3-brain architecture for an intelligent decision and control system

• US 6,581,048 B1
• Filed: 05/10/1999
• Issued: 06/17/2003
• Est. Priority Date: 06/04/1996
• Status: Expired due to Fees

• Current Assignee: IPU Power Management, LLC
• Original Assignee: Paul J. Werbos
• Inventors: Werbos, Paul J.
• Primary Examiner(s): Powell, Mark R.
• Assistant Examiner(s): STARKS, WILBERT L
• Application Number:
  US09/147,338
• Time in Patent Office: 1,499 Days
• Field of Search: 706/16, 706/23, 250/369
• US Class Current: 706/23
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 13/027 using neural networks only
• CPC Class Codes: G05B 13/0285 using neural networks and fuzzy logic
• CPC Class Codes: G06N 3/105 Shells for specifying net layout

Abstract

A method and system for intelligent control of external devices using a mammalian brain-like structure having three parts. The method and system include a computer-implemented neural network system which is an extension of the model-based adaptive critic design and is applicable to real-time control (e.g., robotic control) and real-time distributed control. Additional uses include data visualization, data mining, and other tasks requiring complex analysis of inter-relationships between data.

Claims

1. A computer program product, comprising:

a computer storage medium and a computer program code mechanism embedded in the computer storage medium for causing a computer to control an external physical device, the computer program code mechanism comprising;

a first computer code device configured to create plural decision blocks, each such decision block including plural action choices, a set of exit states, and a set of post-exit states, wherein a subset of the plural decision blocks can be entered from each post-exit state;

a second computer code device configured to receive inputs;

a third computer code device configured to select a post-exit state of one of the plural decision blocks based on the received inputs;

a fourth computer code device configured to select one of the plural decision blocks that can be entered from the selected post-exit state and that maximizes a local cost function for the selected post-exit state; and

a fifth computer code device configured to produce at least one output based on the selected one of the plural decision blocks.

2. The computer code device as claimed in claim 1, wherein the post-exit states of at least one block includes states for failure and changes in goal objectives.

3. The computer code device as claimed in claim 1, further comprising a sixth computer code device configured to update a global cost function based on the selected one of the plural decision blocks.

4. The computer code device as claimed in claim 1, further comprising a sixth computer code device configured to update a local cost estimate corresponding to the selected one of the plural decision blocks without updating a local cost estimate of the plural decision blocks not selected by the fourth computer code device.