Method for constructing a map while performing work
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mapping techniques for autonomous robotic devices are computationally expensive, requiring substantial processing power and memory, and often necessitate additional equipment, which can limit their speed and performance, especially in environments with similar features and measurement noise.
Innovation Solution
A method for mapping a workspace using a robotic device that captures data on its position and movement, recognizes areas, generates a map, and updates it by combining depth measurements from overlapping fields of view, allowing for efficient mapping with minimal processing power and without the need for external beacons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Simultaneous Localization and Mapping (SLAM) with large amounts of captured points and features is used, then mapping accuracy is improved, but computational cost and processing power requirements increase substantially
Solution Approach 1:
The patent extracts only the essential features needed for mapping by using a camera to capture images and identify distinctive visual features (corners, edges, intersections) rather than processing all captured points. This selective extraction of meaningful features reduces computational load while maintaining mapping accuracy.
Solution Approach 2:
The mapping process is segmented into discrete steps: capturing images, identifying features, calculating camera position based on feature matching, and updating the map. This segmentation allows the system to process information in manageable chunks rather than handling all data simultaneously, reducing computational burden.
2Measurement precision
If probabilistic processing and particle filtering with large amounts of redundant data are used, then localization accuracy is improved, but memory requirements increase substantially
Solution Approach 1:
The patent uses a lightweight approach to data storage by maintaining only the current camera position estimate and the map data, rather than storing multiple sets of redundant particle filter data. The system processes images sequentially and discards processed data, using minimal memory while maintaining localization accuracy.
3Measurement precision
If a large complete state vector and error covariance matrix are used to store pose and feature positions, then estimation accuracy is improved, but processing speed decreases
Solution Approach 1:
The patent replaces the complex mathematical machinery of large state vectors and covariance matrices with a simpler geometric approach. By using feature matching and triangulation based on camera images, the system calculates position estimates directly without requiring extensive matrix operations, thereby improving processing speed while maintaining accuracy.
4Measurement precision
If additional equipment for projecting infrared light patterns is used, then positioning accuracy is improved, but device complexity increases
Solution Approach 1:
The patent makes the camera serve multiple functions: it captures images for visual SLAM, identifies features for positioning, and provides depth information through image analysis. This multi-functionality eliminates the need for separate infrared projection equipment while maintaining positioning capabilities through natural visual features in the environment.
5Manufacturing precision
If Laser Distance Sensors are used for high rate data collection, then mapping detail is improved, but sensitivity to lighting and transparent/reflective objects increases
Solution Approach 1:
The patent changes the sensing modality from active laser distance measurement to passive visual feature detection. By using camera images and analyzing visual features (edges, corners, textures) rather than measuring distance directly with lasers, the system achieves mapping detail while becoming insensitive to lighting conditions and transparent or reflective objects that interfere with laser sensors.
Data Source
AI summary
Provided is a method including: capturing, with at least one sensor of a robot, first data indicative of the position of the robot in relation to objects within the workspace and second data indicative of movement of the robot; recognizing, with a processor of the robot, a first area of the workspace based on observing at least one of: a first part of the first data and a first part of the second data; generating, with the processor of the robot, at least part of a map of the workspace based on at least one of: the first part of the first data and the first part of the second data; generating, with the processor of the robot, a first movement path covering at least part of the first recognized area; actuating, with the processor of the robot, the robot to move along the first movement path.