Granted Patent B2

US 12,533,804 · App. 18/288,620 · Granted Jan 27, 2026

Inventor: Erik Mankin (Karlsruhe, DE)

Assignee: PHYSIK INSTRUMENTE (PI) SE & CO. KG

B25J9/1653B25J9/1607B25J9/1664B25J9/1623

Abstract

Claims (83)

1 . A method for controlling a kinematic system, the method comprising:

determining, from a plurality of prespecified spatial units of the work space of the kinematic system, a spatial unit in which a target pose of the kinematic system is located;

estimating a configuration vector of the kinematic system associated with the target pose in accordance with indirect kinematics, wherein:

the estimating is performed in accordance with an interpolation of the indirect kinematics, and

the interpolation for poses in the determined spatial unit is based on predetermined configuration vectors, each of which is associated with a boundary point of the determined spatial unit in accordance with the indirect kinematics;

determining a target configuration vector using the estimated configuration vector; and

actuating the kinematics with the target configuration vector,

wherein the boundary points with which the predetermined configuration vectors are associated include or are the corner points of the prespecified spatial units.

2 . The method according to claim 1 , wherein

in the determining of the target configuration vector, the target configuration vector is calculated using an iterative method, wherein a start vector of the iterative method is selected in accordance with the estimated configuration vector.

3 . The method according to claim 1 , wherein

in the determining of the target configuration vector, it is determined that the target configuration vector is the estimated configuration vector.

4 . The method according to claim 1 , wherein

the plurality of the prespecified spatial units represent a partitioning of a part of the work space.

5 . The method according to claim 4 , wherein

the partitioning of the part of the work space is based on:

a partitioning into spatial units having the same dimensions; and/or

a subdivision of work space coordinates into respective intervals.

6 . The method according to claim 5 , wherein:

the prespecified spatial units have different dimensions and partition hierarchically the part of the work space.

7 . The method according to claim 4 , wherein:

the prespecified spatial units have different dimensions and partition hierarchically the part of the work space.

8 . The method according to claim 1 , wherein

one or more of the prespecified spatial units are hyperrectangles, and the interpolation for poses in the one or more of the prespecified spatial units that are hyperrectangles includes or is a multilinear interpolation, and/or

one or more of the prespecified spatial units are simplexes, and the interpolation for poses in the one or more of the respecified spatial units that are simplex includes or is a barycentric interpolation.

9 . The method according to claim 1 , wherein

the interpolation for poses in one of the prespecified spatial units is based on a sum composed of a spatial unit interpolation and a correction interpolation, wherein

the correction interpolation in each d-simplex of a plurality of d-simplexes that partition a prespecified spatial unit is a respective barycentric interpolation,

at least one corner point of one of the d-simplexes is located in an interior of the prespecified spatial unit, and

d is the dimension of the prespecified spatial unit, and

for each of the d-simplexes, the respective barycentric interpolation interpolates a function which:

assigns a function value to each of those corner points of the d-simplex that are located within the prespecified special unit, the function value corresponding to a difference between:

a configuration vector associated with the corner point according to indirect kinematics; and

a configuration vector associated with the corner point according to the spatial unit interpolation, and

assigns 0 as a function value to those corner points of the d-simplex that are located on the surface of the one prespecified spatial unit.

10 . The method according to claim 1 , wherein

the interpolation for poses in each of the plurality of prespecified spatial units is based on a sum composed of a correction interpolation and a spatial unit interpolation associated with the prespecified spatial unit, wherein

the correction interpolation in each d-simplex of a plurality of d-simplexes that partition a union of the plurality of the prespecified spatial units is a respective barycentric interpolation,

at least one corner point of one of the d-simplexes is located within the union, and

d is the dimension of the prespecified spatial units, and

for each of the d-simplexes, the respective barycentric interpolation interpolates a function which:

assigns a function value to each of those corner points of the d-simplex that are located within the union, the function value corresponding to a difference between:

a configuration vector associated with the corner point according to indirect kinematics; and

a configuration vector associated with the corner point in accordance with the spatial unit interpolation associated with a prespecified spatial unit in which the corner point is located, and

assigns 0 as a function value to those corner points of the d-simplex that are located on the surface of the union.

11 . The method according to claim 1 , further comprising:

providing the plurality of prespecified spatial units and the predetermined configuration vectors, comprising:

obtaining the prespecified spatial units by performing or reading in a partitioning of a part of the work space; and

obtaining the predetermined configuration vectors by calculating the indirect kinematics or by reading in.

12 . The method according to claim 1 , further comprising:

adapting the plurality of prespecified spatial units and the predetermined configuration vectors, comprising:

adding to the plurality of prespecified spatial units a plurality of new spatial units formed by dividing one of the prespecified spatial units;

removing from the plurality of prespecified spatial units the one prespecified spatial unit; and

providing those predetermined configuration vectors, each of which is associated with a point that is no corner point of the one prespecified spatial unit:

by calculating the indirect kinematics or by reading in if the point is no boundary point of the one prespecified spatial unit; and

by estimating in accordance with the interpolation of the indirect kinematics if the point is a boundary point of the one prespecified spatial unit,

wherein the boundary points with which the predetermined poses are associated include or are the corner points of the prespecified spatial units.

13 . A method for controlling a kinematic system, the method comprising:

ascertaining a current configuration vector of the kinematic system;

determining, from a plurality of prespecified spatial units of the configuration space of the kinematic system, a spatial unit in which the ascertained configuration vector is located;

estimating a current pose of the kinematic system that is associated with the ascertained configuration vector in accordance with direct kinematics, wherein

the estimating is performed in accordance with an interpolation of the indirect kinematics, and

the interpolation for configuration vectors in the determined spatial unit is based on predetermined poses of the kinematic system, each of which is associated with a boundary point of the determined spatial unit in accordance with the direct kinematics;

determining a current pose using the estimated pose; and

outputting the determined current pose,

wherein the boundary points with which the predetermined configuration vectors are associated include or are the corner points of the prespecified spatial units.

14 . A control device for controlling a kinematic system, wherein said control device is configured to:

determine, from a plurality of prespecified spatial units of the work space of the kinematic system, a spatial unit in which a target pose of the kinematic system is located;

estimating a configuration vector of the kinematic system associated with the target pose in accordance with indirect kinematics, wherein:

the estimating is performed in accordance with an interpolation of the indirect kinematics; and

the interpolation for poses in the determined spatial unit is based on predetermined configuration vectors, each of which is associated with a boundary point of the determined spatial unit in accordance with the indirect kinematics;

determine a target configuration vector using the estimated configuration vector; and

actuating the kinematic system with the target configuration vector,

wherein the boundary points with which the predetermined poses are vectors include or are the corner points of the prespecified spatial units.

15 . A control device for controlling a kinematic system, wherein said control device is configured to:

ascertain a current configuration vector of the kinematic system;

determine, from a plurality of prespecified spatial units of the configuration space of the kinematic system, a spatial unit in which the ascertained configuration vector is located;

estimate a current pose of the kinematic system that is associated with the ascertained configuration vector in accordance with direct kinematics, wherein

the estimating is performed out in accordance with an interpolation of the indirect kinematics, and

the interpolation for configuration vectors in the determined spatial unit is based on predetermined poses of the kinematic system, each of which is associated with a boundary point of the determined spatial unit in accordance with the direct kinematics;

determine a current pose using the estimated pose; and

outputting the determined current pose,

wherein the boundary points with which the predetermined poses are associated include or are the corner points of the prespecified spatial units.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (20)