Facility Occupancy Mapping Using CAD and Sensor-Based Remapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SLAM techniques require traversing an entire facility to generate a map, which is cumbersome for large facilities, and updating maps to reflect changes requires new satellite images or schematics, making it inefficient for self-driving vehicles to navigate and adapt to changes within the environment.
Innovation Solution
The method involves using a CAD file to generate an occupancy-map image and keyframe graph, which is updated by sensors detecting new features, allowing self-driving vehicles to map and remap facilities electronically without full traversal, using a combination of graphical and non-graphical objects, and LiDAR or vision systems for navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional SLAM techniques are used to generate a facility map, then the map can be created with detailed spatial information, but the entire facility must be traversed which is time-consuming and cumbersome for large facilities
Solution Approach 1:
The system performs preliminary actions by obtaining a CAD file that represents the facility layout before traversal begins. This pre-existing graphical model provides the spatial framework, allowing the system to skip the time-consuming process of mapping the entire facility from scratch while still achieving accurate localization through sensor-based feature detection and comparison with the pre-obtained CAD data
2Adaptability or versatility
If satellite images or new schematics are used to update maps to reflect environmental changes, then the map remains current, but this requires obtaining entirely new data sources which is inefficient
Solution Approach 1:
The system uses sensor data collected during normal operation as feedback to detect changes in the facility environment. By comparing newly detected features against the existing occupancy-map image, the system automatically identifies modifications such as new obstacles or layout changes and updates the map accordingly, eliminating the need for external data sources like new satellite images or schematics
Solution Approach 2:
The mapping system serves itself by using its own sensor measurements to detect environmental changes and automatically update the facility map. The system leverages data collected during its normal navigation tasks to maintain an current map without requiring separate mapping missions or external data sources, making the update process efficient and self-sufficient
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient electronic mapping and remapping of facilities, allowing self-driving vehicles to navigate and adapt to changes without the need for full traversal, improving navigation efficiency and adaptability.
Implementation Method 1
using a combination of graphical and non-graphical objects, and LiDAR or vision systems for navigation
Data Source
AI summary
Systems and methods for electronically mapping a facility are presented. The method comprises obtaining a CAD file that includes graphical representations of a facility. An occupancy-map image is generated based on the CAD file. A sensor, such as a sensor on a self-driving vehicle, is used to detect a sensed feature within the facility. Based on the sensed feature, the occupancy-map image can be updated, since the sensed feature was not one of the known features in the CAD file prior to the sensed feature being detected by the sensor.