Objectives

Implementation of spiking tactile system on a 7-DoF manipulator and development of manipulation algorithms for stable grasping of objects of different shape and softness. Tactile data will be integrated in the control loop of the manipulator to obtain optimal grasps taking the object shape, softness and deformability into account. Different types of end-effectors will be considered, i.e. from parallel grippers to under-actuated multi-fingered hands.

Expected Results

a new set of force controllers, i.e. impedance and admittance controllers, will be obtained allowing different robots, from mobile manipulators to humanoid robots, performing grasping of a large variety of objects. The performance of the new grasping algorithms will be compared to the performance obtained using grasping algorithms based on open loop approaches and those relying on data provided by a wrist-mounted force/torque sensor. The goal is to show that the new algorithms based on the tactile system provide a much higher ratio of successful grasps considering a set of objects and different configurations. The demonstrator will feature a service robot being able to grasp a variety of objects involved in typical home daily tasks of elderly care scenarios.

Planned Secondments

  • IIT-iCub:
    implementation of spiking tactile skin for robots
  • USFD:
    to learn predictive methods
  • UNIBI:
    to learn manipulation of switches

Enrolments (in Doctoral degree/s)

Bielefeld University

Supervisors

S. Terreri, C. Bartolozzi, L. Natale, T. Prescott, R. Haschke

Tags

ART ROB
ESR12: Grasping and Manipulation
This project is funded by the EU Horizon 2020 research and innovation programme under grant agreement No 813713 (NeuTouch)
© 2018 Neutouch