Road surface condition detection based on road surface temperature and solar radiation

摘要

This paper presents a method of road surface condition detection with road temperature and solar radiation by BP neural network. Because road temperature depends on road surface condition (dry, wet, icy) and solar radiation (mapped to season, geographical location, time, air temperature and air humidity), and there is nonlinear causality between them, road surface condition can be detected indirectly with road temperature and solar radiation. In experiment, BP neural network was trained with 2208 group data and validated by 192 group data, the detection accuracy reached 90%. It is feasible to detect road surface condition with road temperature and solar radiation.