Kollisionsvarningssystem för spårvagn - Varningssystem baserat på radardetektioner

Abstract

Tram traffic today has accidents that each year cost money and causes people to seek medical attention. The purpose of this project is to conduct which accidents can be prevented with a forward collision warning system, as well as create a warning system prototype. The work will be carried out at the company Aptiv. A review of the accident statistics and an analysis of the accident types and overall tram traffic has been completed. Tram traffic differs from other traffic since it moves at a slower pace and there is shorter distance to objects around the tram. Most accidents happen due to human error. The warning system has been developed with the help of tram logs. There are two radar sensors who detect moving objects and stationary obstacles. Their data is analysed in a program that calculates time to collision. The system will warn the tram driver by ligthning different colored lamps depending on how time critical the situation is. How well the system works is yet to be investigated. The warning functions have been tested on the tram logs and seem to warn at the right time. The warnings have also been tested on a truck instead of a tram, and here they also appear to warn correctly, but this needs further investigations. A perk with this system is that collisions caused by inattentive drivers can be prevented. A suggestion for further development is to implement adaptive cruise control, so that the tram automatically can regulate its speed regarding to other vehicles or obstacles in front of the tram, as well as conduct a study on how the tram driver interact with the warnings.