Visual Work-Area Mapping From Ortho-Projected Mobile Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the boundaries of a work area for mobile devices like robotic lawnmowers are imprecise and require significant computing resources, especially when using SLAM, and often necessitate additional verification steps that slow down the work function.
Innovation Solution
A two-step method involving external manual movement to define a preliminary boundary, followed by automated movement to confirm and adjust the boundary using visual SLAM and orthographic projection, combined with computer vision algorithms like semantic segmentation, to create a precise visual map of the work area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If environmental information is obtained during externally controlled movement to determine a boundary, then the boundary determination is achieved, but the time required for boundary determination increases
Solution Approach 1:
The patent applies preliminary action by performing boundary determination during teach-in movement before actual work operations begin. The mobile device captures environmental information and determines boundaries in advance, storing this data for later use during automated work functions, thus avoiding time loss during actual operations.
Solution Approach 2:
The system uses the mobile device's own movement and onboard sensors to capture environmental information and determine boundaries autonomously. The device serves itself by using its positioning system, camera, and processing unit to map the environment without requiring external surveying equipment or manual measurement, thereby reducing time while maintaining accuracy.
2Adaptability or versatility
If environmental information is captured during movement to create a visual map, then navigation capability is improved, but the complexity of data processing increases
Solution Approach 1:
The patent introduces an intermediary processing layer where the computing unit acts as a mediator between the mobile device's sensors and the navigation system. Environmental information captured by cameras and sensors is processed by the computing unit to generate simplified boundary data and visual maps, which then guide the automated navigation, reducing the complexity burden on the core navigation system.
Solution Approach 2:
The system creates a simplified copy or representation of the environment through visual maps and boundary data structures. Instead of processing all raw sensor data during navigation, the system uses pre-processed map copies that contain essential geometric information about boundaries and work areas, significantly reducing real-time data processing complexity while maintaining navigation accuracy.
Data Source
Figure 1a~1b
Figure 2a~2b
Figure 2c
AI summary
The invention relates to a method for determining a visual map (420) of at least a part of an environment in which a mobile device, in particular a vehicle or robot that moves at least partially automatically, especially a robotic lawnmower, moves or is intended to move, comprising: providing images of the environment that are or have been obtained when the mobile device moves in the environment, determining an ortho-projected view of the images of the environment, determining the visual map (420) of the environment based on the ortho-projected view of the images of the environment, and providing the visual map of the environment.