Multi-Axis Robot Torque Normalization for Failure Diagnosis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing failure diagnostic methods for multi-axis robots require setting different thresholds for each robot, making it challenging to perform accurate failure diagnosis regardless of the robot in operation.

Innovation Solution

A failure diagnostic device and method that standardize disturbance torque values by calculating representative values and amounts of change, allowing for accurate failure diagnosis using a fixed threshold across multiple robots and operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If different thresholds are set for each robot, then accurate failure diagnosis can be achieved for individual robots, but the system complexity and threshold management become cumbersome

Engineering Contradiction:
Improvefailure diagnosis accuracyVSAvoidthreshold management complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the disturbance torque values through mathematical operations (subtracting representative values and dividing by amounts of change) to create normalized parameters that can be compared across different robots using a single threshold, thereby maintaining diagnostic accuracy while simplifying threshold management

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent creates a universal threshold that can be applied to all robots regardless of their individual characteristics by standardizing the disturbance torque measurements through calculation of representative values and amounts of change, making the diagnostic system universally applicable across multiple robots

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Productivity

If disturbance torque values are used directly for comparison, then the measurement process is simple, but the diagnosis accuracy varies across different robots due to individual differences

Engineering Contradiction:
Improvediagnosis efficiencyVSAvoiddiagnosis accuracy consistency
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies parameter transformation by calculating normalized disturbance torque values using representative values and amounts of change, converting raw torque data into standardized metrics that enable consistent cross-robot comparison while maintaining diagnostic efficiency

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces intermediary calculation steps (representative value calculation and amount of change computation) that act as mediators between raw disturbance torque measurements and final diagnosis, standardizing the data before comparison to ensure consistent accuracy across different robots

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3778157B1Failure diagnostic device and failure diagnostic method
Publication Date: 2022.08.17 NISSAN MOTOR CO LTD
  • EP3778157B1 patent drawingFigure 1
  • EP3778157B1 patent drawingFigure 2~3
  • EP3778157B1 patent drawingFigure 4(a)~4(b)

AI summary

A failure diagnostic device detects the movement position of each of joint shafts included in a multi-axis robot (S03), and detects a disturbance torque (Tq) applied to the joint shaft (S01). The failure diagnostic device determines whether or not the multi-axis robot is executing a predefined routine operation, from the detected movement position, and calculates disturbance-torque reference values from the disturbance torque detected during execution of the predefined routine operation (S07). The failure diagnostic device corrects the disturbance torque by using the disturbance-torque reference values (S09), and performs a failure diagnosis on the multi-axis robot (1) by comparing the corrected disturbance torque (Tq') and a threshold (α) (S11 to S15).