Robot Grid Map Noise Reduction With Contour Simplification

Solution Overview

Problem

Existing grid maps generated by cleaning robots suffer from noise due to sensor errors and environmental factors, leading to poor performance and potential operation errors during cleaning and obstacle avoidance.

Innovation Solution

A map noise reduction apparatus and method that includes a binarization module, inner noise reducing module, outer noise reducing module, and simplification module to refine the grid map by updating pixel values based on surrounding pixels, shortest distances to obstacles, and applying the Douglas-Peucker method for contour simplification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a grid map is generated using sensor data from a cleaning robot, then the robot can store and use the map for navigation and cleaning tasks, but the map contains noise due to sensor errors and environmental factors such as color/material variations of obstacles and slippery floors

Engineering Contradiction:

Improvemap accuracyVSAvoidsensor measurement precision

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent combines multiple processing techniques (binarization, inner noise reduction, outer noise reduction, and contour simplification) into a unified map processing system. The binarization module converts grayscale map data to binary format, the inner noise reducing module processes pixels based on surrounding pixel values, the outer noise reducing module handles boundary noise, and the simplification module refines contours - together these merged modules comprehensively reduce different types of map noise to improve overall map accuracy

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent applies noise reduction processing to the grid map before it is used for navigation and cleaning tasks. By pre-processing the map data through binarization and multiple noise reduction stages, the system eliminates sensor errors and environmental noise in advance, ensuring that the cleaned map data is ready for reliable robot operation without introducing noise during actual cleaning tasks

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If the grid map is displayed to the user, then the user can recognize the cleaning robot's understanding of the house structure, but noise in the map gives the impression of poor performance and may cause operation errors

Engineering Contradiction:

Improveuser recognition of cleaning performanceVSAvoidmap noise

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent combines multiple processing techniques (binarization, inner noise reduction, outer noise reduction, and contour simplification) into a unified map processing system. The binarization module converts grayscale map data to binary format, the inner noise reducing module processes pixels based on surrounding pixel values, the outer noise reducing module handles boundary noise, and the simplification module refines contours - together these merged modules comprehensively reduce different types of map noise to improve overall map accuracy

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent applies noise reduction processing to the grid map before it is used for navigation and cleaning tasks. By pre-processing the map data through binarization and multiple noise reduction stages, the system eliminates sensor errors and environmental noise in advance, ensuring that the cleaned map data is ready for reliable robot operation without introducing noise during actual cleaning tasks

Inventive Principle:

Principle #10Preliminary action

3Productivity

If the robot uses the noisy grid map for navigation and cleaning, then the robot can perform its tasks, but operation errors such as incorrect cleaning patterns and obstacle avoidance navigation may occur

Engineering Contradiction:

Improvecleaning efficiencyVSAvoidoperation accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent combines multiple processing techniques (binarization, inner noise reduction, outer noise reduction, and contour simplification) into a unified map processing system. The binarization module converts grayscale map data to binary format, the inner noise reducing module processes pixels based on surrounding pixel values, the outer noise reducing module handles boundary noise, and the simplification module refines contours - together these merged modules comprehensively reduce different types of map noise to improve overall map accuracy

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent applies noise reduction processing to the grid map before it is used for navigation and cleaning tasks. By pre-processing the map data through binarization and multiple noise reduction stages, the system eliminates sensor errors and environmental noise in advance, ensuring that the cleaned map data is ready for reliable robot operation without introducing noise during actual cleaning tasks

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP4053505A1Noise reduction apparatus and method for map of robot

Publication Date: 2022.09.07 MIELE & CO KG

Figure 1(a)~1(b)
Figure 2
Figure 3

AI summary

A map noise reduction apparatus and method of a robot according to the exemplary embodiment of the present disclosure reduce a noise such as an outer noise and an inner nose which may be caused due to a sensor characteristic and an environmental factor while generating a grid map and simplify a contour of the grid map to provide a neater and clear grid map.