Environment Map Merging Path Planning for Accurate SLAM

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing simultaneous localization and mapping (SLAM) techniques face challenges in accurately merging large environment maps, particularly when the moving range of a movable apparatus is wide, as they do not prioritize the accuracy of merging environment maps.

Innovation Solution

An information processing device is configured to calculate the position and orientation of a movable apparatus, create environment maps, determine a second environment map for merging, and determine a traveling path to ensure high accuracy in merging the maps by focusing on common image features and their distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If environment maps are merged based on relative position and orientation of robots, then the efficiency of creating large environment maps is improved, but the accuracy of merging the environment maps deteriorates

Engineering Contradiction:

Improveefficiency of creating large environment mapsVSAvoidaccuracy of merging environment maps

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions by determining a traveling path in advance that ensures sufficient distribution of similar image features between environment maps before merging occurs. This pre-planning of the traversal path guarantees that when maps are merged, there are enough matching features to achieve high accuracy without requiring multiple robots to simultaneously cover the same areas.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback by evaluating the distribution and number of similar image features between environment maps to determine whether to create a new map or merge existing maps. The merging decision is based on feedback from feature comparison, ensuring that maps are only merged when sufficient similar features exist to maintain high merging accuracy.

Inventive Principle:

Principle #23Feedback

2Area of stationary object

If the moving range of the movable apparatus is widened, then the coverage of the environment map is improved, but the time required to create the environment map increases

Engineering Contradiction:

Improvecoverage of environment mapVSAvoidtime required to create environment map

Core Design Contradiction:

Area of stationary objectVSLoss of time

Solution Approach 1:

The system merges environment maps created by the same or different movable apparatuses to efficiently expand coverage. By combining multiple partial environment maps through feature-based merging, the system achieves wide-area coverage without requiring a single robot to traverse the entire area, thus reducing the time required to create comprehensive environment maps.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The system determines optimal traveling paths in advance that maximize the distribution of similar image features. This preliminary path planning ensures that when maps need to be merged, sufficient overlapping features exist, enabling efficient map consolidation that expands coverage without requiring exhaustive exploration of all areas.

Inventive Principle:

Principle #10Preliminary action

3Device complexity

If environment maps are merged without considering feature distribution, then the complexity of the merging process is reduced, but the accuracy of the merged map deteriorates

Engineering Contradiction:

Improvecomplexity of merging processVSAvoidaccuracy of merged map

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms that evaluate the distribution and quantity of similar image features between environment maps before and during merging. This feedback-driven approach automatically adjusts the merging process to ensure sufficient feature overlap, maintaining high accuracy without requiring complex manual intervention or overly complicated merging algorithms.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system changes parameters related to feature matching thresholds and merging criteria based on the distribution characteristics of image features. By dynamically adjusting these parameters, the system maintains simple merging processes while ensuring that accuracy requirements are met through adaptive parameter selection rather than complex procedural steps.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20240329655A1Information processing device, movable apparatus, information processing method, and storage medium

Publication Date: 2024.10.03 CANON KK

AI summary

An information processing device includes a position and orientation measurement unit to calculate a position and an orientation of a movable apparatus based on measurement information of a first sensor that measures an environment in surroundings of the movable apparatus. A first environment map is created in a vicinity of the movable apparatus based on the measurement information of the first sensor and the position and the orientation, and a second environment map to be merged with the first environment maps determined. Information regarding the second environment map is acquired and a traveling path of the movable apparatus is determined to merge the first environment map and the second environment map based on the information regarding the second environment map.