the context in which that HMI was used.

This study presents a methodology for comparative analysis of seat and suspension parameters on a system level to achieve minimum occupant head displacement and acceleration, thereby improving occupant ride comfort. A lumped-parameter full-vehicle ride model with seat structures, seat cushions and five occupants has been used. Two different vehicle masses are considered. A low amplitude pulse signal is provided as the road disturbance input. The peak vertical displacement and acceleration of the occupant’s head due to the road disturbance are determined and used as measures of ride comfort. Using a design of experiments approach, the most critical seat cushion, seat structure and suspension parameters and their interactions affecting the occupant head displacement and acceleration are determined. An optimum combination of parameters to achieve minimum peak vertical displacement and acceleration of the occupant’s head is identified using a response surface methodology.

This paper presents results of a study conducted to compare driving behavior and performance of drivers in two different fixed-base driving simulators (namely, FAAC and STI) while performing a same set of distracting tasks under geometrically similar freeway and traffic conditions. The FAAC simulator had a wider three-screen road view with steering feedback as compared to the STI simulator which had a single screen and narrower road view and had no steering feedback. Twenty four subjects (12 younger and 12 mature) drove each simulator and were asked to perform a set of nine different tasks involving different distracting elements such as, using a cell phone, operating the car radio, retrieving and selecting a map from map pocket in the driver's door, collecting coins to pay toll, etc.

A new inflator specification, the “inflator thrust variable,” was developed to better explain measured mid-sized male, instrumented test dummy responses in the chest-on-module test condition. Specifically, controlled laboratory experiments were conducted with non-production, driver airbag modules with inflators of various outputs and gas constituents in an effort to assess their effects on a pertinent occupant response. Regression analyses showed that the inflator thrust variable is a better predictor of the observed variation in peak viscous criterion responses than either peak tank pressure or the related pressure rise rate when inflators of differing gas composition were compared.

The fuel energy consumption of subsonic air transportation is examined. The focus is on identification and quantification of fundamental engineering design tradeoffs which drive the design of subsonic tube and wing transport aircraft. The sensitivities of energy efficiency to recent and forecast technology developments are also examined.

This paper presents results of a three-phase research project aimed at understanding how future automotive interior materials should be selected or designed to satisfy the needs of the customers. The first project phase involved development of 22 five-point semantic differential scales to measure visual, visual-tactile, and evaluative characteristics of the materials. Some examples of the adjective pairs used to create the semantic differential scales to measure the perceptual characteristics of the material are: a) Visual: Light vs. Dark, Flat vs. Shiny, etc., b) Visual-Tactile: Smooth vs. Rough, Slippery vs. Sticky, Compressive vs. Non-Compressive, Textured vs. Non-Textured, etc., c) Evaluative (overall perception): Dislike vs. Like, Fake vs. Genuine, Cheap vs. Expensive, etc. In the second phase, 12 younger and 12 older drivers were asked to evaluate a number of different automotive interior materials by using the 22 semantic differential scales.

This paper describes a research project currently in-progress to develop a parametric model of a vehicle for use in early design stages of a new vehicle program. The model requires key input parameters to define the kind of new vehicle to be designed — in terms of details such as its intended driver/user population, vehicle type (e.g. 2-box, 3-box designs), and some key exterior and interior dimensions related to its size and proportions. The model computes and graphically displays interior package, ergonomics zones for driver controls and displays, and field of views through window openings. It also allows importing or inputting and superimposing and manipulating exterior surfaces created by a designer to assess compatibility between the interior occupant package and the vehicle exterior.

Analyses were performed to determine the sensitivity of stopping sight distance on vertical curves to driver eye height and other parameters entering into the stopping sight-distance equations. Sight distance was found to be relatively insensitive to eye height. On a given hill crest, sight distance for a driver whose eye height is 6-inches lower than the design eye height (3.75 ft) is only 5% less than the design sight distance. On the other hand, stopping distance is very sensitive to travel speed, pavement friction and reaction time. For example, a 1.8 mph decrease in speed reduces stopping distance by the same amount that a 6-inch decrease in eye height reduces sight distance. Also, sight distance is about 2.5 times more sensitive to obstacle height than eye height. It is argued that reductions in travel speed since the introduction of the 55-mph speed limit compensate for any recent or projected decreases in driver eye height.

As intelligent automated vehicle technologies evolve, there is a greater need to understand and define the role of the human user, whether completely hands-off (L5) or partly hands-on. At all levels of automation, the human occupant may feel anxious or ill-at-ease. This may reflect as higher stress/workload. The study in this paper further refines how perceived workload may be determined based on occupant physiological measures. Because of great variation in individual personalities, age, driving experiences, gender, etc., a generic model applicable to all could not be developed. Rather, individual workload models that used physiological and vehicle measures were developed.

This paper presents results of a study conducted to determine differences in older (over age 55) and younger (under age 35), male and female drivers while entering and exiting vehicles with three different body styles - namely, a large sedan, a minivan and a full-size pick-up truck. Thirty-six drivers (males and females, ages 25 to 89 years) who participated in this study were first measured for their anthropometric, strength and body flexibility measures relevant to the entry/exit tasks. They were asked to first get in each vehicle and adjust their preferred seating position. Then, they were asked to get in the vehicle and their entry time was measured. Their entry maneuver was also video taped and they were asked to rate the level of ease/difficulty (using a 5-point scale) in entering. Similar procedure and measurements were conducted during their exit from each vehicle.

Vehicle dynamics development relies on subjective assessments (SA), which is a resource-intensive procedure requiring both expert drivers and vehicles. Furthermore, development projects becoming shorter and more complex, and increasing demands on quality require higher efficiency. Most research in this area has focused on moving from physical to virtual testing. However, SA remains the central method. Less attention has been given to provide better tools for the SA process itself. One promising approach is to introduce computer-tablets to aid data collection, which has proven to be useful in medical studies. Simple software solutions can eliminate the need to transcribe data and generate more flexible and better maintainable questionnaires. Tablets’ technical features envision promising enhancements of SA, which also enable better correlations to objective metrics, a requirement to improve CAE evaluations.

With recent advances in microprocessors and data storage technologies, vehicle users can now bring or access large amounts of data in vehicles for purposes such as communication (e.g. e-mail, phone books), entertainment (e.g. music and video files), browsing and searching for information (e.g. on-board computers and internet). The challenge for the vehicle designer is how to design data displays and retrieval methods to allow data search and manipulation tasks by managing driver workload at safe acceptable levels. This paper presents a data retrieval menu system developed to assess levels of screens (depth of menu) that may be needed to select required information when a vehicle is equipped with the capability to access audio files, cell phone, PDA, e-mail and “On-star” type functions.

It is well understood that driver's steering input strongly affects lateral vehicle dynamics and excessive steering command may result in unstable vehicle motion. In a certain driving condition, it is possible for a skilled driver to prevent vehicle rollover with better perceptive capability of judging conditions and responding faster with smooth compensatory actions. This paper investigates the possibility of using active steering and wheel torque control to assist drivers in avoiding vehicle rollovers in emergency situations. The effectiveness of steering control alone and combination of steering/wheel torque control in recovery from unstable vehicle roll condition was demonstrated through simulation of both low and high vehicle speeds.

This study was conducted to develop and validate a multidimensional measure of shift quality as perceived by drivers during kick-down shift events for automatic transmission vehicles. As part of the first study, a survey was conducted among common drivers to identify primary factors used to describe subjective gear-shifting qualities. A factor analysis on the survey data revealed four semantic subdimensions. These subdimensions include responsiveness, smoothness, unperceivable, and strength. Based on the four descriptive terms, a measure with semantic scales on each subdimension was developed and used in an experiment as the second study. Twelve participants drove and evaluated five vehicles with different gear shifting patterns. Participants were asked to make kick-down events with two different driving intentions (mild vs. sporty) across three different speeds on actual roadway (local streets and highway).

ADAS (Advanced Driver Assistance System) functions can help the driver avoid accidents or mitigate their effect when they occur, and are pre-cursors to full autonomous driving (SAE defined as Level 4+). The main goal of this work is to develop a Model-Based system to actuate the Evasive Maneuver Assist (EMA) function. A typical scenario is the situation in which longitudinal Autonomous Emergency Braking (AEB) is too late and the driver has to adopt an evasive maneuver to avoid an object suddenly appearing on the road ahead. At this time, EMA can help improve the driver’s steering and braking operation in a coordinated way. The vehicle maneuverability and response performance will be enhanced when the driver is facing the collision. The function will additionally let the vehicle steer in a predetermined optimized trajectory based on a yaw rate set point and stabilize the vehicle. The EMA function is introduced with some analysis of benchmarking data.