Système de reconstruction de trajectoire pour véhicules deux roues motorisés

The drivers of Powered Two Wheels vehicles are considered among the most vulnerable road users, as attested by the number of crashes increasing every year. The significant part of mortalities related to single vehicle “without identifying a third party” is related to the loss of control in bends. These thesis work is based on an instrumented motorcycle platform with a multi-sensor system. We have proposed algorithms to accurately reconstruct motorcycle trajectories achieved when negotiating bends. This system is intended to objectively evaluate and examine the behavior of drivers when negotiating bends in order to better train them. The goal is to lead them to adopt a safe trajectory in order to improve the road safety. Data required for the trajectory reconstruction are acquired using an instrumented motorcycle embedding several redundant sensors (reference sensors and low-cost sensors) that measure the rider’s actions (roll, steering) and the motorcycle behavior (position, velocity, acceleration, odometry, heading and attitude). This work is a part of the ARN project VIROLO++. The solution we have proposed allows to reconstruct bikes trajectories in bends with acceptable accuracy. The developed algorithm will be used to objectively evaluate and examine how riders negotiate bends. The embedded system carrying this algorithm can be used for the initial training and retraining in order to better train motorcycle drivers to estimate a safe trajectory and thus ensure safety when taking bends.