Details

Autor(en) / Beteiligte

• Tieck, J. Camilo Vasquez
• Weber, Sandro
• Stewart, Terrence C.
• Kaiser, Jacques
• Roennau, Arne
• Dillmann, Rüdiger

Titel

A spiking network classifies human sEMG signals and triggers finger reflexes on a robotic hand

Ist Teil von

• Robotics and autonomous systems, 2020-09, Vol.131, p.103566, Article 103566

Ort / Verlag

Elsevier B.V

Erscheinungsjahr

2020

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• The interaction between robots and humans is of great relevance for the field of neurorobotics as it can provide insights on how humans perform motor control and sensor processing and on how it can be applied to robotics. We propose a spiking neural network (SNN) to trigger finger motion reflexes on a robotic hand based on human surface Electromyography (sEMG) data. The first part of the network takes sEMG signals to measure muscle activity, then classify the data to detect which finger is being flexed in the human hand. The second part triggers single finger reflexes on the robot using the classification output. The finger reflexes are modeled with motion primitives activated with an oscillator and mapped to the robot kinematic. We evaluated the SNN by having users wear a non-invasive sEMG sensor, record a training dataset, and then flex different fingers, one at a time. The muscle activity was recorded using a Myo sensor with eight different channels. The sEMG signals were successfully encoded into spikes as input for the SNN. The classification could detect the active finger and trigger the motion generation of finger reflexes. The SNN was able to control a real Schunk SVH 5-finger robotic hand online. Being able to map myo-electric activity to functions of motor control for a task, can provide an interesting interface for robotic applications, and a platform to study brain functioning. SNN provide a challenging but interesting framework to interact with human data. In future work the approach will be extended to control also a robot arm at the same time. •A method to control a humanoid robot hand with muscle signals recorded with a non-invasive sEMG sensor from a human.•Motion representation using motor primitives and activated as reflexes.•The classification of the sEMG signals and the generation of the motion performed by a SNN implemented with the Nengo simulator to control a Schunk SVH 5-finger Hand using ROS.•With this representation of the robot kinematics we have a flexible and extensible approach that can be adapted to different robots.•The approach works with no modifications with a left or a right robot hand and/or human arm.