Experimental Testbed Development for Long-Duration Deployments with Large Teams of Autonomous Micro Air Vehicles

Abstract

We propose to develop an experimental testbed to enable evaluation of coordinated MAV perception, control, and deployment algorithms over long durations with large teams of robots in a complex, but controlled, indoor experimental environment. The system will consist of thirty autonomous small-scale micro air vehicles as well as charging docks to enable continuous experimental evaluation over long durations (e.g., more than 10^2 - 10^3 power cycles) and an operator station for experiment monitoring. Critical to the success of robust autonomous systems operating over long durations is the development of algorithms that enable adaptation to platform and environmental changes and varying operator requirements. However, the evaluation of these algorithms is challenging due to the need to create an experimental system capable of operating over long periods of time without requiring operator interaction or oversight. Further, the time-scales associated with the autonomous system to sense, perceive, and react can vary greatly depending on the platform characteristic performance and environment variability. We propose to develop an experimental testbed that will enable long duration experimentation with a team of thirty small-scale autonomous micro air vehicles that exhibit fast response times and limited power cycle durations, prompting rapid system evolution and the need for autonomous adaptation. The system will leverage and build on existing hardware and software capabilities developed by the Robust Adaptive Systems Lab in the Field Robotics Center and Robotics Institute at Carnegie Mellon University. The proposed system will enable research in the areas of adaptive, resource-aware deployments, coordinated and cooperative exploration, vehicle introspection and robust, adaptive state estimation with inference-based plant and sensor model adaptation.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 08, 2016
Source ID
N000141512929

Entities

People

  • Nathan Michael

Organizations

  • Massachusetts Institute of Technology
  • Office of Naval Research
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Distributed Systems and Data Platform Development
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy
  • Autonomy - Autonomous System Control