Today, Bosch has introduced a predictive road condition service that can help make sure self-driving vehicles remain safe even on wet and icy roads. The company says the technology can give automated vehicles that seat-of-the-pants feel -- you know that sensation when you're on the driver's seat that tells you the road's condition? Bosch management board member Dr. Dirk Hoheisel says the service can alert AVs to hazards "before critical situations can develop." The technology takes multiple possible weather forecast scenarios from Finnish company Foreca into consideration, so a vehicle that uses it knows how and where it can drive autonomously. Bosch says this can prevent vehicles from having to hand over controls to a human driver at the first sign of poor road conditions.

I introduce Forecastable Component Analysis (ForeCA), a novel dimension reduction technique for temporally dependent signals. Based on a new forecastability measure, ForeCA finds an optimal transformation to separate a multivariate time series into a forecastable and an orthogonal white noise space. I present a converging algorithm with a fast eigenvector solution. Applications to financial and macroeconomic time series show that ForeCA can successfully discover informative structure, which can be used for forecasting as well as classification. The R package ForeCA accompanies this work and is publicly available on CRAN.