Side Collision Avoidance Systems: Better Agreement Between Effectiveness Predictions and Real-world Data

1999-01-0493

03/01/1999

Event
International Congress & Exposition
Authors Abstract
Content
Considerable effort has been invested in the development of models to predict the effectiveness of side collision avoidance systems (“SCAS”). These estimates, based on reliability theory, indicate that SCAS can produce a measurable improvement in safety, but that safety improvement is sensitive to the method of sensor use.
The support of real-world data for these models is inconclusive. Objectively measured “right-clear” data show varying improvement with the use of SCAS, yet professional drivers of large vehicles (buses, heavy trucks) report favourable responses to the idea of SCAS use. Unscientific surveys of the general driving public support this favorable reaction, and also indicate a significant perceived cost to near miss incidents.
This paper proposes a mathematical model of SCAS reliability that takes the above factors into consideration. This model is heavily based on earlier work in the field.
The model predicts decreases in accident rates from 20% to 95%, depending on style of usage of the SCAS. Reduction in occurrence of near miss incidents is predicted to be between 5% and 80%, depending on usage style.
Usage style (i.e. use of the sensor as a substitute for or as an addition to direct observation) is determined to be the critical variable affecting overall performance.
Sensitivity analysis indicates that the assumed proportion of lane changes with potential conflict is somwhat important. The model is not generally sensitive to other factors
Meta TagsDetails
DOI
https://doi.org/10.4271/1999-01-0493
Pages
11
Citation
Hackney, R., "Side Collision Avoidance Systems: Better Agreement Between Effectiveness Predictions and Real-world Data," SAE Technical Paper 1999-01-0493, 1999, https://doi.org/10.4271/1999-01-0493.
Additional Details
Publisher
Published
Mar 1, 1999
Product Code
1999-01-0493
Content Type
Technical Paper
Language
English