a sliding window filter, the sliding window filter adapted to receive a data set from the physical system, the data set comprising a plurality of data points, the sliding window filter selecting multiple data windows in the data set, with each of the data windows including a subset plurality of the data points in the data set, and with adjacent windows in the multiple data windows overlapping in the data set, the sliding window filter adapted to generate upper confidence bounds and lower confidence bounds for each data point using each of the multiple data windows that includes the data point, the sliding window filter adapted to select an upper confidence bounds and a lower confidence bounds for each data point that results in the smallest confidence interval between the upper confidence bounds and lower confidence bounds for that data point, and wherein the sliding window filter is adapted to generate a filtered estimate of the data set from the selected upper confidence bounds and lower confidence bounds for each data point; and

a trend change detection mechanism, the trend change detection mechanism adapted to determine a first convex hull for a set of upper confidence bounds and a second convex hull for a set of lower confidence bounds, the trend change detection mechanism adapted to compare the first convex hull and the second convex hull to determine a transition point in the data set, the trend change detection mechanism adapted to determine an estimated trend of the data set based on the transition point and the set of upper confidence bounds and the set of lower confidence bounds.