Skip to main content

Trinity College Dublin, The University of Dublin

Trinity Menu Trinity Search



You are here Research > RAIL - Robotics & Innovation Lab

Human Robot Interaction Control Interfaces

As part of our research into assistive robotics, we have been interested in investigating effective control interfaces to robotic systems which will eventually be used by untrained individuals in unstructured environments.

It is our belief that fully autonomous systems will not be viable in the near-future, and that assistive robotics will be able to provide a greater level of care and service if this fact is acknowledged through incorporation of the human into the robot’s control by design.

As part of this investigation we have developed a novel control interface method to allow for more intuitive semi-autonomous control, for which we have a patent pending.

In order to evaluate the effectiveness of various control interfaces, we launched a research project to investigate how untrained individuals from a range of demographics would interact with a robotic system in a home-environment. In order to accomplish this in a short time frame, with limited risk of injury or damaging equipment, we developed a simulation of Robbie and a home environment in which the robot would have to accomplish a set of basic common tasks through human guidance, using the open source software Blender.


Robot moving around the simulated home environment


Simulator instruction screen shown to all participants


The control embodiments used during the simulation

This testing evaluated a number of human-robot control interfaces and morphologies, with both a physical dedicated controller investigated (in the form of an xbox controller) and an application running on an Android tablet which we expect will prove to be the more conventional control interface to commercialized assistive robots.