Social Mobile Manipulator

My current Master’s thesis focuses on Mobile Manipulators, which are robots consisting of a mobile base with a robotic arm mounted on top. In this research, we address the challenge of navigating a Mobile Manipulator through environments with humans, classifying this work under Social Robotics.

In this problem, we cannot solely rely on the usual robot navigation criteria, where a robot is deemed successful if it avoids obstacles. When interacting with humans, robots must also consider human comfort to ensure a positive work environment and coexistence. Therefore, parameters such as speed, trajectory, and other characteristics need to be adjusted. A Social Robot may ensure human comfort during its navigation, for example, the Mobile Manipulator may navigate slowly when a human is nearby, and the arm may move lower to avoid sensitive areas, such as the eyes.

On left, Social Regions by Patomak et al., showing a region where people might feel comfortable or uncomfortable if a robot is there. On right, RRT* result for a simplified 2D Mobile Manipulator with a holonomic mobile base and a 2-link planar robotic arm, going from a starting point to an end point.

We are currently testing a RRT* based algorithm to consider social norms in 2D using MatLab. We plan to extend for 3D, use a non-holonomic base and test decoupled (compute arm and base movement separetely) algorithms, consider groups of people and run on a real mobile-manipulator present in our laboratory.