Method, computer program product and robot control system for the contact-based localization of objects that can be moved when manipulated by robot, and robot

Patent metadata

Field	Value
Publication number	US-12076865-B2
Application number	US-202017610168-A
Country	US
Kind code	B2
Filing date	May 11, 2020
Priority date	May 17, 2019
Publication date	Sep 3, 2024
Grant date	Sep 3, 2024

Abstract

At least one robot is disclosed for object manipulation, on the basis of a probabilistic approach without object detection by visual sensors, a probability distribution is determined approximately by an iterative, recursive application of a Bayes filter algorithm, wherein state transition probabilities obtained, in a nondeterministic motion model for system simulation, are multiplicatively linked with measurement probabilities obtained at the start of the application. For this purpose, a physics simulator which completely includes the physics of the system with respect to forces and dynamics and the physical system relationships resulting therefrom, and a controller, which controls the physics simulator while at the same time the simulation system state is fed back and which influences the compliance of the robot with respect to the robot-object environment on the basis of control variables, are incorporated, and measurement results of axis-specific measurements are taken into account in the measurement model for the plausibility checks.

First claim

The invention claimed is:

1. A method for a contact-based localization of objects that can be moved when manipulated by a robot, wherein for a state associated with the object localization of a system formed by at least one manipulated-variable-controlled robot, an object that is movable for the robot during the object manipulation and a robot-object environment that enables the object manipulation as system components, and in the course of a system state estimation, the method comprises:
determining, by an iterative-recursive application of a Bayes filter algorithm, a probability distribution that localizes the object, wherein said determining by the iterative-recursive application of the algorithm obtains: (i) state transition probabilities, in a nondeterministic motion model for system simulation, from modeled transitions of states of the system, and fil) measurement probabilities, in a measurement model for system simulation, wherein an initial probability distribution is used at the start;
using an approximation algorithm in the iterative-recursive determination of the probability distribution;
controlling, by a controller execution of a physics simulator with a simulation system state being fed back and, on the basis of a manipulated variable fed to the controller, wherein the controller controls the physics simulator in such a way that during the system state simulation the robot behaves compliantly with respect to the robot-object environment, wherein the physics simulator is configured to use the physics of the system in regard to forces and dynamics and the resultant physical relationships of the system components;
performing model axis-specific measurements, on the robot, wherein measurement results of the axis-specific measurements are axial position, axial velocity, axial acceleration and/or axial torque of a robot ROB, wherein the robot ROB is configured to be manipulated to move the object.

2. The method as claimed in claim 1 , wherein said determining the probability distribution comprises using a particle filter algorithm for approximating the Bayes filter algorithm in the iterative-recursive application of the Bayes filter algorithm.

3. The method as claimed in claim 1 , wherein the manipulated variable is formed from a setpoint position, setpoint velocity and/or a setpoint acceleration.

4. The method as claimed in claim 1 , wherein the controller is operated in an impedance controlled manner which together with the manipulated variable fed to the controller causes the robot to behave compliantly with respect to the robot-object environment.

5. The method as claimed in claim 1 , wherein the measurement of the axis-specific measurements are measurement results of axial position, axial velocity, axial acceleration and axial torque of the robot ROB.

6. The method as claimed in claim 1 , said method further comprising:
forming a multiplicative combination of the measurement probabilities and system state transitions probabilities.

7. A computer program product, comprising a computer readable hardware storage device having computer readable program code stored therein, said program code executable by a processor of a computer system to implement a method for a contact-based localization of objects that can be moved when manipulated by a robot, wherein for a state associated with the object localization of a system formed by at least one manipulated-variable-controlled robot, an object that is movable for the robot during the object manipulation and a robot-object environment that enables the object manipulation as system components, and in the course of a system state estimation, the method comprises:
determining, by an iterative-recursive application of a Bayes filter algorithm, a probability distribution that localizes the object, wherein said determining by the iterative-recursive application of the algorithm obtains: (i) state transition probabilities, in a nondeterministic motion model for system simulation, from modeled transitions of states of the system, and (ii) measurement probabilities, in a measurement model for system simulation, wherein an initial probability distribution is used at the start;
using an approximation algorithm in the iterative-recursive determination of the probability distribution;
controlling, by a controller execution of a physics simulator with a simulation system state being fed back and, on the basis of a manipulated variable fed to the controller, wherein the controller controls the physics simulator in such a way that during the system state simulation the robot behaves compliantly with respect to the robot-object environment, wherein the physics simulator is configured to use the physics of the system in regard to forces and dynamics and the resultant physical relationships of the system components;
performing model axis-specific measurements, on the robot, wherein measurement results of the axis-specific measurements are axial position, axial velocity, axial acceleration and/or axial torque of a robot ROB, wherein the robot ROB is configured to be manipulated to move the object.

8. The computer program product as claimed in claim 7 , wherein said determining the probability distribution comprises using a particle filter algorithm for approximating the Bayes filter algorithm in the iterative-recursive application of the Bayes filter algorithm.

9. The computer program product as claimed in claim 7 , wherein the manipulated variable is formed from a setpoint position, setpoint velocity and/or a setpoint acceleration.

10. The computer program product as claimed in claim 7 , wherein the controller is operated in an impedance controlled manner which together with the manipulated variable fed to the controller causes the robot to behave compliantly with respect to the robot-object environment.

11. The computer program product as claimed in claim 7 , wherein the measurement results of the axis-specific measurements are measurement results of axial position, axial velocity, axial acceleration and axial torque of the robot ROB.

12. A robot control system for the contact-based localization of objects that can be moved when manipulated by robot, characterized by the computer program product as claimed in claim 7 for carrying out the method.

13. The robot with a robot control system as claimed in claim 12 .

14. The computer program product as claimed in claim 7 , said method further comprising:
forming a multiplicative combination of the measurement probabilities and system state transitions probabilities.

Assignees

• Siemens Ag

Inventors

• Kast Bernd
• Schmitt Philipp Sebastian
• Wirnshofer Florian

Classifications

• G05B2219/50391

  Robot · CPC title

• G05B19/4155

  characterised by program execution, i.e. part program or machine function execution, e.g. selection of a program · CPC title

• B25J13/088

  with position, velocity or acceleration sensors · CPC title

• G05B2219/39018

  Inverse calibration, find exact joint angles for given location in world space · CPC title

• G05B2219/39013

  Locate movable manipulator relative to object, compare to stored gridpoints · CPC title