Robot Self-Position Mapping Across Slopes Using Divided 2D Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-position identification methods for robots fail to accurately determine position in environments with changing heights and postures, such as slopes, where conventional two-dimensional maps fail to account for changes in posture and height, causing misalignment in sensor data alignment.
Innovation Solution
A method for generating and connecting divided maps using odometry and markers to create integrated maps, which are divided into stable sections, and generate connection information between these maps based on sensor information, allowing for seamless movement in three-dimensional environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a two-dimensional advance map is used for self-position identification, then the system complexity is low and implementation is simple, but the position identification accuracy deteriorates in environments with changing heights and postures such as slopes
Solution Approach 1:
The patent divides the three-dimensional space into multiple two-dimensional maps, each corresponding to a specific height level or stable section. This segmentation allows the system to maintain simple two-dimensional map structures while accurately representing three-dimensional environments with varying heights and postures, resolving the contradiction between system simplicity and positioning accuracy.
Solution Approach 2:
The patent transitions from traditional two-dimensional maps to a multi-layered two-dimensional map system that incorporates vertical dimension information. By organizing multiple 2D maps at different height levels with defined positional relationships, the system captures three-dimensional spatial information while maintaining the computational advantages of two-dimensional processing.
2Adaptability or versatility
If the robot moves on a slope causing height and posture changes, then the robot can navigate various terrains, but the sensor observation point changes causing collation failure between advance map and sensor information
Solution Approach 1:
The patent segments the navigation space into multiple stable sections, each with its own two-dimensional map. When the robot moves between sections with different heights or postures, the system switches between corresponding maps, ensuring that sensor observations always match the current stable section's map, thereby maintaining collation reliability across varied terrains.
Solution Approach 2:
The patent implements a dynamic map selection mechanism that adapts to the robot's current position and posture. As the robot moves through different stable sections, the system dynamically switches between corresponding two-dimensional maps, ensuring that the active map always reflects the current observation conditions, thus maintaining reliable sensor collation.
3Measurement precision
If divided maps are generated for each stable section to improve position identification accuracy, then the position identification accuracy improves in three-dimensional environments, but the device complexity increases
Solution Approach 1:
The patent segments the three-dimensional environment into multiple stable sections, each represented by a two-dimensional map. This segmentation approach maintains manageable map complexity while improving position identification accuracy, as each individual map remains simple but collectively they cover the full three-dimensional workspace.
Solution Approach 2:
The patent organizes multiple two-dimensional maps in a multi-layered structure that represents three-dimensional space. By maintaining the simplicity of two-dimensional maps while adding vertical layering, the system improves position identification accuracy without proportionally increasing the complexity of individual map structures.
Data Source
AI summary
An information processing device includes a self-position identification unit. The self-position identification unit generates a divided map for each stable section in which the height or posture of the traveling robot is stable. The self-position identification unit generates connection information indicating a positional relationship between the divided maps on the basis of sensor information obtained during movement between the divided maps.