A Hybrid Intelligent System to Detect Anomalies in Robot Performance

Nuño Basurto ^[1] ; Ángel Arroyo ^[1] ; Carlos Cambra ^[1] ; Álvaro Herrero ^[1]
1. [1] Universidad de Burgos
  
  Universidad de Burgos
  
  Burgos, España
Localización: Hybrid Artificial Intelligent Systems: 16th International Conference, HAIS 2021. Bilbao, Spain. September 22–24, 2021. Proceedings / coord. por Hugo Sanjurjo González, Iker Pastor López , Pablo García Bringas , Héctor Quintián Pardo , Emilio Santiago Corchado Rodríguez , 2021, ISBN 978-3-030-86271-8, págs. 415-426
Idioma: inglés
Texto completo no disponible (Saber más ...)
Resumen
- Although self-diagnosis is required for autonomous robots, little effort has been devoted to detect software anomalies in such systems. The present work contributes to this field by applying a Hybrid Artificial Intelligence System (HAIS) to successfully detect these anomalies. The proposed HAIS mainly consists of imputation techniques (to deal with the MV), data balancing methods (in order to overcome the unbalancing of available data), and a classifier (to detect the anomalies). Imputation and balancing techniques are subsequently applied in for improving the classification performance of a well-know classifier: the Support Vector Machine. The proposed framework is validated with an open and recent dataset containing data collected from a robot interacting in a real environment.