Title :
Detecting anomalies in humanoid joint trajectories
Author :
Marcolino, Fernando ; Jiuguang Wang
Author_Institution :
Univ. do Estado da Bahia, Salvador, Brazil
Abstract :
We present a semi-supervised anomaly detection system for humanoid robots that operates on trajectories with varying lengths, resolutions, and time shifts. The proposed approach utilities optimization to extract a model from joint trajectories under normal operation and seek to identify anomalous behaviors that deviates significantly from the known model. Compared to previously proposed approaches in humanoid anomaly detection that identified only high-level faults, our approach can detect subtle defects in the robot and at the same time, is capable of generalizing to higher-level behaviors. The system is demonstrated on a simulated model of the Atlas humanoid robot, with several experimental scenarios demonstrating detection of both joint-level anomalies and behaviors such as falling.
Keywords :
fault diagnosis; humanoid robots; learning (artificial intelligence); legged locomotion; optimisation; robot dynamics; Atlas humanoid robot; anomalous behaviors; behavior detection; high-level fault identification; humanoid joint trajectories; joint-level anomaly detection; model extraction; optimization; semisupervised anomaly detection system; subtle defect detection; Actuators; Detectors; Humanoid robots; Joints; Legged locomotion; Trajectory; Humanoid robots; anomaly detection; optimization;
Conference_Titel :
Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ International Conference on
Conference_Location :
Tokyo
DOI :
10.1109/IROS.2013.6696722