Vision-Based Lawn Mower Navigation Using 3D Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous lawn mowers rely on boundary wires for navigation, which are costly, cumbersome, and limited by the mobile nature of the mowers, restricting available computing resources for more sophisticated navigation techniques.
Innovation Solution
The use of feature extraction and object recognition techniques to generate a three-dimensional point cloud (3DPC) for autonomous machine navigation, allowing the mower to define work region boundaries without wires and navigate efficiently within the region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If boundary wires are used for navigation, then the mower can stay within predefined boundaries, but the system becomes costly and cumbersome to install and maintain
Solution Approach 1:
The patent extracts the boundary definition function from the physical wire infrastructure and relocates it to the autonomous mower's onboard vision system. The mower captures images, detects natural or artificial features in the environment, and uses these features to define and maintain work region boundaries without requiring external wire installations.
Solution Approach 2:
The patent replaces the mechanical boundary wire system with an optical vision-based system. Instead of using physical wires that detect mower position through electrical contact, the system uses cameras and image processing algorithms to detect visual features and calculate mower position and orientation relative to work region boundaries.
2Ease of operation
If boundary wires are used for navigation, then navigation is simplified, but moving boundaries becomes difficult and time-consuming
Solution Approach 1:
The patent implements a dynamic boundary definition system where work region boundaries are not fixed physical installations but are dynamically determined by the mower's vision system. Boundaries can be easily redefined by capturing new images of the work region and detecting different visual features, allowing rapid adaptation to changing maintenance needs without physical reinstallation.
Solution Approach 2:
The system allows boundary parameters to be changed by detecting different visual features in the environment. By changing which features are detected and how they are interpreted, the effective boundary configuration can be modified without changing the physical environment, enabling flexible adaptation to different work region requirements.
3Extent of automation
If more sophisticated navigation techniques are implemented, then navigation capability is improved, but computing resources are exceeded due to mobile platform limitations
Solution Approach 1:
The patent applies partial action by implementing a two-stage processing approach: comprehensive image processing and feature extraction are performed offline when computing resources are abundant, while only essential navigation functions are executed in real-time on the mobile platform. This reduces online computational requirements while maintaining sophisticated navigation capabilities.
Solution Approach 2:
The system performs preliminary processing of navigation data during offline periods when the mower is stationary or charging. Complex image analysis, feature matching, and path planning computations are completed beforehand, so that during active mowing operations, the mower only needs to execute pre-computed navigation commands, significantly reducing real-time computing resource consumption.
Data Source
AI summary
Autonomous machine navigation techniques may determine vision-based pose data based on feature data and object recognition data extracted from images. The vision-based pose data may be used to generate a three-dimensional point cloud that represents at least a work region. The vision-based pose data may be used to determine an operational vision-based pose relative to the three-dimensional point cloud.


