Downloadable ContentDownload PDF
Development and Implementation of Advanced Driver Assistance Algorithms
This thesis shows the design and performance of Advanced Driver Assistance System (ADAS) algorithms on a S32 NXP image processing board. The system was integrated onto a 2016 Chevrolet Camaro and included a monocular camera, two front-facing LIDARs and two blind-spot LIDARs. Communication between the sensors and processing board is possible through a common physical network used by vehicles and satellites known as Control Area Network (CAN). Algorithms like vehicle and lane detection are combined with sensor and vehicle signals to produce more complex algorithms like collision warning and suggested driving velocities. Due to a non-disclosure agreement with NXP, code will not be provided but software flow diagrams describing the logic will. The system was integrated onto a 2016 Chevrolet Camaro as part of the EcoCAR3 university competition sponsored by General Motors and the U.S. Department of Energy. The system was developed over the course of one year in the competition.