David J. LeBlanc

Driving behaviors in early stage dementia: a study using in-vehicle technology

According to the Alzheimers Association (2011), (1) in 8 people age 65 and older, and about one-half of people age 85 and older, have Alzheimers disease in the United States (US). There is evidence that drivers with Alzheimers disease and related dementias are at an increased risk for unsafe driving. Recent advances in sensor, computer, and telecommunication technologies provide a method for automatically collecting detailed, objective information about the driving performance of drivers, including those with early stage dementia. The objective of this project was to use in-vehicle technology to describe a set of driving behaviors that may be common in individuals with early stage dementia (i.e., a diagnosis of memory loss) and compare these behaviors to a group of drivers without cognitive impairment. Seventeen drivers with a diagnosis of early stage dementia, who had completed a comprehensive driving assessment and were cleared to drive, participated in the study. Participants had their vehicles instrumented with a suite of sensors and a data acquisition system, and drove 1-2 months as they would under normal circumstances. Data from the in-vehicle instrumentation were reduced and analyzed, using a set of algorithms/heuristics developed by the research team. Data from the early stage dementia group were compared to similar data from an existing dataset of 26 older drivers without dementia. The early stage dementia group was found to have significantly restricted driving space relative to the comparison group. At the same time, the early stage dementia group (which had been previously cleared by an occupational therapist as safe to drive) drove as safely as the comparison group. Few safety-related behavioral errors were found for either group. Wayfinding problems were rare among both groups, but the early stage dementia group was significantly more likely to get lost.

Heavy-Truck Drivers’ Following Behavior With Intervention of an Integrated, In-Vehicle Crash Warning System: A Field Evaluation

Objective: This study is designed to evaluate heavy-truck drivers’ following behavior and how a crash warning system influences their headway maintenance. Background: Rear-end crashes are one of the major crash types involving heavy trucks and are more likely than other crash types to result in fatalities. Previous studies have observed positive effects of in-vehicle crash warning systems in passenger car drivers. Although heavy-truck drivers are generally more experienced, driver-related errors are still the leading factors contributing to heavy-truck-related rear-end crashes. Method: Data from a 10-month naturalistic driving study were used. Participants were 18 professional heavy-truck drivers who received warnings during the last 8 months of the study (treatment period) but not during the first 2 months (baseline period). Time headway and driver’s brake reaction time were extracted and compared with condition variables, including one between-subjects variable (driver shift) and five within-subjects variables (treatment condition, roadway types, traffic density, wiper state, and trailer configuration). Results: The presence of warnings resulted in a 0.28-s increase of mean time headway with dense on-road traffic and a 0.20-s increase with wipers on. Drivers also responded to the forward conflicts significantly faster (by 0.26 s, a 15% enhancement) in the treatment condition compared with responses in the baseline condition. Conclusion: Positive effects on heavy-truck drivers’ following performance were observed with the warning system. Application: The installation of such in-vehicle crash warning systems can help heavy-truck drivers keep longer headway distances in challenging situations and respond quicker to potential traffic conflicts, therefore possibly increasing heavy-truck longitudinal driving safety.

Learning Drivers’ Behavior to Improve Adaptive Cruise Control

Traditionally, vehicles have been considered as machines that are controlled by humans for the purpose of transportation. A more modern view is to envision drivers and passengers as actively interacting with a complex automated system. Such interactive activity leads us to consider intelligent and advanced ways of interaction leading to cars that can adapt to their drivers. In this article, we focus on the adaptive cruise control (ACC) technology that allows a vehicle to automatically adjust its speed to maintain a preset distance from the vehicle in front of it based on the driver’s preferences. Although individual drivers have different driving styles and preferences, current systems do not distinguish among users. We introduce an approach to combine machine learning algorithms with demographic information and behavioral driver models into existing automated assistive systems. This approach can reduce the interactions between drivers and automated systems by adjusting parameters relevant to the operation of these systems based on their specific drivers and context of drive. We also learn when users tend to engage and disengage the automated system. This approach sheds light on the kinds of dynamics that users develop while interacting with automation and can teach us how to improve these systems for the benefit of their users. While generic packages such as Weka were successful in learning drivers’ behavior exclusively based on the ACC’s sensors, we found that improved learning models could be developed by adding information on drivers’ demographics and a previously developed model about different driver types. We present the general methodology of our learning procedure and suggest applications of our approach to other domains as well.

Accelerated Evaluation of Automated Vehicles in Car-Following Maneuvers

The safety of automated vehicles (AVs) must be assured before their release and deployment. The current approach to evaluation relies primarily on 1) testing AVs on public roads or 2) track testing with scenarios defined in a test matrix. These two methods have completely opposing drawbacks: the former, while offering realistic scenarios, takes too much time to execute and the latter, though it can be completed in a short amount of time, has no clear correlation to safety benefits in the real world. To avoid the aforementioned problems, we propose accelerated evaluation, focusing on the car-following scenario. The stochastic human-controlled vehicle (HV) motions are modeled based on 1.3 million miles of naturalistic driving data collected by the University of Michigan Safety Pilot Model Deployment Program. The statistics of the HV behaviors are then modified to generate more intense interactions between HVs and AVs to accelerate the evaluation procedure. The importance sampling theory was used to ensure that the safety benefits of AVs are accurately assessed under accelerated tests. Crash, injury and conflict rates for a simulated AV are simulated to demonstrate the proposed approach. Results show that test duration is reduced by a factor of 300 to 100 000 compared with the non-accelerated (naturalistic) evaluation. In other words, the proposed techniques have great potential for accelerating the AV evaluation process.

Accelerated Evaluation of Automated Vehicles Using Piecewise Mixture Models

The process to certify highly automated vehicles has not yet been defined by any country in the world. Currently, companies test automated vehicles on public roads, which is time-consuming and inefficient. We proposed the accelerated evaluation concept, which uses a modified statistics of the surrounding vehicles and the importance sampling theory to reduce the evaluation time by several orders of magnitude, while ensuring the evaluation results are statistically accurate. In this paper, we further improve the accelerated evaluation concept by using piecewise mixture distribution models, instead of single parametric distribution models. We developed and applied this idea to forward collision control system reacting to vehicles making cutin lane changes. The behavior of the cutin vehicles was modeled based on more than 403,581 lane changes collected by the University of Michigan Safety Pilot Model Deployment Program. Simulation results confirm that the accuracy and efficiency of the piecewise mixture distribution method outperformed single parametric distribution methods in accuracy and efficiency, and accelerated the evaluation process by almost four orders of magnitude.

Development and validation of an errorable car-following driver model

An errorable car-following driver model was presented in this paper. This model was developed for evaluating and designing of active safety technology. Longitudinal driving was first characterized from a naturalistic driving database. The stochastic part of longitudinal driving behavior was then studied and modeled by a random process. The resulting stochastic car-following model can reproduce the normal driver behavior and occasional deviations without crash. To make this model errorable, three error-inducing behaviors were analyzed. Perceptual limitation was studied and implemented as a quantizer. Next, based on the statistic analysis of the experimental data, the distracted driving was identified and modeled by a stochastic process. Later on, time delay was estimated by recursive least square method and was modeled by a stochastic process as well. These two processes were introduced as random disturbance of the stochastic driver model. With certain combination of those three error-inducing behaviors, accident/incident could happen. Twenty-five crashes happened after eight million miles simulation (272/100M VMT). This simulation crash rate is higher by about twice with 2005 NHTSA data (120/100M VMT).

Anticipatory speed control model applied to intersection left turns

This paper presents a new speed control model applicable to real-world driving. It is developed for intersection left turns and is based on anticipated acceleration reference (AAR) inputs. This addresses combined visual anticipation of lateral and longitudinal accelerations for the approach to an intersection where both stopping and turning outcomes are possible. The relationship between the AAR and the resulting vehicle accelerations are studied for both stopping and turning events using naturalistic driving data. A closed-loop model is developed, including braking to rest when the left turn is not attempted and for the turn and exit stages when it is. Parameter ranges are estimated, and as a demonstration of model applicability, Monte Carlo simulations are conducted to generate representative left turns using a full simulation model. Extension of the AAR model to other speed control problems, for example, driving on curved roads, is also discussed.

Longitudinal driving behavior with integrated crash-warning system

This study created the most extensive set of naturalistic data that has ever been gathered on the following behavior of drivers when interacting with a forward crash-warning system. For the purposes of this paper, data from the naturalistic driving study of the Integrated Vehicle-Based Safety System (IVBSS) program were used. IVBSS data collected from a total of 108 drivers, representing 81,163 steady state following events and 20,096 forward conflict events were extracted and compared. Drivers were from three age groups (younger, middle-aged, and older) and balanced between two gender groups. Three objective measures were used in this study: mean time headway, minimum time to collision, and proportion of time drivers spent in time headway of 1 s or less. Drivers used the research vehicles for 40 days, with the system not activated for the first 12 days and activated for the following 28 days. A linear mixed model was used for the data analysis. Results of this study show that drivers have a tendency to follow more closely when the warning system is activated. It is recommended that a visual display for feedback on real-time safe following distance may help drivers keep a safer distance. This study also observed age-related self-regulation behavior when other vehicles were being followed and showed that older drivers tended to follow farther away from the leading vehicle.

Gap Acceptance During Lane Changes by Large-Truck Drivers—An Image-Based Analysis

This paper presents an analysis of rearward gap acceptance characteristics of drivers of large trucks in highway lane change scenarios. The range between the vehicles was inferred from camera images using the estimated lane width obtained from the lane tracking camera as the reference. Six-hundred lane change events were acquired from a large-scale naturalistic driving data set. The kinematic variables from the image-based gap analysis were filtered by the weighted linear least squares in order to extrapolate them at the lane change time. In addition, the time-to-collision and required deceleration were computed, and potential safety threshold values are provided. The resulting range and range rate distributions showed directional discrepancies, i.e., in left lane changes, large trucks are often slower than other vehicles in the target lane, whereas they are usually faster in right lane changes. Video observations have confirmed that major motivations for changing lanes are different depending on the direction of move, i.e., moving to the left (faster) lane occurs due to a slower vehicle ahead or a merging vehicle on the right-hand side, whereas right lane changes are frequently made to return to the original lane after passing.

Accelerated Evaluation of Automated Vehicles in Lane Change Scenarios

It is important to rigorously and comprehensively evaluate the safety of Automated Vehicles (AVs) before their production and deployment. A popular AV evaluation approach is Naturalistic-Field Operational Test (N-FOT) which tests prototype vehicles directly on public roads. Due to the low exposure to safety-critical scenarios, N-FOTs is time-consuming and expensive to conduct. Computer simulations can be used as an alternative to N-FOTs, especially in terms of generating motions of the surrounding traffic. In this paper, we propose an accelerated evaluation approach for AVs. Human-controlled vehicles (HVs) were modeled as disturbance to AVs based on data extracted from the Safety Pilot Model Deployment Program. The cut-in scenarios are generated based on skewed statistics of collected human driver behavior, which amplifies riskier testing scenarios while reserves its statistical information so that the safety benefits of AV in non-accelerated cases can be accurately estimated. An AV model based on a production vehicle was tested. Results show that the proposed method can accelerate the evaluation process by at least 100 times.Copyright