Dynamo: A Model Transition Framework for Dynamic Stability Control and Body Mass Manipulation

Abstract

Ground robots, such as the iRobot PackBot, have saved hundreds of lives in Iraq and Afghanistan by helping soldiers safely inspect and disarm improvised explosive devices (IEDs) (Figure 1). However, even state-of-the-art robot control systems fail to approach the adaptive, versatile mobility demonstrated by humans and animals on an everyday basis. In order to extend the applicability of ground robots to a wider range of missions, fundamental advances are needed to provide robust mobility in unstructured, dynamic, natural environments. One of the common criticisms of currently deployed UGVs is that they are too slow. The typical standoff distance for an EOD team investigating an IED is 200-500 meters. At top speed (5.8 mph), a PackBot would take over 3 minutes to travel downrange to an IED 500 meters away. Though a defensive perimeter is setup around the EOD teams, 3 minutes is long enough to setup and deploy mortars or for snipers to take action. Reducing this time window is critical. The Robotic Systems Joint Project Office (RSJPO) has identified the need for faster UGVs as a top priority for Army and Marine Corps warfighters [Jackowski 10]. We envision UGVs that can travel 30-45 mph or faster which would enable their use in rapid reconnaissance and high-optempo infantry assault missions. At 45 mph, a high-speed UGV could reach an IED 500 meters away in just 25 seconds. However, high-speed UGVs are difficult to control manually, particularly on the diret or sand surfaces that are common on the battlefield. On these surfaces, the vehicle can ski or lose traction while cornering and can become airborne from even modest terrain variation.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2011
Accession Number
ADA553515

Entities

People

  • Brian Yamauchi

Organizations

  • iRobot

Tags

Communities of Interest

  • Autonomy
  • Ground and Sea Platforms
  • Weapons Technologies

DTIC Thesaurus Topics

  • Adaptive Systems
  • Artificial Intelligence
  • Center Of Gravity
  • Collision Avoidance
  • Computer Programs
  • Control Systems
  • Explosive Devices
  • Feasibility Studies
  • Friction
  • Improvised Explosive Devices
  • Machine Learning
  • Reinforcement Learning
  • Simulators
  • Transitions
  • Two Dimensional
  • Unmanned Ground Systems
  • Unmanned Ground Vehicles

Fields of Study

  • Computer science

Readers

  • Educational Psychology
  • Robotics and Automation.
  • Unmanned Aerial System (UAS) Autonomous Capabilities and Mission Reconnaissance.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - DoD AI Strategy
  • Autonomy