Title :
A proposal for Neuro-ITS over the connected vehicles network
Author_Institution :
Adv. Res. div, DENSO Corp., Nisshin, Japan
Abstract :
On the basis of the recent technical trend of automated vehicles and connected vehicles, we have been proposing the new application concept Neuro-ITS which has three major technical features such as multi-viewpoint tracking, human appearance prediction, and collective intelligence. Especially by combining the collective intelligence with sensing control, we will drastically reduce the hectic tasks to collect and teach huge size of GT (ground truth) which has been serious bottleneck of conventional machine learning. Also it will greatly improve the performance of environment understanding beyond perception. In this article, we focus on the collective intelligence and investigate the technical realization regarding the evolutionary process of ET (estimated truth) towards GT.
Keywords :
evolutionary computation; intelligent transportation systems; learning (artificial intelligence); neural nets; road vehicles; ET; GT; automated vehicles; collective intelligence; connected vehicle network; connected vehicles; estimated truth; evolutionary process; ground truth; human appearance prediction; machine learning; multiviewpoint tracking; neuro-ITS; sensing control; Cameras; Mobile communication; Roads; Spatiotemporal phenomena; Trajectory; Vehicles;
Conference_Titel :
Intelligent Vehicles Symposium (IV), 2015 IEEE
DOI :
10.1109/IVS.2015.7225653