Adaptive combination of support robot's degree-of-freedom for an easier control

(Adaptive Freiheitsgradeinbettung als kooperatives Userinterface für einen Assistenzroboter)


An assistive robotic arm is a valuable support for people with severely limited upper limb mobility. However, its use is often strenuous and time-consuming because the seven-movement options (degrees of freedom) of a robotic arm are matched by only two dimensions of a typical input device (e.g., joystick). Therefore, it is frequently necessary to manually switch between movement options.

Goals and procedure

In this project, the use of assistive robotic arms will be facilitated by the application of deep neural networks. A sensor-based situation recognition will be combined with an algorithm-based control to form an adaptive AI-based control system. Unlike with automatic control, the user retains control. The project focuses on three main aspects:

  1. A neural network developing suggestions for movement options based on training data generated in virtual reality with VR controllers.
  2. Exploring data glasses as a possibility for displaying feedback in a user-centered design process for human-robot communication.
  3. Elicitation of requirements and evaluation of the system by stakeholders using a participatory approach.

Innovations and perspectives

The movement options suggested by the system help users to simplify the control of robotic assistance systems. This can significantly increase the autonomy of those affected by upper limb mobility restrictions in their daily lives.