Search

Reliable measures of driving skill are important for driving simulators to play a larger role in driver education and training. In this study we measured the adequacy of left turns of 12 novice and 12 experienced drivers using a new performance measure. We called this Entry Lane Position Error (ELPE) and defined it as the perpendicular distance from the center of the lane being entered to the center of the vehicle's width at the point of new lane entrance. The different between average scores of ELPE for the novice and experienced groups was highly significant. In a driving simulator the participant's position and the road database are always available. Participants should be able to receive near real-time feedback as to the goodness of their lane turns. Visual playback of left turns is also easily implemented on most driving simulators We hope to investigate and analyze other simple performance measures that can reflect the skill level of novice drivers.

In order to be an effective tool for driver evaluation and education, driving simulators need to be better designed to reduce simulator sickness. This study investigated driving in four environments (country, suburban, city, and curves) using a simulator. When driving on straight roads (city and suburban environments) subjects reported less simulator sickness then driving in the city environment (which included left and right turns) and on curves. A mini-SSQ was used to measure the accumulation of simulator sickness over trials. From trial 1 to trial 5, reported simulator sickness increased linearly. From trial 5 through 8, the rate of increase in simulator sickness decreased. We suggest that the rapid and distorted optic flow experienced while executing turns and driving on curves in driving simulators makes a substantial contribution to simulator sickness.

This article reports on a study that used a driving simulator to display conditions of high and low optic flow. The study included 30 drivers who were requested to produce vehicle speeds of either 30 or 60 mph with the geometric field of view at 25, 55, or 85 visual degrees. Drivers overestimated the production of 30 by 20 mph. It was also found that the production of speed was highly dependent on the geometric field of view. These results suggest that optic flow presented in a driving simulator display does not correspond with optic flow found in real-world driving.

Movement disorders associated with Essential Tremor (ET) and Parkinson’s disease (PD) can negatively impact use of the upper limb for many precision tasks, including handwriting. Both ET and PD can be assessed through clinical tests which are, however, relatively subjective. This assessment approach possesses inherent logistical and resolution limitations. To address this, here we present objective computerized metrics intended to assess and quantify the extent to which static writing samples display the effects of ET and PD. Specifically, these metrics are tested in their ability to measure tremor by comparing unaffected writing samples with those affected by artificially induced tremor on healthy subjects, and also by comparing healthy writing samples with symptomatic writing samples collected from PD patients reporting micrographia. Our findings indicate that the presented metrics can be utilized for assessment, leading to a toolset capable of objectively monitoring static handwriting changes associated with symptom variations in ET and/or PD patients.Copyright © 2015 by ASME

Active feedback control is utilized in this study in order to regulate pen-tip deviations in a novel pen design with the aim to minimize the effects of hand tremors on handwriting. The pen comprises a pendulum-like pen-rod that swings inside a tubular shaped pen casing, and between the pen and the casing, certain compliance and active actuation is considered. Since by the nature of the system dynamics, arbitrary pole placement is not possible in the design of the controller, a nonlinear optimization scheme is constructed to design the controller gains. With these gains, pen-tip deviations are minimized (≈ −47 dB) when the pen casing is subjected tremor-induced cyclic disturbances, and pen-tip response against impulsive perturbations is satisfactorily improved (settling time ≈ 1 sec) while keeping the controller effort around 2 N. Simulation studies are presented comparing the efficacy of the proposed controller with respect to a passively controlled pen, along with trade-offs within the design parameter space.Copyright © 2015 by ASME

Essential Tremor (ET) manifests itself in people in the form of uncontrolled shaking, most notably during skeletal muscle usage. Severity of ET is commonly assessed through a set of clinical handwriting tests, the subjective analyses of which limit resolution and availability. To address this, our group is pursuing objective and computerized metrics to assess and quantify the extent of tremor in pre-existing static writing samples. In the current study, we tested these metrics by comparing unaffected writing samples with writing affected by artificially-induced tremor. Our findings suggest that these metrics can be utilized for assessment, leading to a toolset for objective monitoring of static handwriting changes associated with ET symptoms.

Using a virtual reality driving simulator, the behavior of 12 novice and 12 experienced drivers was recorded during three right-to-left lane-change scenarios. The first scenario had no vehicles in the left lane, the second had one rear-approaching vehicle in the left lane, and the third had three vehicles approaching from the rear left lane. Each lane-change maneuver was composed of 1) a preparatory period, 2) the actual steering initiative from the right lane to the left lane, and 3) a post-lane change period and was to be accomplished while maintaining a specified highway velocity. Lane deviation data showed that novice drivers had significantly more variance in lane position during the preparatory and post-lane change periods than experienced drivers. Novice drivers also spend significantly less time looking at the vehicle's speedometer and mirrors. This study suggests that virtual reality driving simulators may be useful in helping novice drivers acquire the skills necessary for safe lane change maneuvers.

Guihua Yang, Farnaz Baniahmad, Beverly K. Jaeger, and Ronald R. Mourant Virtual Environments Laboratory 334 Snell Engineering Center Northeastern University Boston, MA 02115 Although driving often involves considerable use of a vehicle’s mirrors, collecting data on the frequency and duration of mirror glances can be difficult. First, data collection in the real world may be time-consuming and unsafe. Second, low -cost simulators have used visual inserts within the projected image to simulate the inside mirror. When such inserts are used, a driver’s search and scan behavior differs significantly from that in the real world. Both eye movements and eye accommodation processes are different for mirror inserts then for actual mirrors. This paper describes a low-cost driving simulator that allows actual rear-vision mirrors to be used. Thus, eye-mirror scan behavior in our simulator emulates that of the real world. Our simulator, with the actual vehicle mirrors, is composed of two networked PCs, two projectors, and two screens. The forward view is projected to the front screen from the server, and the rear view is projected to the back screen from the client. The scene and autonomous vehicl es in the server and client are synchronized by communication between the server and client. A driver seated in the vehicle body/buck views the front driving scene on the front screen and views the rear driving scene in either the left-side mirror or inside mirror. A small video camera, which is synchronized with the computer-generated visual scene, is used to record the operator’s glances to the vehicle's mirrors. We plan to gather information and data during a study driver behavior when making lane changes under various situations. The results may be relevant to understanding the process of attaining driving skills, and could contribute to the content of driver training and education.

Parkinson's disease (PD) is the second most common neurodegenerative disorder. The debilitating nature of this medical condition can negatively impact use of the upper limb for tasks such as handwriting, which affects daily lives of many PD patients. A two-week self-administrated handwriting therapeutic exercise protocol, Amplified Air Writing (AAW), has been conducted by six PD subjects who have demonstrated micrographia symptoms, to investigate its influence on their writing samples. To study handwriting changes, several computerized metrics developed by our group to quantify micrographia were implemented. Metric results demonstrate significant temporary improvement for post-AAW handwriting performance; cursive samples especially show clear benefits for the subjects. Therefore, this self-administrated AAW exercise has potential to impart benefits on a larger PD population in their handwriting.

Findings from the Simulator Sickness Questionnaire (SSQ) evaluated several factors related to the physiological effects of immersive virtual environments (IVE) exposure. Subjects conducted locomotion activities within a selected IVE by traversing a three-dimensional computer hallway setting using either a mouse-driven static simulator or a treadmill-operated dynamic simulator system. Two levels of rendered visual detail were also compared for their SSQ effects. Simulator Sickness Questionnaire analyses reveal an effect of gender such that the females were significantly more affected by simulator activities than the male subjects. Length of time in the simulator was also found to have a significant physiological effect on the participants in the tested range of 13–23 minutes of exposure. Longer time intervals were associated with significantly greater symptoms of simulator sickness and perceived discomfort. A comparison of scores between distance judgment and movement production activities produced no significant results, leading to the conclusion that the task objectives were not different from one another in simulator effects on the participants. Individuals who used the static simulator were significantly more affected than those with similar exposure times in the dynamic simulator. Analysis of two levels of detail in the IVEs revealed a tendency for more richly textured imagery to yield higher Total Severity SSQ scores with nearly significant differences. In accordance with previously established simulator sickness profiles, the static simulator pattern of symptoms resembled those of visually-dependant IVEs (cybersickness), while the dynamic simulator symptomology was more analogous to that of motion-based IVEs (classic simulator sickness).

A divided-attention task was used to measure the ability of young and older drivers to obtain information from an in-vehicle display. Performance with the in-vehicle display was compared with performance with information superimposed on the driving scene. Ten young and 10 older drivers drove on a curvy road by using a fixed-based driving simulator. Older drivers were less accurate in obtaining information from the in-vehicle display, the average lane-position error of older drivers was greater, and older drivers spent more time driving outside their lanes. These results suggest that the use of in-vehicle displays, in their present configuration, is not appropriate for older drivers. When viewing information superimposed on the driving scene, older drivers were much more accurate and controlled their vehicles better. This indicates that the poor performance of older drivers with the in-vehicle display was due to vision-related changes, such as longer eye accommodation times, rather than cognitive processes.

Detection of changes in micrographia as a manifestation of symptomatic progression or therapeutic response in Parkinson's disease (PD) is challenging as such changes can be subtle. A computerized toolkit based on quantitative analysis of handwriting samples would be valuable as it could supplement and support clinical assessments, help monitor micrographia, and link it to PD. Such a toolkit would be especially useful if it could detect subtle yet relevant changes in handwriting morphology, thus enhancing resolution of the detection procedure. This would be made possible by developing a set of metrics sensitive enough to detect and discern micrographia with specificity. Several metrics that are sensitive to the characteristics of micrographia were developed, with minimal sensitivity to confounding handwriting artifacts. These metrics capture character size-reduction, ink utilization, and pixel density within a writing sample from left to right. They are used here to “score” handwritten signatures of 12 different individuals corresponding to healthy and symptomatic PD conditions, and sample control signatures that had been artificially reduced in size for comparison purposes. Moreover, metric analyses of samples from ten of the 12 individuals for which clinical diagnosis time is known show considerable informative variations when applied to static signature samples obtained before and after diagnosis. In particular, a measure called pixel density variation showed statistically significant differences ( $p ) between two comparison groups of remote signature recordings: earlier versus recent, based on independent and paired t -test analyses on a total of 40 signature samples. The quantitative framework developed here has the potential to be used in future controlled experiments to study micrographia and links to PD from various aspects, including monitoring and assessment of applied interventions and treatments. The inherent value in this methodology is further enhanced by its reliance solely on static signatures, not requiring dynamic sampling with specialized equipment.

In a simulated driving scenario, we investigated the production of vehicle velocity at 30 and 60 mph while varying the amount of optical flow (roadside trees versus no trees) and the geometric field of view (25, 55, and 85 horizontal degrees). Participant drivers were significantly less accurate in producing vehicle velocities of 30 mph as contrasted with those of the 60 mph target set. On average, the produced velocities were overestimated by 20 mph for the 30 mph target set. This may be due to an inability of driving simulators to produce authentic optical flows when traversing at 30 mph or lower velocities.