Nathan P. Koenig

Design and use paradigms for Gazebo, an open-source multi-robot simulator

Simulators have played a critical role in robotics research as tools for quick and efficient testing of new concepts, strategies, and algorithms. To date, most simulators have been restricted to 2D worlds, and few have matured to the point where they are both highly capable and easily adaptable. Gazebo is designed to fill this niche by creating a 3D dynamic multi-robot environment capable of recreating the complex worlds that would be encountered by the next generation of mobile robots. Its open source status, fine grained control, and high fidelity place Gazebo in a unique position to become more than just a stepping stone between the drawing board and real hardware: data visualization, simulation of remote environments, and even reverse engineering of blackbox systems are all possible applications. Gazebo is developed in cooperation with the Player and Stage projects (Gerkey, B. P., et al., July 2003), (Gerkey, B. P., et al., May 2001), (Vaughan, R. T., et al., Oct. 2003), and is available from http://playerstage.sourceforge.net/gazebo/ gazebo.html.

Mobile human-robot teaming with environmental tolerance

We demonstrate that structured light-based depth sensing with standard perception algorithms can enable mobile peer-to-peer interaction between humans and robots. We posit that the use of recent emerging devices for depth-based imaging can enable robot perception of non-verbal cues in human movement in the face of lighting and minor terrain variations. Toward this end, we have developed an integrated robotic system capable of person following and responding to verbal and non-verbal commands under varying lighting conditions and uneven terrain. The feasibility of our system for peer-to-peer HRI is demonstrated through two trials in indoor and outdoor environments.

2010 Special Issue: Communication and knowledge sharing in human-robot interaction and learning from demonstration

Inexpensive personal robots will soon become available to a large portion of the population. Currently, most consumer robots are relatively simple single-purpose machines or toys. In order to be cost effective and thus widely accepted, robots will need to be able to accomplish a wide range of tasks in diverse conditions. Learning these tasks from demonstrations offers a convenient mechanism to customize and train a robot by transferring task related knowledge from a user to a robot. This avoids the time-consuming and complex process of manual programming. The way in which the user interacts with a robot during a demonstration plays a vital role in terms of how effectively and accurately the user is able to provide a demonstration. Teaching through demonstrations is a social activity, one that requires bidirectional communication between a teacher and a student. The work described in this paper studies how the users visual observation of the robot and the robots auditory cues affect the users ability to teach the robot in a social setting. Results show that auditory cues provide important knowledge about the robots internal state, while visual observation of a robot can hinder an instructor due to incorrect mental models of the robot and distractions from the robots movements.

Robot life-long task learning from human demonstrations: a Bayesian approach

Programming a robot to act intelligently is a challenging endeavor beyond the skill level of most people. Trained roboticists generally program robots for a single purpose. Enabling robots to be programmed by non-experts and to perform multiple tasks are both open challenges in robotics. This paper presents a framework that allows life-long robot task learning from demonstrations. To make that possible, the paper introduces a task representation based on influence diagrams, and a method to transfer knowledge between similar tasks. A novel approach to influence diagram learning is presented along with a demonstration method that allows non-experts to teach tasks to the robot in an intuitive manner. The results from three user studies validate that the approach enables both a simulated and a physical robot to learn complex tasks from a variety of teachers, refining those tasks during on-line performance, successfully completing the tasks in different environments, and transferring knowledge from one task to another.

Learning from Demonstration: A Study of Visual and Auditory Communication and Influence Diagrams

Learning from demonstration utilizes human expertise to program a robot. We believe this approach to robot programming will facilitate the development and deployment of general purpose personal robots that can adapt to specific user preferences. Demonstrations can potentially take place across a wide variety of environmental conditions. In this paper we study the impact that the users visual access to the robot, or lack thereof, has on on teaching performance. Based on the obtained results, we then address how a robot can provide additional information to a instructor during the LfD process, to optimize the two-way process of teaching and learning. Finally, we describe a novel Bayesian approach to generating task policies from demonstration data.