Autonomous Machine Navigation With Vision-Based Boundary Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous lawn mowers face limitations in navigation due to limited computing resources and the impracticality of using boundary wires, which are costly, cumbersome, and difficult to reconfigure.
Innovation Solution
A method for autonomous machine navigation using a combination of non-vision-based sensors and a vision system to determine and correct the pose of the machine within a work region, employing training modes to generate a 3D point cloud and update positions based on image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If boundary wires are used to define work region boundaries, then navigation reliability is improved, but device complexity and installation cost increase
Solution Approach 1:
The patent extracts the boundary definition function from physical wires and implements it through vision-based detection of natural or artificial features in the environment. The system removes the need for boundary wires by using cameras and image processing to identify and track boundary features such as edges, lines, or markers in the visual field, thereby reducing device complexity while maintaining navigation reliability.
Solution Approach 2:
The patent replaces the mechanical boundary wire system with a vision-based detection system. Instead of using physical wires that require installation and maintenance, the system uses optical sensors (cameras) and computer vision algorithms to detect and interpret visual features that define work region boundaries, substituting a mechanical system with an optical and computational one.
2Measurement precision
If more sophisticated navigation systems are implemented, then navigation precision is improved, but computing resource requirements increase
Solution Approach 1:
The patent implements partial action by using a hybrid navigation approach that combines simple non-vision-based pose estimation (sufficient for basic navigation) with selective vision-based corrections (applied only when needed to improve precision). This avoids the excessive computational burden of using full vision-based navigation continuously, thereby reducing computing resource requirements while maintaining navigation precision.
Solution Approach 2:
The patent introduces an intermediary approach by using vision-based pose data not as the primary navigation source but as a corrective element that refines the pose estimates from non-vision sensors. This intermediary role of vision data allows the system to achieve higher navigation precision without requiring the computational resources of a fully vision-based system, as vision is used selectively to correct rather than continuously compute.
3Measurement precision
If vision-based pose correction is continuously applied, then navigation precision is improved, but energy consumption increases
Solution Approach 1:
The patent implements periodic action by applying vision-based pose correction at specific intervals or under specific conditions rather than continuously. The system periodically updates the pose estimate using vision data when the machine is in suitable conditions (e.g., when visual features are detectable), and relies on non-vision sensors between updates. This periodic application of vision-based correction maintains navigation precision while significantly reducing energy consumption compared to continuous vision processing.
Data Source
AI summary
Autonomous machine navigation techniques may generate a three-dimensional point cloud that represents at least a work region based on feature data and matching data. Pose data associated with points of the three-dimensional point cloud may be generated that represents poses of an autonomous machine. A boundary may be determined using the pose data for subsequent navigation of the autonomous machine in the work region. Non-vision-based sensor data may be used to determine a pose. The pose may be updated based on the vision-based pose data. The autonomous machine may be navigated within the boundary of the work region based on the updated pose. The three-dimensional point cloud may be generated based on data captured during a touring phase. Boundaries may be generated based on data captured during a mapping phase.