Perception Navigation Marker Obstruction Resolution at Worksites
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Perception-based navigation systems in mobile machines face issues with object detection due to environmental factors and activities at worksites, leading to incorrect localization and navigation, which can cause autonomous operations to cease until issues are resolved.
Innovation Solution
A computer-implemented analytic map processor with a marker identification module and detection module to identify and resolve marker obstructions, utilizing an onboard electronic controller with a data gathering module to analyze worksite data and generate obstruction resolutions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If perception sensors and rangefinders are used for localization and navigation, then the mobile machine can determine its location and navigate about the worksite, but object detection can be negatively affected by environmental factors and activities at the worksite leading to incorrect localization
Solution Approach 1:
The patent introduces an intermediary verification process that compares detected objects against a pre-stored worksite map. The system uses the map as a reference framework to validate perception sensor data, filtering out false detections caused by environmental factors. This intermediary layer reconciles the conflict between using perception sensors for navigation and maintaining reliable object detection despite environmental interference.
Solution Approach 2:
The system implements feedback by continuously comparing real-time perception sensor data with the stored worksite map and adjusting localization accordingly. When environmental factors cause detection issues, the feedback mechanism uses the map reference to correct localization errors, maintaining accuracy despite unreliable individual sensor readings.
2Measurement precision
If the mobile machine ceases operation until object detection issues are resolved, then localization accuracy can be maintained, but productivity is reduced due to operational interruptions
Solution Approach 1:
The patent applies preliminary action by pre-storing a detailed worksite map before operations begin. This pre-prepared reference framework allows the system to quickly resolve detection issues without ceasing operation, as the map provides immediate reference data for correcting localization errors. This eliminates the need to stop operations while maintaining localization accuracy.
Solution Approach 2:
The system dynamically adjusts its reliance between perception sensor data and the stored worksite map based on detection confidence levels. When environmental factors reduce detection reliability, the system automatically increases weight on the map reference, allowing continuous operation without compromising accuracy. This dynamic adaptation prevents operational interruptions while maintaining precision.
3Reliability
If multiple perception sensors are deployed to improve object detection, then detection coverage is enhanced, but device complexity increases
Solution Approach 1:
The patent uses a simplified copy of the worksite environment stored in the worksite map as a reference framework. This digital copy provides the detection reference structure without requiring complex sensor arrays, reducing device complexity while maintaining detection reliability through comparison with the stored environmental model.
Data Source
AI summary
To assist operation of a perception-based location and navigation system, a computer-implemented system and method can determine if a physical marker at a worksite is perceptibly obstructed. Worksite data can be gathered from the physical worksite and/or a worksite map in computer readable format. The worksite data is analyzed with respect to the worksite map to determine if the physical marker is obstructed. To resolve the marker obstruction, an obstruction resolution can be generated and output.