Discovering and plotting the boundary of an enclosure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mapping methods for autonomous robotic devices, such as SLAM, require substantial memory, processing power, and additional equipment, and struggle with incomplete depth maps due to gaps in data collection, especially in low-light environments and areas with transparent or reflective surfaces.
Innovation Solution
A process using one or more cameras and sensors to discover and plot the boundary of an enclosure by combining readings within successively overlapping fields of view, allowing for the identification and closure of gaps in the map without the need for additional equipment or high processing power, using techniques like stereoscopic imaging and structured light projection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SLAM methods use laser distance sensors with high data collection rates, then mapping precision is improved, but device complexity and processing power requirements increase substantially
Solution Approach 1:
The patent extracts and removes the laser distance sensor component from the mapping system, replacing it with a camera-based visual mapping approach. This eliminates the need for high-rate distance sensing hardware while maintaining mapping functionality through image processing and feature detection algorithms.
Solution Approach 2:
The patent substitutes the mechanical/optical laser distance sensing system with a camera-based visual system. Instead of using laser ranging to measure distances, the system uses image capture, feature extraction, and computer vision algorithms to achieve mapping, thereby reducing hardware complexity.
2Measurement precision
If SLAM methods use image processing techniques with multiple processing stages, then measurement precision is improved, but use of energy and processing power increase
Solution Approach 1:
The patent removes complex multi-stage image processing pipelines including probabilistic processing and particle filtering. Instead, it employs simplified feature detection algorithms that directly identify corners, lines, and geometric structures from images without requiring extensive computational stages.
Solution Approach 2:
The patent uses lightweight, computationally inexpensive image processing algorithms that can be executed with minimal processing power. These simplified algorithms trade some computational robustness for significantly reduced energy consumption and processing requirements.
3Measurement precision
If SLAM methods require additional equipment to project infrared patterns, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent removes the infrared pattern projection equipment entirely from the system. Instead of using structured light or infrared patterns to enhance feature detectability, the system relies on natural visual features in the environment captured by the camera, eliminating the need for additional active illumination or projection devices.
4Area of stationary object
If depth maps are constructed using distance sensors, then area coverage is improved, but gaps in data collection occur in low-light and reflective areas
Solution Approach 1:
The patent replaces distance sensor-based depth mapping with camera-based visual mapping. The camera system captures visual information that is less sensitive to lighting conditions and reflective surfaces, using feature detection and triangulation methods to construct accurate spatial maps without the gaps that plague sensor-based approaches in challenging environments.
Data Source
AI summary
Provided is a process that includes: obtaining a first version of a map of a workspace; selecting a first undiscovered area of the workspace; in response to selecting the first undiscovered area, causing the robot to move to a position and orientation to sense data in at least part of the first undiscovered area; and obtaining an updated version of the map mapping a larger area of the workspace than the first version.