Safe Manipulation of Humans in Robot-driven Physical Human-Robot Interaction

• Autores: Francisco J. Ruiz Ruiz
• Directores de la Tesis: Jesús Manuel Gómez de Gabriel (dir. tes.), Juan Manuel Gandarias Palacios (codir. tes.)
• Lectura: En la Universidad de Málaga ( España ) en 2023
• Idioma: español
• Tribunal Calificador de la Tesis: Mahdi Tavakoli (presid.), Anthony Mandow (secret.), Iván Maza Alcañiz (voc.)
• Programa de doctorado: Programa de Doctorado en Ingeniería Mecatrónica por la Universidad de Málaga
• Materias:
  • Ciencias tecnológicas
    • Tecnología de la instrumentación
      • Ingeniería de control
• Enlaces
  • Tesis en acceso abierto en: RIUMA
• Resumen

  • Due to the emergence of collaborative robots, humans and robots have been working in close proximity, and in some tasks, even sharing a common goal. Under these circumstances, physical contact between human and robot has a high probability of occurrence, which has made the field of physical Human-Robot Interaction (pHRI) a hot research topic. Despite the impact that pHRI applications could have in society, the current state-of-the-art in this field is focused on collaboration, i.e., on how human and robot work together towards the completion of a given task minimizing the physical interaction between both. However, there still a lack of works about pHRI in which robots have a proactive role. This thesis tackles the challenge of studying and implementing robot-driven pHRI tasks. To this end, such a challenge is decomposed. First, a compliant underactuated gripper specifically designed for human limb grasping is introduced. The kinetostatic model of the gripper is computed, allowing for the estimation of the grasping and interaction forces. Secondly, a method to estimate the parameters of the kinematic model of a human upper-limb is presented. Such a method relies on the kinesthetic information of a robotic manipulator that moves the human limb, requiring only a simple ascendant motion. Finally, two controllers under the assist-as-needed paradigm are developed. The first one serve as a balance assistant with a collaborative manipulator. The second one assist the user in the following of a predetermined Cartesian path. In general, although some extra considerations should be taken into account the outcomes of this work lay the foundations for the implementation of a robot-driven pHRI task.