Agile and Dexterous Robot for Inspection and EOD Operations

U.S. Army / TARDEC, 2007-present

The EOD All-Terrain Biped® (EOD/ATB) is a robotic platform with wheels, legs and arms capable of driving, crawling, walking and manipulating objects for inspection and explosive ordnance disposal tasks. Limb coordination technology provides independent control of balance and posture and intelligent behaviors automatically select locomotion mode based on the operational environment.

Phase I involved the preliminary design of the ATB® prototype. Simulated locomotion and manipulation behaviors were developed as well as an intuitive operator interface.

In Phase II the ATB prototype is being built and tested. Sensors include an IMU and a scanning laser rangefinder for object and terrain recognition. Performance goals include driving on flat terrain, crawling on steep terrain, walking on stairs, opening doors and grasping objects.

Anticipated benefits of the terrain-adaptive mobility and dual-arm dexterity of the ATB platform include increased robot agility and autonomy for EOD operations, reduced operator workload and reduced operator training and skill requirements.


Tactical All-Terrain Biped

DARPA / U.S. Navy SPAWAR, 2005-2006


The goal of the Tactical All-Terrain Biped (T/ATB) project was to design and build a two-legged robot that is strong, agile and adaptive. The legs-only prototype demonstrates walking on uneven terrain using a real-time stereo vision system. Simulation studies show control of a full-body configuration capable of transitioning between walking and crawling.


Scout

Internal R&D / Education, 2005


The Scout robot and American Android's Behavior Development Studio® software illustrate the educational potential of ATB technology.


Shorty

Internal R&D, 2003


Shorty is an early ATB prototype demonstrating simultaneous control of balance and posture.


In Situ Training of Robonaut for In-Space Assembly, Maintenance and Servicing

NASA, 2005


Control technology for in-the-field training of NASA's Robonaut for tool manipulation and multi-robot collaboration was developed.


In Situ Training of Anthropomorphic Robots

NASA, 2001-2004


Control technology enabling human operators to teach anthropomorphic robots, in the field, how to perform new complex tasks was developed. Building upon existing inverse kinematics, rule-based control and neural network learning technology, the training method enhances robot capabilities through operator supervision.

Rules provide a modular and hierarchical way to specify plans of action, and neural networks provide a context-dependent form of skill memory. The FieldTrainer system enables the online construction of rule-based plans through verbal dialog between operator and robot. It also uses verbal, visual and manual cues such as spoken words, hand gestures and the pushing of buttons, along with neural network learning, to augment nominal rule-based motion and thereby shape robot behavior.

 
  home | about us | contact


© 2010 American Android Corp.