This paper analyzes the effects of daylight and daylight saving time (DST) on pedestrian and motor vehicle occupant fatalities in the United States. Multivariate analyses of county level data from the Fatality Analysis Reporting System for 2-week periods in 1998 and 1999 are used. Results show that full year daylight saving time would reduce pedestrian fatalities by 171 per year, or by 13% of a...

While most building simulation tools model occupancy using simple 24-hour profiles, researchers are applying machine learning and other advanced modeling approaches to simulate individual occupants and their interactions with buildings. For building designers to fully benefit from these increasingly advanced occupant models, visualizations must ultimately reveal subtle yet informative patterns ...

This paper proposes a methodology that allows the combination of the Stability Factor and handling characteristic to predict the probability of rollover per single vehicle crash. Paper No.202-O Simulation of Occupant Kinematics in Vehicle Rollover Dummy Model versus Human Model Norbert Praxl, Markus Schönpflug, Jiri Adamec Institute for Legal Medicine Munich, Germany Abstract Currently, 25% of ...

Road traffic accidents are responsible for over 3000 deaths per year in the UK, according to Department for Transport (2004a) figures. Although progress is being made in a number of areas, vehicle occupant fatalities have not been falling in line with casualty reduction targets for the year 2010. A sample of 1185 fatal vehicle occupant cases was considered, from ten UK police forces, from the y...

In this paper, we propose a new HVAC (heating, ventilation, and air conditioning) control strategy as part of the smart energy system that can balance occupant comfort against building energy consumption using ubiquitous sensing and machine learning technology. We have developed ZigBee-based wireless sensor nodes and collected realistic temperature and humidity data during one month from a labo...

A new method of predicting automobile occupant posture is presented. The Cascade Prediction Model approach combines multiple independent predictions of key postural degrees of freedom with inverse kinematics guided by data-based heuristics. The new model, based on posture data collected in laboratory mockups and validated using data from actual vehicles, produces accurate posture predictions fo...

AbstmctA red-time stereo-based head detection algorithm i s presented and evaluated as an occupant posture analysis system for ”smart” airbag deployment. The algorithm uses several constraints to limit the number of head ellipse ”candidates” found in the image. These constraints are based on shape, size and depth of the occupant’s head. Results of ground truth experiments show $at the bead dete...

Knowledge of human presence and interaction in a vehicle is of growing interest to vehicle manufacturers for design and safety purposes. We present a framework to perform the tasks of occupant detection and occupant classification for automatic child locks and airbag suppression. It operates for all passenger seats, using a single overhead camera. A transfer learning technique is introduced to ...

One hundred ninety-five rear impacts with both front- and rear-seat occupants in the struck car, where at least one occupant sustained permanent disability, were selected for study. There was a significantly higher disability risk for the female rear-seat occupant compared with the male driver. Furthermore, a higher risk was found for female rear-seat occupants compared with female front-seat p...