Compliant Robot Boundary Control via Adaptive Restoring Force

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for controlling flexible robots in restricted areas can result in unexpected and undesired massive return movements when border monitoring is activated, especially if the robot is far from the border, leading to unpredictable behavior.

Innovation Solution

A method and controller that differentiate between a first and second restoring force based on the robot's position relative to the border, with the first force being temporarily smaller or zero when border monitoring is activated, and a direction-dependent, time-dependent stiffness of a virtual spring that increases resistance for movements away from the border, reducing the likelihood of such movements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If a high restoring force is applied to drive the robot back from the restricted area when boundary monitoring is activated, then the robot is quickly returned to the boundary, but this causes unexpected and undesired massive reversal movements leading to unpredictable behavior

Engineering Contradiction:
Improvespeed of returning robot to boundaryVSAvoidpredictability of robot behavior
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The control system dynamically adjusts the restoring force based on the robot's position relative to the boundary. When the robot is far from the boundary, a lower restoring force is applied to prevent sudden movements. As the robot approaches the boundary, the restoring force increases to ensure timely return. This dynamic adjustment resolves the contradiction by making the force adaptive rather than constant.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The control system changes the parameter of restoring force magnitude based on the robot's distance from the boundary. By monitoring the robot's position and adjusting the force parameter accordingly, the system prevents massive reversal movements while maintaining effective boundary enforcement. This parameter change approach allows the system to balance speed and predictability.

Inventive Principle:
Principle #35Parameter changes

2Stability of the object's composition

If the robot is technically bound to its current position with high stiffness and damping, then the robot remains stable at its position, but this prevents the robot from being driven back out of the restricted area when needed

Engineering Contradiction:
Improvestability of robot positionVSAvoidability to return robot to boundary
Core Design Contradiction:
Stability of the object's compositionVSEase of operation

Solution Approach 1:

The control system dynamically adjusts stiffness and damping parameters based on the robot's position. When the robot is within the restricted area, high stiffness and damping are applied to maintain stability. When the robot approaches or crosses the boundary, these parameters are reduced to allow the restoring force to effectively move the robot back. This dynamic parameter adjustment resolves the contradiction between stability and ease of operation.

Inventive Principle:
Principle #15Dynamics

3Reliability

If border monitoring is activated when the robot is already far from the border within the restricted area, then boundary monitoring coverage is maximized, but this immediately results in a massive reversal movement of the robot

Engineering Contradiction:
Improveboundary monitoring coverageVSAvoidsmoothness of robot operation
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The control system applies a preliminary counteracting force that is limited in magnitude when boundary monitoring is activated. Instead of immediately applying the full restoring force that would cause a massive reversal movement, the system applies a controlled, limited force that gradually guides the robot back to the boundary. This preliminary anti-action prevents the harmful effect while maintaining monitoring coverage.

Inventive Principle:
Principle #9Preliminary anti-action

Solution Approach 2:

The control system cushions the potential massive reversal movement by applying a limited restoring force that acts as a buffer. This cushioning effect prevents sudden, jerky movements while still enforcing the boundary constraint. The limited force acts as a protective measure that maintains both monitoring coverage and operational smoothness.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentEP3117968B1Control of a flexibly controlled robot
Publication Date: 2020.11.11 KUKA DEUT GMBH
  • EP3117968B1 patent drawingFigure 1~3
  • EP3117968B1 patent drawingFigure 4

AI summary

A method according to the invention for controlling a compliant robot (1) comprises the step of: performing boundary monitoring of the robot; wherein a first restoring force (T1), which drives the robot from a current position (x0) in a restricted area (S: x>s) to a boundary (s) of this area, is specified by the control system (S40), provided that the robot is already in this position in the restricted area when the boundary monitoring is activated, and a second restoring force (T2), which drives the robot from the position (x0) to the boundary (s), is specified by the control system (S70), provided that the robot only assumes this position in the restricted area after the boundary monitoring has been activated, wherein the first restoring force is at least temporarily smaller than the second restoring force, in particular equal to zero, and/or wherein a first restoring force (T),which drives the robot from a current position (x0) in a restricted area (S: x>s) back to a boundary (s) of this area, regardless of the distance of the position to the boundary, in particular equal to zero, is specified by the control system (S40), provided that the robot is moved by a distance towards the boundary or parallel to the boundary (d(x0-s)/dt<0), and another restoring force (T), which is greater than the first restoring force, is specified by the control system (S40), provided that the robot is moved by the same distance away from the boundary (d(x0-s)/dt>0), in particular if the robot is already in this position in the restricted area when the boundary monitoring is activated.