What is anomaly detection and how can it be utilized in robots ?
In simple words, anomaly detection is a process of identifying a value which is quite different from the rest of the values in that slot. For example if you plot a graph showing grades of 5 students i.e. 50, 45, 47, 25, 49 then out of these values 25 is the anomaly. This concept is used in machine learning to detect the most uncertain event. Even in the field of robotics, using this process a robot will be able to make decisions more accurately depending upon the situations and use its previous failure as a reference to reach the right answer. Machine learning is a must which is used in highly intelligent robots these days which means they are also equipped with anomaly detection system and maybe even with z score analysis which is a remedy of anomaly detection.