• Jānis Pekša Riga Technical University, Faculty of Computer Science and Information Technology, Institute of Information Technology (LV)



Kalman filter, noise covariances, time-series, forecasting


The article considers the road monitoring weather-stations which collects raw observations that are processed to be able to make the necessary forecasting for future decisions. For the road maintainers those predictions are crucial to make decisions daily. When it comes to the winter season when road safety is very important; however, the road condition is also affected by the snow and icing. In order to improve safety on the road network the road maintainers are trying to use every possible way to be able to provide it. A number of methods have been studied and compared to clarify the parameter required by Kalman filter, which can be improved by making forecasting more accurate. Several road monitoring weather-stations are merged into one region because they are relatively close to each other and it is assumed that there are common conditions in one region that may indicate changes in road conditions. The corresponding algorithms are applied for each region and then compared to each other. Adaptive Kalman filter is generalized in the relevant article in order to have a general understanding of how to correctly apply the approach. The main result of this article is a comparison with the different methods, which are finally compiled in a single table.


Download data is not yet available.


J. B. Edwards, “Speed adjustment of motorway commuter traffic to inclement weather,” Transp. Res. Part F Traffic Psychol. Behav., vol. 2, no. 1, pp. 1–14, Mar. 1999,

Y.-Y. Nguwi and A. Z. Kouzani, “Detection and classification of road signs in natural environments,” Neural Comput. Appl., vol. 17, no. 3, pp. 265–289, Jun. 2008,

J. P. Gibbs and W. G. Shriver, “Can road mortality limit populations of pool-breeding amphibians?,” Wetl. Ecol. Manag., vol. 13, no. 3, pp. 281–289, Jun. 2005.

C. Taylor, S. Pollard, S. Rocks, and A. Angus, “Selecting Policy Instruments for Better Environmental Regulation: a Critique and Future Research Agenda,” Environ. Policy Gov., vol. 22, no. 4, pp. 268–292, Jul. 2012.

J. Honaker and G. King, “What to Do about Missing Values in Time-Series Cross-Section Data,” Am. J. Pol. Sci., vol. 54, no. 2, pp. 561–581, Apr. 2010.

G. Bishop, G. W.-P. of SIGGRAPH, undefined Course, and undefined 2001, “An introduction to the kalman filter,”

F. R. Johnston, J. E. Boyland, M. Meadows, and E. Shale, “Some properties of a simple moving average when applied to forecasting a time series,” J. Oper. Res. Soc., vol. 50, no. 12, pp. 1267–1271, Dec. 1999.

E. Jacquier, A. Kane, and A. J. Marcus, “Geometric or Arithmetic Mean: A Reconsideration,” Financ. Anal. J., vol. 59, no. 6, pp. 46–53, Nov. 2003.

J. Huber, K. T.-M. Letters, and undefined 2001, “On the similarity of classical and Bayesian estimates of individual mean partworths,” Springer.

A. Likas, N. Vlassis, J. V.-P. recognition, and undefined 2003, “The global k-means clustering algorithm,” Elsevier.

T. M.-I. S. processing magazine and undefined 1996, “The expectation-maximization algorithm,”




How to Cite

J. Pekša, “ADAPTIVE KALMAN FILTER FORECASTING FOR ROAD MAINTAINERS”, ETR, vol. 2, pp. 109–113, Jun. 2019, doi: 10.17770/etr2019vol2.4134.