Being Transparent about Transparency: A Model for Human-Robot Interaction

The current paper discusses the concept of human-robot interaction through the lens of a model depicting the key elements of robot-to-human and robot-of-human transparency. Robot-to-human factors represent information that the system (which includes the robot but is broader than just the robot) needs to present to users before, during, or after interactions. Robot-of-human variables are factors relating the human (or the interactions with the human; i.e., teamwork) that the system needs to communicate an awareness of to the users. The paper closes with some potentials design implications for the various transparency domains to include: training and the human-robot interface (including social design, feedback, and display design). Human-Robot Interaction The world is on the cusp of a robotics revolution given advances in technology and software relating to robotic systems. Yet, despite cutting-edge technology, fundamental research is still needed to better understand the dynamics of human-robot interaction. Robotic systems in the future will likely possess greater autonomy than current systems. Further, robotic systems within the military may be operated in hostile, complex situations and may be given the authority to execute lethal decisions within the battlespace (Arkin, 2009). The interactions between humans and robots will also likely get more complex as systems increase in their autonomy. For instance, instead of the teleoperation of contemporary military robotic systems such as Uninhabited Aerial Vehicles (UAVs), future operators will likely execute Copyright © 2013, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. supervisory control of multiple robots. This evolution of robotic capabilities coupled with increased supervisory control from humans adds additional layers of complexity in the human-robot interaction, thus making the humans’ trust of robotic systems a key aspect of the overall humanrobot system. Understanding the trust a human has of a robotic system is important because trust will influence whether or not the human relies on the technology (Lyons & Stokes, 2012), and history shows that our reliance behavior is often suboptimal. There are numerous examples of major accidents in the aviation community being attributed to mis-calibrated reliance on technology. Thus, the research community is called to improve the trust calibration process when humans interact with advanced technological systems. Yet, there is a surprisingly little systematic guidance in these regards. Having shared context between humans and robots will be a critical facet of the overall system performance of human-robot teams (Stubbs, Wettergreen, & Hinds, 2007) and this will likely facilitate “appropriate” reliance on the robotic system (Lee & See, 2004). The current paper suggests that transparency between the robot and the human is one mechanism to facilitate effective interactions between humans and their robotic teammates. The terms robot and autonomous system are used interchangeably in the current paper to represent a system that must operate autonomously in a dynamic environment under some conditions of uncertainty. Robot-to-Human Transparency With added capability comes added responsibility, at least in theory. This raises questions concerning accountability with systems that operate under some degree of autonomy. A Model for Human-Robot Interaction Trust and Autonomous Systems: Papers from the 2013 AAAI Spring Symposium