Body-Machine Interfaces (BMI)
We investigate various methods to control the DLR Light-Weight-Robot III with five-finger-hand via brain interfaces. The LWR's various soft-robotics control schemes are embedded in a human-friendly state-based control architecture which enables the development of complex interaction scenarios. The LWR-III is equipped with the modular DLR five-finger hand for grasping and holding objects. Similar to the robot, this hand is equipped with joint torque sensors for impedance control which makes it robust against uncertainties in the environment such as the position of objects to be grasped.
We use surface EMG for controlling the grasp force of the hand as well as the position of the arm in a user-intuitive manner. To this end, an adaptive system learns the correspondence between the human hand position and orientation and the muscular activity measured at the skin surface. Thereby we can control to move the arm and grasp an object in teleoperation / telemanipulation. Such control schemes are also applicable in rehabilitation and orthoses environments.
We are furthermore investigating the control of the DLR-LWR-III with five-finger-hand through human cortical implants. In this control scheme, neural signals recorded in the human motor cortex are decoded in continuous motion commands, that are executed by the robot. Additionally a binary state signal is used to trigger grasping actions with the robotic hand. This combination of state-of-the-art robotics and advanced neuro-prosthesis can potentially allow a person with severe physical disabilities to interact with their environment again. Recent results on this research has been published in Nature in 2012. Further information can be found here.
Other interfaces, including invasive communication with the human peripheral nervous system as well as surface EEG control are within the realm of our research spectrum.
A short video showing the German Chancellor Angela Merkel testing the robot that has also been used in our brain-controlled robotics experiments can be found below.