A LANGUAGE-BASED SOFTWARE FRAMEWORK FOR MISSION PLANNING IN AUTONOMOUS MOBILE ROBOTS

Open Access
Author:
Shah, Mirza Akbar
Graduate Program:
Computer Science and Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 01, 2011
Committee Members:
  • John Joseph Hannan, Thesis Advisor
Keywords:
  • Software Engineering
  • Software Framework
  • Autonomy
  • Autonomous Robots
  • Autonomous Vehicles
  • Planning
  • Domain-Specific Programming Languages
  • Artificial Intelligence
  • Distributed Planning
  • Hierarchical Planning
  • Multi-Agent Planning
Abstract:
Autonomous mobile robots have many uses in both commercial and military applications. Creating the computer software that implements the autonomy of these robots is a daunting task, which has resulted in the development of many tools across both industry and academia to help mitigate development complexity and costs in the form of software libraries, frameworks, and application programmer interfaces. This thesis describes a new software framework, the Modular Planning Framework and Language (MPFL), for developing the mission planning aspect of a mobile robot’s autonomy. MPFL is unique in that it looks at the problem of vehicle planning as a programming language with a corresponding runtime compiler that processes the language. This approach to autonomy development leverages concepts from programming language theory to create a planning framework that supports component reuse, strong data type verification, and the ability to reason about complex autonomy solutions in a piecemeal fashion.