To ensure that artificial intelligent robotic systems absorb the proper data from their sensors and combination of data from each sensor to form observational data sets, we must dictate what is worthy of memory storage in what data must be dumped from the data sets. Since we expect artificial intelligence systems to learn, modify their behavior and adapt we need to address the absorption of observational data and held observational data will be used in the future. That is to say future events in which the artificially intelligent robotic systems will be involved with and how it will respond.

The absorption of observational data is key point indeed. But so is the dumping of erroneous data, data which was used in error or decisions made From said data which led to incorrect answers or something other than the best possible choice in that particular event. One way to absorb data from the sensors is to continually created XML spreadsheets of data during the activity of the artificially intelligent robotic system. Then held the robotic system systematically check the datasets for those particular events against previous datasets which were either programmed into the unit or which the units created to best modify any use only the best datasets or XML spreadsheets.

Then the artificially intelligent robotic system will have absorb the new data or modify the old spreadsheet or master spreadsheet and then dump the old data. Speaking primarily to the issue of mobility and motion sensors the artificially intelligent robotic system may adjust its stability control systems based on such things as wind, surface traction, angle of incidence, angle of tilt, speed of unit or weight of object being carried. All these factors may indeed prevent a disruptive of event in the robotic movement if additional datasets are indeed connected to the master datasets through variation triggers of data.

Although this particular topic can get extremely complicated very quickly, I thought it might be of interest to you to speak directly to the concept of when and why new data from artificially intelligent robotic systems must be absorbed with the sensors present situations not yet encountered. It is important for the robotic artificially intelligent system to learn just as a child must learn to trial and error when it first walks. Consider all this in 2006.