Robot Localization Correction for Odometry Drift and Posture Reset

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Wheeled robots face positioning errors due to wheel slippage, unequal wheel diameters, and inaccurate wheel base distances, leading to deviations from planned paths, which existing odometry systems fail to accurately correct.

Innovation Solution

A localization correction method using a processor-based system that acquires initial and post-motion position information, establishes a transformation model, calculates compensation values, and generates reset commands to adjust the robot's localization, incorporating TOF or structured light technology and lidar for accurate positioning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If odometry system is used for position estimation, then the robot can determine its position based on wheel revolution, but positioning accuracy deteriorates due to wheel slippage, unequal wheel diameter, and inaccurate wheel base distance

Engineering Contradiction:
Improveposition estimation capabilityVSAvoidpositioning accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary correction system that includes a processor-based system with camera and sensor arrays. This intermediary system captures images and sensor data, processes them to determine actual position and orientation, and generates correction values that compensate for odometry errors caused by wheel slippage, unequal wheel diameters, and inaccurate wheel base distances.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the purely mechanical odometry system (which relies on wheel encoders and mechanical measurements) with an optical and sensor-based system. The processor-based system uses camera images and sensor data to optically determine the robot's actual position and orientation, substituting the mechanical wheel-revolution-to-linear-translation method with a vision-based measurement approach that is not affected by wheel slippage or mechanical imperfections.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If wheel encoder odometry is used, then the robot can transform wheel revolution to linear translation, but navigation accuracy deteriorates due to accumulated errors from wheel slippage and mechanical variations

Engineering Contradiction:
Improvenavigation capabilityVSAvoidnavigation accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where the processor-based system continuously monitors the robot's actual position and orientation by processing camera images and sensor data. The system compares the actual position with the planned course, calculates deviation, and generates correction values that are fed back to the navigation system to compensate for accumulated errors from wheel slippage and mechanical variations.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent replaces the mechanical wheel encoder odometry system with an optical feedback system. Instead of relying on mechanical wheel revolutions to estimate position, the system uses camera images and sensor data to directly measure the robot's actual position and orientation, providing accurate feedback that is not affected by mechanical imperfections or wheel slippage.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If traditional odometry correction methods are used, then some positioning errors can be compensated, but accumulated errors cannot be effectively eliminated and automatic posture correction is not achieved

Engineering Contradiction:
Improvepositioning error compensationVSAvoidaccumulated error elimination
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent replaces traditional mechanical odometry correction methods with a processor-based optical system. The system uses camera images and sensor data to directly measure the robot's actual position and orientation, then uses image processing algorithms to calculate precise correction values that effectively eliminate accumulated errors, enabling reliable automatic posture correction.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a processor-based system as an intermediary between the robot's motion and its position estimation. This intermediary system processes camera images and sensor data to determine actual position and orientation, generating correction values that reliably eliminate accumulated errors and enable automatic posture correction, rather than relying on imperfect mechanical correction methods.

Inventive Principle:
Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This method effectively eliminates accumulated positioning errors and enables automatic posture correction of the robot, improving its navigation accuracy and adaptability.

Implementation Method 1

incorporating TOF or structured light technology and lidar for accurate positioning

Methodology Applied
Scientific EffectTime of Flight: Time of Flight

Implementation Method 2

incorporating TOF or structured light technology and lidar for accurate positioning

Methodology Applied
Scientific EffectLIDAR: LIDAR

Data Source

PatentUS11260528B2Localization correction method for robot, computer readable storage medium, and robot
Publication Date: 2022.03.01 UBTECH ROBOTICS CORP LTD
  • US11260528B2 patent drawing
  • US11260528B2 patent drawing
  • US11260528B2 patent drawing

AI summary

An localization correction method for a robot, comprising acquiring first position information of the robot in a first coordinate system; acquiring second position information of the robot in a second coordinate system after the robot executes a motion command; establishing a transformation model between the first position information and the second position information based on the first coordinate system and the second coordinate system; calculating a compensation value according to the transformation model; and generating a reset command according to the compensation value, and adjusting the localization of the robot according to the reset command.