Method of processing signals within a neural network to position arobot

• US 5,504,841 A
• Filed: 11/19/1993
• Issued: 04/02/1996
• Est. Priority Date: 11/06/1990
• Status: Expired due to Term

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Tani, Jun
• Primary Examiner(s): Moore, David K.
• Assistant Examiner(s): HAFIZ, TARIQ R
• Application Number:
  US08/154,995
• Time in Patent Office: 865 Days
• Field of Search: 395/11, 395/22, 395/24, 395/27, 395/81-89, 395/94, 395/904, 382/1, 382/12-15, 901/14-19
• US Class Current: 700/246
• CPC Class Codes: B25J 9/161 Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
• CPC Class Codes: B25J 9/1664 characterised by motion, path, trajectory planning

Abstract

A signal processing method for efficiently searching an optimum solution in a neural network by including a term of a nonlinear resistance in an equation of motion and changing such nonlinear resistance periodically. According to the method, the range of absolute values of connection weights between units in the neural network is limited by the equation of motion, hence preventing a prolonged search time that may otherwise be caused by excessive extension of the search scope beyond the requisite. A plurality of patterns are previously embedded or stored in the neural network and, upon input of a predetermined key pattern, the nonlinear resistance is changed periodically to recall a pattern similar to the key pattern, whereby any desired pattern can be searched or retrieved with rapidity and facility out of the complicated patterns. A process of calculating the next position of an articulated robot corresponding to an optimum solution is repeated while periodically changing a nonlinear resistance included in another equation of the positional energy of the robot, thereby acquiring the data of the robot path up to a desired goal.

Claims

1. A method of processing signals within a Hopfield neural network to successively position a robot from an initial position to a desired position by selecting an optimum path for the robot between the initial position and the desired position, comprising the steps of:

storing a plurality of patterns in the Hopfield neural network, the patterns representative of paths of the robot;

searching within the neural network for a stored pattern, the selected stored pattern representative of successive coordinates of the optimum path of the robot, the searching based upon data representative of the initial position of the robot, the desired position of the robot, undesired path coordinates and an equation of motion, the equation of motion including a term of nonlinear resistance;

changing the value of the nonlinear resistance as a function of a periodic equation, wherein a range of absolute values of connection weights between units is limited by the equation of motion, wherein said equation of motion is expressed as

space="preserve" listing-type="equation">mx+f(x, ω

t)=-ε

∇

E(x)

in which the term of said nonlinear resistance is represented by

space="preserve" listing-type="equation">f(x, ω

t)={d.sub.0 sin (ω

t)+d.sub.1 }x+d.sub.2 x.sup.2 sgn (x); and

successively positioning the robot in accordance with the selected stored pattern.

2. A method of processing signals within a neural network to successively position a robot from an initial position to a desired position by determining an optimum path for the robot between the initial position and the desired position, comprising the steps of:

receiving a data representative of the initial position of the robot, the desired position of the robot and undesired path coordinates;

repetitively calculating, within the neural network, successive coordinates of the optimum path of the robot based upon the actual motion of the robot, the data representative of the initial position of the robot, the desired position of the robot, undesired path coordinates and an equation of motion, the equation of motion including a term of nonlinear resistance, the nonlinear resistance having a value which changes as a function of a periodic equation, wherein a range of absolute values of connection weights between units is limited by the equation of motion, wherein said equation of motion is expressed as

space="preserve" listing-type="equation">mx+f(x, ω

t)=-ε

∇

E(x)

in which the term of said nonlinear resistance is represented by

space="preserve" listing-type="equation">f(x, ω

t)={d.sub.0 sin (ω

t)+d.sub.1 }x+d.sub.2 x.sup.2 sgn (x); and

successively positioning the robot in accordance with the calculated successive coordinates.

3. A method of processing signals within a Hopfield neural network to successively position a robot from an initial position to a desired position by selecting an optimum path for the robot between the initial position and the desired position, comprising the steps of:

storing a plurality of patterns in the Hopfield neural network, the patterns representative of paths of the robot;

searching within the neural network for a stored pattern, the selected stored pattern representative of successive coordinates of the optimum path of the robot, the searching based upon data representative of the initial position of the robot, the desired position of the robot, undesired path coordinates and an equation of motion, the equation of motion including a term of nonlinear resistance, wherein said equation of motion is expressed as

space="preserve" listing-type="equation">mx+f(x, ω

t)=-ε

∇

E(x)

in which the term of said nonlinear resistance is represented by

space="preserve" listing-type="equation">f(x, ω

t)={d.sub.0 sin (ω

t)+d.sub.1 }x+d.sub.2 x.sup.2 sgn (x);

changing the value of the nonlinear resistance as a function of a periodic equation; and

successively positioning the robot in accordance with the selected stored pattern.

4. A method of processing signals within a neural network to successively position a robot from an initial position to a desired position by determining an optimum path for the robot between the initial position and the desired position, comprising the steps of:

receiving a data representative of the initial position of the robot, the desired position of the robot and undesired path coordinates;

repetitively calculating within the neural network, successive coordinates of the optimum path of the robot based upon the actual motion of the robot, the data representative of the initial position of the robot, the desired position of the robot, undesired path coordinates and an equation of motion, the equation of motion including a term of nonlinear resistance, the nonlinear resistance having a value which changes as a function of a periodic equation, wherein said equation of motion is expressed as

space="preserve" listing-type="equation">mx+f(x, ω

t)=-ε

∇

E(x)

in which the term of said nonlinear resistance is represented by

space="preserve" listing-type="equation">f(x, ω

t)={d.sub.0 sin (ω

t)+d.sub.1 }x+d.sub.2 x.sup.2 sgn (x); and

successively positioning the robot in accordance with the calculated successive coordinates.