Modeling and simulation for exploring human-robot team interaction requirements

Abstract

Small-sized and micro-robots will soon be available for deployment in large-scale forces. Consequently, the ability of a human operator to coordinate and interact with largescale robotic forces is of great interest. This paper describes the ways in which modeling and simulation have been used to explore new possibilities for human-robot interaction. The paper also discusses how these explorations have fed implementation of a unified set of command and control concepts for robotic force deployment. Modeling and simulation can play a major role in fielding robot teams in actual missions. While live testing is preferred, limitations in terms of technology, cost, and time often prohibit extensive experimentation with physical multi-robot systems. Simulation provides insight, focuses efforts, eliminates large areas of the possible solution space, and increases the quality of actual testing.

"Where am i?" acquiring situation awareness using a remote robot platform

“…1 An average of 30% of each run was spent acquiring SA while no other task was being done. 2 Despite this time spent trying to acquire SA, users often expressed confusion about where their robots were located relative to various landmarks and whether their robots were near obstacles. For example, users backed robots into walls, asked "have I been here before?…”

“…1 Data from Subject 3's Run 2 was not analyzed due to problems with the robot system during the run. 2 During Subject 4's Run 3, the autonomous mode was primarily used. Since the robot did most of the driving, the subject did not pan the camera or robot to acquire SA.…”

Human swarm interaction for radiation source search and localization

“…The tasking of the robots by the human user introduced later deviates away from the notion of swarm robotics. In [9], Dudenhoeffer et al conduct modeling and simulation for exploring HRI requirements and suggest that with high level of automation where the operator serves mainly a monitoring role; situation awareness may be negatively impacted. Their study suggests that emphasis on monitoring alone ignoring collaboration roles in multi-robot interaction poses a significant problem to the overall swarm due to degradation of situational awareness.…”

Multimodal controls for soldier/swarm interaction

“…must not only permit exchange of detailed or specific information on an individual robot, but permit the user to gain an understanding of intent and operations of a multi-part entity that has a cognition of its own, and also has potentially hundreds or thousands of members at both micro (individual member) and macro (swarm) levels [2]. Soldier-swarm interaction is a critical aspect of swarm control, especially in disrupted or degraded conditions: the Soldier must be kept cognizant of swarm operations through an interface that allows him or her to monitor status and/or institute corrective actions.…”