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.
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.
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
implementation of spiking tactile skin for robots
learn predictive methods
learn manipulation of switches
My interest in Robotics and Cognitive Science was sparked when I joined the RoboCup @Home Team TOBI. In 2017, we won the WorldChampionship in service robotics and for my Bachelor's Thesis I've work with one of these robots. During that time, I learned how robots drive, manipulate objects, form decision processes and how they can learn themselves. For my Master's Thesis, I explored the field of Reinforcement Learning in conjunction with Representation Learning. It was very exciting for me to acquire new skills in this field of research. My plan is to combine both of my main interest for my PhD.
What I like most about NeuTouch is the mix of many different sciences which try to achieve a common goal. The possibility to experience a different culture for three years motivated my choice as well. I hope that this time will create many interesting opportunities for exciting collaborations.
Robotics, machine learning
B.Sc. Cognitive Computer Science (2013-2017)
M.Sc. Intelligent Systems (2017-2019)