Failure Prediction Using Robot Execution Data

Robust execution of robotic tasks is a difficult learning problem. Whereas correctly functioning sensors’ statements are consistent, partially corrupted or otherwise incomplete measurements will lead to inconsistencies within the robot’s learning model of the environment. So, methods of prediction (classification) of robot failure detection with erroneous or incomplete data deserve more attention.Studies have shown that the techniques combining to classification has become an effective tool in increasing the efficiency and accuracy the classifieds. The primary goal of the evaluation was to analyze the impact of erroneous data on predictive robot fault detection accuracy .And then show category, increases performance classification despite noise with combined categories. In this regard have used the data set associated with torque obtained from a humanoid robot. In this paper the performance of base-level classifiers and meta-level classifiers is compared. Bagged Naïve Bayes is found to perform consistently well across different settings

کلیدواژه ها:

fault detection ، incomplete data ، and Meta classifiers