729 research outputs found
Sort by
On the Assessment of Criteria for Driver Impairment;In Search of the Golden Yardstick for Driving Performance
Get PDFMost traffic accidents may be attributed to driver impairment as a consequence of factors such as inattention, fatigue, intoxication, etc. It is now technically feasible to monitor and diagnose driver behaviour with respect to impairment with the aid of a limited number of in-vehicle sensors, however, a valid framework for the evaluation of driver impairment is still lacking. In order to provide an acceptable definition of driver impairment, a method is needed to assess both absolute and relative criteria. The seemingly paradoxical goal is to develop a definition of impaired driving that is consistent, yet adaptable to inter-individual differences
Event-Based Driver Performance Assessment
Get PDFDriving is an intermittent control task during which drivers manage their work across a number of driving and non-driving sub-tasks. The multitasking nature forces drivers to adopt situated safety margins (tolerances) in terms of time-headway (THW), time-to-collision (TTC), time-to-linecrossing (TLC), etc. Unacceptable situations (i.e. violations of safety margins defined here as events) are characterized by corrective maneuvers or actions. The frequency with which corrective actions are necessary (bandwidth), the hazard level of the situation that inspired these actions (performance), and the rate and magnitude of the responses to these situations (effort) provide a rich signature of how drivers manage their task(s). We hypothesize that drivers’ perception of performance and effort are founded in the characteristics of experienced events. This is explored by comparing driving characteristics of bus drivers who drive on the shoulder of a highway with and without the support of a haptic lateral support system (LSS). Subjective performance and effort scores extracted from a usability questionnaire and objective ones from our event-based analysis show highly significant correspondence when comparing supported versus unsupported driving. This provides validating support for the adopted event-based approach. The proposed approach offers not only sensitive metrics of driver performance and effort to evaluate ITS applications but also explanatory power by exposing the various strategically different ways drivers are affected by these systems. This method of quantifying and analyzing driver data affords new opportunities to evaluate driver responses to ITS applications
Test-Retest Reliability of Standard Deviation of Lane Position as Assessed on a PC-Based Driving Simulator
Get PDFDriving is an everyday activity that is commonly affected by neurologic disorders and medical treatments. A frequently used metric for assessing driving ability is the standard deviation of lane position (SDLP), or the amount that subjects “swerve” within their driving lane. This measurement has been used with individuals under the influence of alcohol, illicit drugs, and prescribed medications in both on-road and simulator studies. Although good test-retest reliability is critical if one is to measure change in individuals over time, there is surprisingly limited data regarding the test-retest reliability of SDLP. Objective. To examine the test-retest reliability of SDLP in subjects tested at (1) a 3-month retest interval (a time frame common to clinical trials), and (2) a year or longer retest interval (a time period over which one might track changes in neurologic patients. Methods. Group 1 completed retesting an average of 84 (s.d. = 8.1) days after their initial simulator assessment. Both HIV negative (HIV-; n = 16) and positive (HIV+; n = 13) subjects were included to explore short-term reliability in control and mildly ill patient groups. All HIV+ subjects were medically asymptomatic, and unlikely to experience HIV-related changes over this interval. Two HIV+ subjects were neuropsychologically (NP) impaired. Group 2 (n = 31), a different cohort, was retested an average of 19.8 (8.3) months after baseline. All subjects completed NP evaluations at baseline and follow-up, with NP status rated on a scale of 1 (above average) to 9 (severe impairment) by a clinician blinded to simulator performance. Twelve subjects (39%) were NP impaired. In order to examine reliability in a stable neurologic cohort, all subjects were selected because they remained at the same level of NP functioning at follow-up. SDLP was assessed in both groups using an interactive PC-based driving simulator that consisted of a monitor, steering wheel, and brake/accelerator pedals. Participants were required to maintain lane position while holding a constant speed (55 mph) and responding to divided attention tasks in the corner of the monitor. Group 2 completed an existing, standardized scenario (TOPS), while Group 1 completed a newly developed driving scenario. Both simulations lasted approximately 7 minutes. Results. Combined reliability for Group 1 was .74. Test-retest reliability was .68 for the HIVand .83 for the HIV+ subjects. For Group 2, SDLP was significantly correlated with NP functioning at baseline (r = .5, p = .005) and follow-up (r = .48, p = .006), with impaired subjects evidencing a higher SDLP than NP normal subjects at both baseline (mean of 1.9 vs 1.2; p = .006) and follow-up (1.7 vs 1.1, p = .01). Combined test-retest reliability for Group 2 was .86. The NP normal group had a test-retest reliability of .74; test-retest reliability for the NP impaired group was .87. Conclusions. SDLP is a reliable measure for periods ranging from months to years when assessed in cognitively stable subjects. As such, this may serve as a useful tool in tracking the effects of neurologic disorders and pharmacologic treatments on driving abilities
The Dimensions of Driver Performance during Secondary Manual Tasks
Get PDFThis analysis identified the underlying dimensions of driver performance using data obtained from drivers engaged in secondary manual tasks. Randomly chosen subjects balanced for age and gender used one of five advanced navigation and communication systems while driving on a closed roadway. Fifteen driver performance variables were averaged and standardized across subjects for 79 tasks. There were high correlations between all variables. Principal Component Analysis (PCA) found that the vector of loadings defining the first principal component (PC1) was positive for all 15 variables, accounting for 61 percent of the total variation across all tasks. It is interpreted as overall driver demand. PC2 loaded with one sign on event detection and response variables, but opposing sign on visual-manual workload variables. It identified tasks making drivers more inattentive to outside events than expected, given a task’s visual-manual workload, and accounted for 17 percent of total variation. It is interpreted as low-workloadbut-high-inattentiveness. PC3 had loadings of opposing sign for peripheral vs. central event variables (5 percent of total variation). It is interpreted as peripheral insensitivity. The first three components together accounted for 83 percent of total variation, which is deemed substantial. Thus most of the information available through the 15 original variables can be summarized by only three PC variables. Because the vectors of loadings defining the components are orthogonal to each other as defined by PCA, no single variable by itself can capture all the important variations in driver performance during secondary manual tasks. A multivariate design and analysis is required
Stimulus-Response Compatibilitiy Effects for Warning Signals and Steering Responses
Get PDFStimulus-response compatibility is relevant to the way a collision avoidance system signals a hazard. Using the location of a warning tone as the signal, standard spatial compatibility effects predict that it would be most beneficial to have the tone correspond to the desired response direction. However, because drivers typically turn away from sounds created by hazards, they may adopt a frame of reference where turning away from the warning tone is more compatible than responding toward it. This issue was examined in an experiment in which subjects responded to tones in the left or right ear by turning a steering wheel clockwise or counterclockwise, with the meaning of the tones manipulated to simulate warning signals. Two groups received typical compatibility instructions (tone instructions), and two received instructions specifying that the tone was a warning signal (warning instructions) indicating either the location of the danger (from which they were to turn away) or the escape direction (toward which they were to turn). The compatibility effect was in the same direction and of the same magnitude for both the warning instructions and the tone instructions. This outcome implies that instructions to turn away from danger did not cause subjects to adopt an avoidance frame of reference and that spatial correspondence was the overriding factor. The results suggest that collision avoidance systems should signal the escape direction, but these results need to be verified in simulated and actual driving conditions
Car Following by Optical Parameters
Get PDFA model for car following based solely on optical parameters was developed and compared with performance of human drivers in a simulator. The model uses the optical size of the back of the car being followed and the first derivative of its optical size as inputs. The model consists of two components: one that accelerates to maintain the visual size of the leading car, and another that accelerates to minimize changes in the rate of change of the visual size of the leading car. The simulator presented drivers with a leading car that was changing its velocity according to a sum of non-harmonic sines. Comparisons of human drivers’ performance with the models’ show a high degree of similarit
Effects of a Night Vision Enhancement System (NVES) on Driving: Results from a Simulator Study
Get PDFThree related experiments looked at the effects of an NVES on driving performance, with differences in image size ratio, lateral position and direct/indirect viewing as parameters. The experiments used experienced drivers in a fixed based virtual-reality driving simulator. Experiment 1 found that subjects using an NVES gained time to assess the situation and choose an appropriate response, which was seen in terms of better control of braking and swerving. Contrary to expectations, subjects did not drive significantly faster when using the NVES. Experiment 2 found that a 1:2 display ratio resulted in better anticipatory control without any adverse effects from differences in recognition distances. When using an NVES display displaced laterally from the normal line of sight, drivers kept the vehicle closer to the middle of the road. They also found the displaced position less favourable than one in the normal line of sight, although there were no strong negative effects of a the displacement. Experiment 3 compared a virtual (collimated) display to a direct viewing Flat Panel, with the hypothesis that reduced need of accommodation would lead to smoother driving. The results showed some differences between the two display types, although they were small compared to the effects of learning. Altogether the experiments confirmed that an NVES leads to an indisputable improvement in the drivers’ anticipatory control, and hence has considerable safety potential. This work has also emphasised the need to consider the combined effects of an NVES as a system on driving, rather than to do classical controlled experiments
Time-Sharing of a Visual In-Vehicle Task While Driving: The Effects of Four Key Constructs
Get PDFWhat affects time-sharing of a visual in-vehicle task while driving? Four key constructs are considered: (1) time pressure to complete the in-vehicle task, (2) interference of concurrent driving, (3) postponed processing while looking away from the display, and (4) the cost of task partitioning (due to forgetting and the need to reacquire last point of gaze). To examine these effects, 24 drivers were instructed to plan routes to destinations on an electronic map while driving a simulator at three levels of visual demand of the concurrent driving task, and then while parked. To force task partitioning while parked, the map was intermittently occluded. There were six combinations of viewing and occlusion time. Total task time increased significantly from 11.0 s while parked to 19.5 s on a sharp curve. Total glance time at the display, however, remained unchanged in all four levels of visual demand (12.0±0.6 s). Thus, the time pressure imposed by driving resulted in shorter glances at the display, but subjects maintained constant total glance time. Interference from concurrent driving was not significant enough to negate this effect. With task occlusion of 1 s, when viewing time decreased from 2 to 1 s, total task time increased from 15.9 to 19.7 s. Total glance time, however, remained unchanged (10.4±0.4 s) in all six timing combinations. Thus, the costs of task partitioning and the benefits of postponed processing were either small or cancelled each other in this task. The task occlusion method, as described in this paper, provides a framework for understanding the effects of in-vehicle tasks on driving
A Simulation Study of Path and Speed Through Double-Lane Roundabouts
Get PDFThis study assessed the effects of geometric curvature and lane demarcation on drivers’ selection of path and speed in double-lane roundabouts. By means of a PC-based simulator, path and speed data were collected as subjects drove twice through six roundabouts. The six roundabouts varied in terms of pavement markings and geometric curvature on the entry and exit. Seventy-five participants were tested using a fixed-base driving simulator. The results showed that drivers maintained lane position better when the roundabouts had lane demarcation than when the roundabouts had no lane demarcation. Furthermore, lane-tracking behavior for participants exposed to roundabouts with pavement markings was similar to lane-tracking behavior observed in a recent field study. Observations of speed indicated that drivers drove faster though roundabouts with a large central island radius as opposed to a roundabout with a smaller central island radiu
An Examination of the Efficacy of a Brief Educational Program on Driver Distraction
Get PDFA study examined the influence of a brief educational intervention on self-reports of past and intended future distracted driving behavior, and on perception of the danger of various driver distractions (e.g. map reading, grooming, etc). Respondents completed a series of rating scales, reporting (A) the frequency with which they had participated in various distracting activities while driving in the past, (B) the frequency with which they expected to participate in those activities in the future, and (C) the perceived danger of those activities. Half of the respondents completed the rating scales after first watching a series short animated videos from the driver education program. The remaining respondents (controls) completed the surveys prior to viewing the video segments. As compared to the control subjects, participants who viewed the video segments prior to filling out the rating scales reported reliably higher levels of perceived danger for a number of distracting activities. Video segments did not, however, appear to affect respondents’ anticipated future behavior