Socially-aware navigation and object transportation for mobile manipulators in static environments

Abstract

This work addresses the challenge of navigating a robotic mobile manipulator, carrying a load, in a static, human-populated environment while respecting social norms. In other terms, it tackles the problem of Social Navigation for Mobile Manipulators, aiming to develop a socially-aware mobile manipulator. Here, we propose a novel technique that not only ensures collision-free navigation but also respects human social preferences and considers the object being transported, building harmonious coexistence between robots and people. While previous research in social robotics focused on mobile robots for social navigation, this work seeks to integrate a robotic arm to a mobile base and actuate both mobile platform and robotic arm concurrently. This setup allows the system to minimize social discomfort while transporting objects. We propose two distinct techniques: the first computes coordinated base–arm motions to minimize discomfort, suitable for simpler two-dimensional environments. The second is a decoupled strategy, planning first the base and then the arm, making it faster to find solutions for three-dimensional scenarios. In addition, we introduce a novel three-dimensional discomfort model, which, to the best of our knowledge, has not been previously defined. Our results demonstrate that the proposed methods outperform traditional approaches that consider only the mobile base.