Human-Robot Trust in Emergency Evacuations - Trust is required for humans to interact with other agents in high-risk, time-critical situations even if they have never interacted with the agents before. Robots must understand why a human makes a trust decision in order to effectively aid the human in these situations. Currently, robots are prone to commit errors that negatively impact their ability to be trusted. Even when the robot is performing correctly, nearby humans may interpret its actions as strange, and thus less trustworthy.
We use emergency evacuations as our example of high-risk, time-critical situations. We began by examining actual evacuations and determining how guidance robots can best help. We then developed and evaluated several methods for a robot to communicate directional information to evacuees. We have also studied the effect of situational risk and the robot's previous performance on a participant's decision to use the robot in a future interaction.
Overtrust of Robots in Emergency Evacuation Scenarios at Human Robot Interaction (HRI) 2016:Paper, Video (YT: Overtrust Emergency Guidance Robot Demo Video), Press Release
Timing Is Key for Robot Trust Repair at International Conference on Social Robotics 2015: Paper
Assessment of Robot Guidance Modalities Conveying Instructions to Humans in Emergency Situations at Ro-Man 2014: Paper
Modeling Human-Robot Trust in Emergencies at AAAI Spring Symposium 2014: Paper
Building and Maintaining Trust Between Humans and Guidance Robots in an Emergency at AAAI Spring Symposium 2013: Paper
Trust in Emergency Evacuation Robots at SSRR 2012: Paper
Information Propagation Applied to Robot-Assisted Evacuation at ICRA 2012: Paper
Incorporating a Model of Human Panic Behavior for Robotic-Based Emergency Evacuation at Ro-Man 2011: Paper
Emergency Evacuation Robot Design at EPRRSD 2011: Paper
Adaptive User Interface for Vehicle Swarm Control - For my Masters Thesis I created a robotic swarm control interface and tools to assist a user in controlling a simulated swarm. The tools used the Gram-ART symbolic clustering algorithm to learn which commands were frequently issued to the vehicles. It them combined those commands into behaviors and gave the user the option of issuing a whole behavior to a vehicle instead of individual commands.
LabRatTM - The LabRat was designed by Rolla Engineered Solutions, LLC. It is a miniature robot with two motors, whiskers, infrared sensors and an optical mouse sensor. The poster presented at the IROS 2009 conference is below.
Self Organizing Project Teams - With some help from collaborators, I came up with an idea to allow researchers to choose which projects they would like to work on. I then wrote a program in Python to simulate researchers in this organization as compared to a more traditional organization. My paper (below) was accepted to the International Joint Conference on Neural Networks 2009.
After completing this research, I made some modifications to compare the IJCNN simulations with a third research organization inspired by Adaptive Resonance Theory. Below is the poster I presented on this at the International Conference on Cognitive and Neural Systems 2009.