NASA SBIR 2005 Solicitation


SUBTOPIC TITLE:Intelligent Modular Systems
PROPOSAL TITLE:Broad Application of a Reconfigurable Motor Controller

SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Barrett Technology
625 Mount Auburn St.
Cambridge ,MA 02138 - 4555
(617) 252 - 9000

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
William   Townsend
625 Mount Auburn St.
Cambridge, MA  02138 -4555
(617) 252 - 9000

An ultra-miniature (<50 grams) high-performance brushless-motor controller, code named 'Puck', has been developed by Barrett for Earth-based mobile-manipulation use where efficiency, low mass, and robustness are critical attributes. Application of a distributed intelligent system will enable these Pucks to be used liberally across a variety of NASA's future satellite, articulated-machine, and robotic applications requiring low-level robust brushless commutation directed by high-level task intelligence. In addition to supporting space-vector control of brushless motors, each Puck carries a virtually unburdened 32-bit DSP running at 80 MHz with plenty of memory and high-speed serial communications to neighboring Pucks. So even as machines increase in degrees of freedom - and therefore complexity - the excess computational power provided by the Pucks increases proportionately, working in tandem to overcome increasingly complex controls issues. While another Barrett proposal submitted to NASA is focused on developing the hardware for a space-qualified Puck controller, this proposal focuses on a control architecture that leverages the distributed DSPs. Phase I will build an architecture that best leverages a distributed network of Pucks, such as a solution of computationally-intensive kinematics equations (e.g. Jacobian matrix), and local tasks, such as estimating precise realtime velocities and supporting series-elastic-actuator (SEA) strain-gages.

The proposed architecture for modular, scalable, distributed servomotor electronics designed for use in multi-axis and wheeled robots presents a significant opportunity to address important orbital, interplanetary, and moon/Mars robotics applications involving integrated systems for mobility and manual dexterity. The construction of mining facilities and habitat on the moon and Mars and the servicing of orbiting spacecraft without excessive exposure to human astronauts are two immediate areas that are directly impacted by this proposed research and development. The SBIR supports the increasing importance of complex machines requiring articulated servo-driven machines for all aspects of NASA exploration.

As the military and commercial sectors continue to ramp up the use of robotics, especially lightweight mobile platforms, the proposed distributed network of motor-controller modules would reduce system weight and control complexity of robotic systems including search & rescue robots, scout robots, advanced wheelchairs, and in-house robotic assistants. Furthermore, brushless DC servomotors are rapidly replacing brushed motors as the technology of choice as both the cost decreases and power increases of Neodymium-Iron magnets over Samarium-Cobalt. The torque density and reliability of permanent-magnet motors have always exceeded those of brushed motors, but the large size and poor power efficiency of the support electronics generally outweigh the benefits except where power conservation is not important and when the controllers can be located off the machine. Providing a power-efficient module that integrates all of the electronic and optical functions into a volume smaller than most motor position sensors suddenly alters the decision point in favor of brushless motors. Most servo-driven machines use multiple servomotors in concert. Each of these systems would benefit from an intelligent means to distribute motion-control intelligence as described in this proposal.

NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.

Architectures and Networks
Autonomous Control and Monitoring
Autonomous Reasoning/Artificial Intelligence
Computer System Architectures
Data Acquisition and End-to-End-Management
Data Input/Output Devices
Database Development and Interfacing
Integrated Robotic Concepts and Systems
On-Board Computing and Data Management
Software Tools for Distributed Analysis and Simulation
Ultra-High Density/Low Power

Form Printed on 09-19-05 13:12