Model-free robot anomaly detection

Safety is one of the key issues in the use of robots, especially when human-robot interaction is targeted. Although unforeseen environment situations, such as collisions or unexpected user interaction, can be handled with specially tailored control algorithms, hard- or software failures typically lead to situations where too large torques are controlled, which cause an emergency state: hitting an end stop, exceeding a torque, and so on-which often halts the robot when it is too late. No sufficiently fast and reliable methods exist which can early detect faults in the abundance of sensor and controller data. This is especially difficult since, in most cases, no anomaly data are available. In this paper we introduce a new robot anomaly detection system (RADS) which can cope with abundant data in which no or very little anomaly information is present.

Authors

• Rachel Hornung
• Holger Urbanek
• Julian Klodmann
• Christian Osendorfer
• Patrick van der Smagt

Keywords

• Detectors
• Training data
• Training
• Kernel
• Support vector machines
• Robot sensing systems
• Anomaly Detection
• Collision
• Sensor Data
• Human-robot Interaction
• Use Of Robots
• False Positive
• Validation Data
• Support Vector Machine
• False Positive Rate
• Dimensionality Reduction
• Second Derivative
• Goodness-of-fit Test
• Data Space
• Number Of Centers
• Mahalanobis Distance
• Distance Metrics
• End-effector
• Back Projection
• Cartesian Position
• Fault Identification
• Linear Projection
• High-dimensional Data Space
• Novelty Detection
• Fewer Dimensions
• Dimensional Data
• False Negative
• Training Speed