Wheel-rail contact analysis system using spectral signatures for train automation and traffic management

The transport industry is subject to a lot of improvement aiming to autonomous transportation systems. Railway is also concerned. An autonomous vehicle must be able to perceive and analyze its environment with a view to being able to adapt its driving. In this paper we focus on the wheel-rail contact analysis, and the adherence evaluation, as it conditions loads of variables such as braking, traction and maximum speed. It also has an impact on security distance between trains. Adherence can be degraded by the presence of pollution on the rail surface. We present an imaging system, using a multispectral camera, capable of detecting and recognizing pollutant. The pollution is then associated to an adherence coefficient.