Agricultural soils are variable in space and time, as well as in depth. This variability necissitates variable managment decisions with the ultimate goal of maximising crop growth and yield, and soil physical, chemical and biological charachteristics while minimising the environmental foot print from crop production. This can be achieved by implementing an integrated framework of sensing, modelling and control to optimise the farm input resources (e.g., fertilisers, pesticides, seeds, water used for irrigation, manure and tillage), responding the spatio-temopral variabilities exists in top majority of agricultural fields.  At the Precision Scoring Group in the Department of Environment we are working on development and optimisation of advanced precision agriculture solutions in creal crops and potato. To achive this target new soil and crop sensing technologies combined with machine learning algorithms are being developed and optimised, so as to optimise the soil managment decisions.