Extended task space control for robotic manipulators

Claim 3

The method of claim 1 wherein said computing a first command kinematic quantity comprises the computing of a commanded acceleration from a commanded velocity and a commanded position, and from behavior forces, behavior accelerations, behavior velocities and behavior positions.

Claim 4

The method of claim 1 wherein said computing a first command kinematic quantity comprises the computing of a commanded velocity from a commanded acceleration and a commanded position.

Claim 5

The method of claim 1 wherein: said common space is a motion space common to which all of said behavior kinematic quantities are transformable; and said controlling said joint actuators is preceded by second transforming at least some of said command kinematic quantities from said motion space to an actuator space of said joint actuators.

Claim 6

The method of claim 4 wherein: said robot further comprises joint position sensors for detecting positions of said joints; and said second transforming comprises computing a transform to said actuator space based upon current angles of said joints detected by said joint position sensors.

Claim 7

The method of claim 5 wherein: said decomposing defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces of each of said behaviors; said first transforming comprises transforming the behavior kinematic quantities of each respective behavior from the corresponding behavior space to said motion space.

Claim 8

The method of claim 6 wherein said decomposing step defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces each having a dimensionality corresponding to respective degrees of freedom of the robot.

Claim 9

The method of claim 7 wherein said decomposing step defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces each having a dimensionality that does not exceed a maximum number of degrees of freedom of said robot.

Claim 10

The method of claim 1 wherein said decomposing step comprises the decomposing of said task into behavior kinematic quantities of said behaviors wherein each said behavior is governed in accordance with a corresponding input parameter of the type including force set points, joint singularity angle, joint limit angle, and obstacle location, said method further comprising reprogramming selected ones of said input parameters each sample interval, whereby to dynamically control said plural behaviors.

Claim 11

The method of claim 1 wherein said command kinematic quantities comprise acceleration, velocity and position, said behavior kinematic quantities comprise force, acceleration, velocity and position, and wherein the computing of said command kinematic quantity comprises equating a sum of proportionate differences between command and behavior accelerations, velocities and positions, respectively, with a sum of measured forces of said robot and of behavior forces.

Claim 12

The method of claim 10 wherein said computing of said command kinematic quantity further comprises making said proportionate differences proportionate in accordance with inertia damping and stiffness characteristics respectively, of said robot.

Claim 15

The method of claim 12 wherein: said common space is a motion space common to which all of said behavior forces, accelerations, velocities and positions are transformable; and said controlling said joint actuators is preceded by second transforming said command accelerations, velocities and positions from said motion space to an actuator space of said joint actuators.

Claim 16

The method of claim 13 wherein: said robot further comprises joint position sensors for detecting positions of said joints; and said second transforming comprises computing a transform to said actuator space based upon current angles of said joints detected by said joint position sensors.

Claim 17

The method of claim 14 wherein: said decomposing defines the behavior forces, accelerations, velocities and positions of each of said behaviors in respective behavior spaces of each of said behaviors; said first transforming comprises transforming the behavior forces, accelerations, velocities and positions of each respective behavior from a corresponding one of said respective behavior space to said motion space.

Claim 18

The method of claim 15 wherein the decomposing step defines the behavior forces, accelerations, velocities and positions of each of said behaviors in respective behavior spaces each having a dimensionality corresponding to respective degrees of freedom of the robot.

Claim 19

The method of claim 16 wherein the decomposing step defines the behavior forces, accelerations, velocities and positions of each of said behaviors in respective behavior spaces each having a dimensionality that does not exceed the maximum number of degrees of freedom of an arm.

Claim 20

The method of claim 17 wherein the decomposing step defines the behavior forces, accelerations, velocities and positions of an avoidance of one of a joint limit, joint singularity or obstacle, for which the corresponding behavior kinematic quantity is a function of the proximity of the joint limit, joint singularity or obstacle, respectively.

Claim 21

The method of claim 12 wherein the decomposing step defines the behavior forces, accelerations, velocities and positions of said behaviors wherein each said behavior is governed in accordance with a corresponding input parameter of the type including force set points, joint singularity angle, joint limit angle, obstacle location, and wherein any one and all said input parameters are reprogrammable at each sample interval, whereby to provide dynamic control of said plural behaviors.

Claim 22

The method of claim 12 wherein said computing said command accelerations comprises equating a sum of proportionate differences between command and behavior accelerations, velocities and positions, respectively, with a sum of measured forces of said robot and of said behavior forces.

Claim 23

The method of claim 20 wherein said computing said command accelerations further comprises making said proportionate differences proportionate in accordance with inertia, damping and stiffness characteristics, respectively, of said robot.

Claim 24

A robot control system for operating a robot in successive sampling intervals to perform a task in an environment of objects, said robot having plural joints and joint actuators, said system comprising: means for decomposing said task into behavior kinematic quantities of plural behaviors to be exhibited simultaneously in respective behavior spaces by said robot, said plural behaviors including exerting force on at least one of said objects; means for transforming said behavior kinematic quantities from said respective behavior spaces to a common space; and means for computing, in said common space, from said behavior kinematic quantities command kinematic quantities including a force kinematic quantity corresponding to the force to be exerted on said one object and controlling said joint actuators in accordance with said command kinematic quantities; wherein said means for computing comprises a second means for computing a first command kinematic quantity for a next sampling interval from: (a) behavior kinematic quantities, and (b) other command kinematic quantities of a current sampling interval; and third means for computing said other command kinematic quantities for said next sampling interval from said first command kinematic quantity of said next sampling interval and from said other command kinematic quantities of said current sampling interval.

Claim 25

A robot control system for operating a robot in successive sampling intervals to perform a task in an environment of objects, said robot having plural joints and joint actuators, said system comprising: means for decomposing said task into behavior kinematic quantities of plural behaviors to be exhibited simultaneously in respective behavior spaces by said robot, said plural behaviors including exerting force on at least one of said objects; means for transforming said behavior kinematic quantities from said respective behavior spaces to a common space; and means for computing, in said common space, from said behavior kinematic quantities command kinematic quantities including a force kinematic quantity corresponding to the force to be exerted on said one object and controlling said joint actuators in accordance with said command kinematic quantities; wherein said means for computing comprises a second means for computing a first command kinematic quantity for a next sampling interval from: (a) behavior kinematic quantities, and (b) other command kinematic quantities of a current sampling interval; and third means for computing said other command kinematic quantities for said next sampling interval from said first command kinematic quantity of said next sampling interval and from said other command kinematic quantities of said current sampling interval.

Claim 26

The system of claim 22 wherein said second means for computing comprises means for computing a commanded acceleration from commanded velocity and commanded position, and from behavior forces, behavior accelerations, behavior velocities and behavior positions.

Claim 27

The system of claim 22 wherein said second means for computing comprises means for computing a commanded velocity from commanded acceleration and commanded position.

Claim 28

The system of claim 22 wherein: said common space is a motion space common to which all of said behavior kinematic quantities are transformable; and said first means for computing and controlling said joint actuators comprises second means for transforming at least some of said command kinematic quantities from said motion space to an actuator space of said joint actuators.

Claim 29

The system of claim 25 wherein: said robot further comprises joint position sensors for detecting positions of said joints; and said second means for transforming comprises means for computing a transform to said actuator space based upon current angles of said joints detected by said joint position sensors.

Claim 30

The system of claim 26 wherein: said means for decomposing defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces of each of said behaviors; said first means for transforming comprises means for transforming the behavior kinematic quantities of each respective behavior from the corresponding behavior space to said motion space.

Claim 31

The system of claim 27 wherein the means for decomposing defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces each having a dimensionality corresponding to respective degrees of freedom of the robot.

Claim 32

The system of claim 28 wherein said means for decomposing defines the behavior kinematic quantities of each of said behaviors in respective behavior spaces each having a dimensionality that does not exceed the maximum number of degrees of freedom of said arm.

Claim 33

The system of claim 22 wherein said command kinematic quantities comprise acceleration, velocity and position, said behavior kinematic quantities comprise force, acceleration, velocity and position, and wherein said first means for computing comprises means for equating a sum of proportionate differences between command and behavior accelerations, velocities and positions, respectively, with a sum of measured forces of said robot and of behavior forces.

Claim 34

The system of claim 30 wherein the means for computing comprises means for equating the sum of said differences wherein said differences are proportionate in accordance with inertia, damping and stiffness characteristics, respectively, of said robot.