Automotive, Carmenta TrafficWatch

Dynamic most probable path algorithm

April 4, 2022

MASTER THESIS at Carmenta Automotive.

This Mater Thesis project researched how to accurately predict the routes of emergency vehicles and send out preemptive warnings with precision, we need to know a lot about our surroundings. The foundation of the algorithm is a traditional distance based A* graph searching algorithm, however we dynamically take current road conditions and events into account, to produce a prediction that depends on weather, road works, traffic, etc. Another aspect that needs consideration is the minimization of false positives, in others words, warnings when no emergency vehicle is approaching. To achieve this, multiple possible routes are calculated and compared with each other, and only if one single route is deemed significantly more likely than the alternatives, warnings are sent along this route. The algorithm is currently being validated and optimized for implementation in the Nordic Way 3 project.

The Maser thesis results was presented in 2022.