Robot Electronics Architecture
- Created: Tuesday, 01 January 2008
Key features are modularity and expandability.
An electronics architecture has been developed to enable the rapid construction and testing of prototypes of robotic systems. This architecture is designed to be a research vehicle of great stability, reliability, and versatility. A system according to this architecture can easily be reconfigured (including expanded or contracted) to satisfy a variety of needs with respect to input, output, processing of data, sensing, actuation, and power.
The architecture affords a variety of expandable input/output options that enable ready integration of instruments, actuators, sensors, and other devices as independent modular units. The separation of different electrical functions onto independent circuit boards facilitates the development of corresponding simple and modular software interfaces. As a result, both hardware and software can be made to expand or contract in modular fashion while expending a minimum of time and effort.
To ensure modularity and reconfigurability, the architecture incorporates the PC/104 standard [an industry standard for compact, stackable modules that are fully compatible (in architecture, hardware, and software) with personal-computer data- and power-bus circuitry]. This feature also enables minimization of development costs through selection of off-the-shelf PC/104 components whenever possible.
Particularly notable is a capability for modular expansion to enable a single central processing unit (CPU) to supervise the simultaneous operation of a practically unlimited number of actuators. For this purpose, the architecture provides for each actuator a modular real-time control subsystem, independent of other such subsystems. The subsystem contains dedicated electronic hardware that drives the actuator to execute continuously updated arbitrary motions. The architecture includes a provision for control feedback in the form of outputs from any or all of a variety of sensors. Any or all actuators can be run independently and motions updated instantly, without reference to any prior motion profile.
A custom actuator-driver circuit board has been developed for this architecture to satisfy some power and mass constraints pertaining to a specific application. This board is capable of driving 12 motors simultaneously under computer control and is built on a standard PC/104 footprint.
The architecture includes several userand system-friendly features: Two independent inputs for panic buttons or watchdog functions enable manual, computer, or watchdog disablement of any or all boards, without affecting the computer. An independent circuit holds all actuators inactive until the computer sends an enabling signal. A single switch overrides all functions to enable manual control. Lights, test points, and outputs enable both the user and the computer to independently monitor the state of the board and internal circuit functions.
This work was done by Michael Garrett, Lee Magnone, Hrand Aghazarian, Eric Baumgartner, and Brett Kennedy of Caltech for NASA’s Jet Propulsion Laboratory.
Refer to NPO-41784, volume and number of this NASA Tech Briefs issue, and the page number.