Robotic Object Update Mechanism for Misdetection Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Robotic systems face challenges in adapting to unexpected conditions, such as incorrect object identification, which can lead to errors in motion planning and execution, resulting in collisions or grip failures during object transfer.

Innovation Solution

The robotic system employs a mechanism to register expected objects based on physical characteristics and uses sensors to verify the identity of objects by comparing actual measurements with expected data. If discrepancies are found, the system adjusts its motion plan and updates its registration records accordingly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If the robotic system uses object identification to generate motion plans, then the system can automate object transfer tasks, but the system may incorrectly identify objects with similar physical characteristics, leading to detection errors

Engineering Contradiction:
Improveautomation of object transferVSAvoidobject identification accuracy
Core Design Contradiction:
Extent of automationVSMeasurement precision

Solution Approach 1:

The system implements feedback by comparing actual sensor measurements (weight, dimensions) against expected values from registration records. When discrepancies are detected, the system triggers correction procedures including pausing the motion plan, notifying operators, or automatically adjusting parameters to resolve the identification error before execution

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary verification of object identity by comparing multiple physical characteristics (weight, dimensions, appearance) against pre-stored registration records before generating the motion plan. This preliminary check prevents incorrect identification from propagating through the system

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If the system verifies object identity by comparing measurements with registration records, then detection accuracy improves, but the system complexity increases due to additional verification steps

Engineering Contradiction:
Improveobject detection verificationVSAvoidverification mechanism complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses multi-functional sensors that can measure multiple physical characteristics (weight, dimensions, appearance) simultaneously. A single sensor array performs verification across multiple parameters, reducing the need for separate verification devices and minimizing system complexity while maintaining high detection accuracy

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

3Reliability

If the robotic system corrects detection errors by pausing and notifying operators, then object transfer reliability improves, but productivity decreases due to manual intervention requirements

Engineering Contradiction:
Improveobject transfer reliabilityVSAvoidobject transfer throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system dynamically adjusts its response to detection errors based on the severity and type of discrepancy. For minor discrepancies, the system automatically corrects parameters and continues operation. For major discrepancies, the system pauses and notifies operators. This dynamic response strategy maintains high reliability while minimizing productivity loss from manual interventions

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12202145B2Robotic system with object update mechanism and methods for operating the same
Publication Date: 2025.01.21 MUJIN INC
  • US12202145B2 patent drawing
  • US12202145B2 patent drawing
  • US12202145B2 patent drawing

AI summary

A system and method for determining a misdetection of an object and subsequent response is disclosed. A robotic system may use a motion plan, which is derived based on an initial detection result of a package, to transfer the package from a start location to a task location. During implementation of the motion plan, the robotic system may obtain additional sensor data, which can be used to deviate from the initial motion plan and implement a replacement motion plan to transfer the package to the task location.