Pervasive and Mobile Computing 8 (2012) 542–561 Contents lists available at SciVerse ScienceDirect Pervasive and Mobile Computing journal homepage: www.elsevier.com/locate/pmc Human-aware planning for robots embedded in ambient ecologies Marcello Cirillo ∗ , Lars Karlsson, Alessandro Saffiotti AASS Research Center, Örebro University, Sweden article info Article history: Available online 7 December 2011 Keywords: Human-aware planning Planning under uncertainty Human–robot interaction Planning in intelligent environments abstract We address the issue of human–robot cohabitation in smart environments. In particular, the presence of humans in a robot’s work space has a profound influence on how the latter should plan its actions. We propose the use of human-aware planning, an approach in which the robot exploits the capabilities of a sensor-rich environment to obtain information about the (current and future) activities of the people in the environment, and plans its tasks accordingly. Here, we formally describe the planning problem behind our approach, we analyze its complexity and we detail the algorithm of our planner. We then show two application scenarios that could benefit from the techniques described. The first scenario illustrates the applicability of human-aware planning in a domestic setting, while the second one illustrates its use for a robotic helper in a hospital. Finally, we present a five hour-long test run in a smart home equipped with real sensors, where a cleaning robot has been deployed and where a human subject is acting. This test run in a real setting is meant to demonstrate the feasibility of our approach to human–robot interaction. © 2011 Elsevier B.V. All rights reserved. 1. Introduction A very promising convergence is taking place between the fields of ambient intelligence and autonomous robotics. Researchers in both areas are increasingly interested in the possibilities offered by the inclusion of robotic devices inside smart environments. On the one hand, this inclusion could empower a smart environment with unprecedented sensing and actuation capabilities. On the other hand, it could make a robot more robust and autonomous by exploiting the capabilities offered by the other devices in the environment. For instance, a robotic vacuum cleaner could rely on the home localization system to know its position in the home, and hence to optimize its cleaning strategy. The convergence of ambient intelligence and autonomous robotics has given birth to several new research areas, including network robot systems [1], ubiquitous robotics [2], and robot ecologies [3–5]. In all these areas, the main goal is to design intelligent robotic environments, that is, environments where close communication is established among sensing and robotic devices. The coordinated cooperation of such devices can enact and support complex tasks to help the users in their everyday life. The whole environment is often viewed as an ecosystem of components, and human users are considered as the core of this ecosystem. Intelligent robotic environments are expected to open a wide space of new applications, ranging from domestic robots aimed at improving the quality of life of the elderly [6], to service robots operating in a factory outside the confined space of a working cell. Most of these new applications have one aspect in common: robots would operate in the same space as humans, and interact with them as members of the same ecosystem. The cohabitation of humans and robots in the same space induces many difficult technical challenges. Safety and human–robot interaction immediately come to mind, and intensive work has been ongoing on both problems for a few ∗ Corresponding author. E-mail addresses: marcello.cirillo@aass.oru.se (M. Cirillo), lars.karlsson@aass.oru.se (L. Karlsson), alessandro.saffiotti@aass.oru.se (A. Saffiotti). 1574-1192/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.pmcj.2011.11.004