SLAM Map Correction Using CAD Visual Index Registration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The existing methods for generating maps for movable apparatuses, such as AGVs, using SLAM face challenges in accurately matching CAD data with SLAM maps due to accumulated errors, leading to positional deviations and orientation measurement inaccuracies.

Innovation Solution

An information processing device that includes a map acquisition unit, detection unit, registration unit, association unit, and correction unit to detect visual indexes from CAD data, register and associate them with SLAM map features, and correct the SLAM map coordinates to improve alignment and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If SLAM method is used for map generation, then autonomous travel capability is achieved, but accumulated error causes positional deviation from CAD data

Engineering Contradiction:

Improveautonomous travel capabilityVSAvoidposition accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The system performs feedback by comparing SLAM-generated map features with CAD data features, detecting deviations, and correcting the SLAM map coordinates based on the association results. This closed-loop feedback mechanism eliminates accumulated errors while preserving autonomous travel capability.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system creates a corrected copy of the SLAM map by generating new coordinate information based on the association between SLAM features and CAD data features. This copying approach preserves the original autonomous navigation functionality while eliminating positional deviations.

Inventive Principle:

Principle #26Copying

2Manufacturing precision

If scale adjustment is performed using depth sensor data, then partial matching improvement is achieved, but entire map alignment error remains

Engineering Contradiction:

Improvelocal map accuracyVSAvoidglobal position accuracy

Core Design Contradiction:

Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The system segments the map correction problem into multiple feature-level associations between SLAM map elements and CAD data elements. By correcting individual feature coordinates and propagating corrections throughout the map, it achieves both local and global alignment accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system transitions from 2D scale adjustment to 3D coordinate correction by utilizing depth sensor data and performing full 3D association between SLAM map features and CAD data features, thereby achieving comprehensive map alignment.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4266005A1Information processing device, information processing system, method of controlling information processing device, and storage medium

Publication Date: 2023.10.25 CANON KK

Figure 1
Figure 2
Figure 3

AI summary

An information processing device that improves accuracy of a map used to measure a position and an orientation of a movable apparatus includes: a map acquisition unit configured to acquire a map for measuring a position and an orientation of a movable apparatus; a detection unit configured to detect a visual index from a drawing information which corresponds to a real space that the movable apparatus moves or an object in the real space; a registration unit configured to perform registration of a coordinate system between the map and the drawing information; an association unit configured to perform association between the visual index included in the drawing information after the registration and a constituent element included in the map; and a correction unit configured to correct the map based on a result of the association.