Support Structure Forward-Plane Detection for Aisle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous or semi-autonomous transporters in facilities face localization errors that lead to suboptimal navigation, particularly when detecting the forward plane of support structures amidst varying item arrangements.
Innovation Solution
The method involves capturing three-dimensional point cloud data using a sensor, generating a two-dimensional projection in a facility coordinate system, retrieving the expected location of the support structure's forward plane from a stored map, selecting regions that satisfy a positional criterion, determining the actual location of the forward plane, and providing this information to a navigational controller.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional localization methods are used by autonomous transporters, then the navigation system is simpler to implement, but localization errors occur leading to suboptimal navigation
Solution Approach 1:
The patent introduces support structures as intermediary objects between the autonomous transporter and the facility environment. These structures serve as mediators for localization, providing detectable features (forward planes) that enable accurate positioning without requiring complex sensor systems on the transporter itself. The support structures act as a bridge that translates environmental information into navigational data.
Solution Approach 2:
The system creates a digital representation (2D projection) of the physical support structure's forward plane from 3D point cloud data. This copy is then used for localization purposes, allowing the transporter to determine its position by comparing the detected forward plane location with the stored map data, thereby achieving accurate localization without direct complex sensing.
2Measurement precision
If the transporter detects all features in the environment, then localization accuracy improves, but the detection system becomes more complex and computationally intensive
Solution Approach 1:
The patent extracts only the critical feature (the forward plane of support structures) from the complete environmental scene. By using 3D point cloud data to identify and isolate the forward plane, the system eliminates the need to process all environmental features. This extraction approach maintains high localization accuracy while significantly reducing detection complexity compared to analyzing all visible objects and features.
Solution Approach 2:
The system applies different processing qualities to different parts of the environment. Instead of uniformly processing all detected features with high complexity, the patent focuses computational resources specifically on identifying and locating the forward plane of support structures. This localized approach to feature detection maintains accuracy for the critical localization feature while reducing overall system complexity.
Data Source
AI summary
A method includes: capturing, via a navigational sensor of a mobile automation apparatus, three-dimensional point cloud data depicting a portion of an aisle containing a support structure, the support structure having a forward plane facing into the aisle; generating, from the point cloud data, a two-dimensional projection in a facility coordinate system; retrieving, from a stored map, an expected location of the forward plane of the support structure in the facility coordinate system; selecting, from the projection, a subset of regions satisfying a positional criterion relative to the location of the forward plane; determining, based on the selected subset of regions from the projection, an actual location of the forward plane of the support structure in the facility coordinate system; and providing the actual location of the forward plane to a navigational controller of the mobile automation apparatus.