Application and use of quality metrics for the prediction of grasp success and evaluation of artificial hands

Artificial manipulation has been one of the main areas of interest in robotics for decades. Finding a proper grasp to seize objects and design robotic hands capable of such grasps are two of the main problems in such field. This thesis analyze the use of quality metrics to evaluate grasp hypothesis....

Descripción completa

Detalles Bibliográficos
Autor: Rubert Escuder, Carlos
Tipo de recurso: tesis doctoral
Estado:Versión publicada
Fecha de publicación:2018
País:España
Institución:CBUC, CESCA
Repositorio:TDR. Tesis Doctorales en Red
OAI Identifier:oai:www.tdx.cat:10803/663158
Acceso en línea:http://hdl.handle.net/10803/663158
http://dx.doi.org/10.6035/14101.2018.176497
Access Level:acceso abierto
Palabra clave:Grasping
Quality Metrics
Grasp Prediction
Artificial hands
Grasp Evaluation
Machine Learning
Enginyeria, indústria i construcció
68
Descripción
Sumario:Artificial manipulation has been one of the main areas of interest in robotics for decades. Finding a proper grasp to seize objects and design robotic hands capable of such grasps are two of the main problems in such field. This thesis analyze the use of quality metrics to evaluate grasp hypothesis. Different classification algorithms are used with such metrics to predict the success of grasp hypothesis. These algorithms are applied to evaluate artifical hands and their performance to grasp objects.