Robot Force-Control Stopping for Peg-in-Hole Load Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current robot control methods, particularly torque-based force control, face challenges in accurately stopping at target positions due to individual differences in robot manufacturing, leading to excessive loads on both the robot and workpieces, especially when switching from force control to position control, resulting in overshooting and difficulty in releasing workpieces during operations like peg-in-hole fitting.

Innovation Solution

A robot control system that reduces the load on both the robot and workpieces by maintaining torque-based force control during stopping, using a method where the force target value is gradually decreased and the position target value is fixed to the current position upon receiving a stop order, allowing for precise control and minimizing position deviations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If position control is used to stop the robot, then the robot can be stopped at the target position, but the robot overshoots the target position due to individual differences and high responsiveness, causing excessive load on the robot and workpiece

Engineering Contradiction:
Improvestopping position accuracyVSAvoidload on robot and workpiece
Core Design Contradiction:
Measurement precisionVSForce

Solution Approach 1:

The patent changes the control parameter from position-based to torque-based during the stopping process. By switching to torque control and gradually reducing the torque instruction value, the system exploits the physical property of torque to provide smooth deceleration without the overshoot problems inherent in position control, thereby stopping at the target position without excessive load.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent dynamically switches between control modes (position control to torque control) based on the stopping phase. During approach, position control provides accuracy; during stopping, torque control provides smooth deceleration. This dynamic adaptation resolves the contradiction between positioning accuracy and stopping smoothness.

Inventive Principle:
Principle #15Dynamics

2Productivity

If the robot decelerates and stops quickly using position control, then productivity is improved, but the robot cannot stop at the stop target position due to high responsiveness, causing overshoot and excessive load

Engineering Contradiction:
Improvestopping speedVSAvoidstopping position accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent changes the control parameter from position to torque during deceleration. By using torque instruction values that gradually decrease to zero, the system maintains high responsiveness for quick stopping while avoiding overshoot, achieving both high productivity and positioning accuracy.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies preliminary torque reduction before the robot reaches the target position. By gradually reducing the torque instruction value in advance, the system prepares for smooth deceleration, ensuring the robot can stop quickly yet accurately without overshooting.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If position control is used for stopping, then the robot can be controlled to stop, but the control complexity increases due to switching from force control to position control

Engineering Contradiction:
Improvecontrol simplicityVSAvoidcontrol system complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent dynamically switches control modes based on operational phase: position control during approach, torque control during stopping. This dynamic approach maintains ease of operation by using simple position control for most of the movement while employing torque control only when needed for smooth stopping, minimizing overall control complexity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent extracts the stopping function from position control and handles it separately using torque control. By separating the stopping operation from the main position control loop, the system maintains simple position control for positioning while adding a dedicated torque control mechanism only for the stopping phase.

Inventive Principle:
Principle #2Taking out (Extraction)

4Speed

If torque-based force control is used, then responsiveness is improved without being restricted by natural vibration frequency, but positioning accuracy deteriorates due to individual differences causing position deviation

Engineering Contradiction:
ImproveresponsivenessVSAvoidpositioning accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent segments the control process into two phases: approach phase using position control for accuracy, and stopping phase using torque control for responsiveness. By dividing the operation into segments with different control strategies, the system achieves both high positioning accuracy during approach and high responsiveness during stopping.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different control qualities to different phases of the motion: position control with high precision requirements for the approach phase, and torque control with high responsiveness requirements for the stopping phase. This local differentiation of control quality resolves the contradiction between overall positioning accuracy and stopping responsiveness.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3147087B1Robot apparatus, robot controlling method, program, recording medium, and assembly manufacturing method
Publication Date: 2024.01.17 CANON KK
  • EP3147087B1 patent drawingFigure 1
  • EP3147087B1 patent drawingFigure 2
  • EP3147087B1 patent drawingFigure 3

AI summary

There is provided a robot apparatus which is characterized by comprising: a robot comprising a plurality of motors for driving respective joints and a sensor for obtaining force acting on a hand tip; and a controlling unit for obtaining a torque instruction value for each of the plurality of motors such that a force deviation between the force acting on the hand tip and a force target value becomes small, controlling driving of each of the plurality of motors based on the torque instruction value, and performing a stopping process of decreasing the force target value when a stop order for stopping the robot is received.