EAPSI: A Common Sensor and Software Approach For Humanoid Material-Handling Missions in Disaster Scenarios

Hament Blake, Las Vegas NV

Investigators

Abstract

Major disasters like Hurricane Katrina and the Fukushima nuclear accident strike quickly but leave regions devastated for years afterwards. Even a fractional reduction in cleanup time can drastically reduce economic opportunity costs and low social morale while more quickly returning these communities to pre-disaster quality of life. One possibility for reducing cleanup time is the deployment of humanoid robots. In the 2015 DARPA challenge, teams from around the world demonstrated that humanoid robots have the ability to navigate and interact with their world with unprecedented complexity: driving, using tools, opening doors, and manipulating facility controls. The biggest obstacle to overcome before humanoids can be deployed for cleanup is developing better material-handling capabilities like lifting, pushing, pulling, and stacking. As roboticists develop these capabilities, their humanoid robots are evolving different software and sensor approaches. This project will compare the software and sensors of an American and a Korean humanoid being developed for disaster cleanup, identify disparities in configuration, and determine optimum configuration for material-handling missions. The project will be conducted at Korea Advanced Institute for Science and Technology under the mentorship of Dr. Jun Ho Oh, whose team placed first in the 2015 DARPA Robotics Challenge. The humanoids being compared in this study are DRC Hubo 2 platforms. The American and Korean Hubos use many of the same scripts and hardware, but differences have evolved as both groups continue to develop the platform independently. In cases that these differences significantly affect performance, the researcher will use a combination of simulated and real material-handling trials to determine which software and sensor approach produces superior success rates and completion times. This award, under the East Asia and Pacific Summer Institutes program, supports summer research by a U.S. graduate student and is jointly funded by NSF and the National Research Foundation of Korea.