Row Vision Edge Detection for GPS-Limited Tractor Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated vehicle operations, such as farming vehicles, face challenges due to unreliable GPS signals and limited processing power, leading to inaccurate environmental understanding and delayed navigation.
Innovation Solution
A row vision system that uses camera images and operator input to detect edges between surfaces, employing machine learning models to select the most accurate edge for navigation, thereby reducing processing requirements and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If GPS information is used for automated navigation, then wireless communication convenience is improved, but signal reliability deteriorates due to fading, shadowing, and interference
Solution Approach 1:
The patent introduces an intermediary system (computer vision system with cameras and image processing) that mediates between the unreliable GPS wireless communication and the navigation function. Instead of relying solely on GPS signals, the system uses visual intermediaries (camera images) to detect edges and determine vehicle position and orientation, thereby resolving the contradiction by finding an alternative information channel that is not subject to radio frequency interference.
2Measurement precision
If computer vision systems process entire images to identify edges, then edge detection accuracy is improved, but processing time and power consumption increase
Solution Approach 1:
The patent applies segmentation by dividing the image processing task into distinct stages: first identifying candidate edge regions using efficient algorithms, then applying more computationally intensive edge detection methods only to those candidate regions. This segmentation allows the system to maintain high accuracy while reducing overall processing time and power consumption by avoiding exhaustive analysis of the entire image.
Solution Approach 2:
The patent implements partial action by performing comprehensive edge detection only on candidate regions identified as potentially containing relevant edges, rather than processing the entire image with the same level of detail. This selective approach applies excessive processing only where necessary, optimizing the balance between accuracy and computational efficiency.
3Ease of manufacture
If embedded controllers are used for automated operation, then on-vehicle processing is improved, but processing power and resources are limited causing delays
Solution Approach 1:
The patent applies dynamics by creating a flexible, adaptive processing architecture that can dynamically adjust the level of processing based on computational resources available and task priorities. The system can shift processing loads between the embedded controller and external computing resources, and can adjust the complexity of image processing algorithms based on real-time performance requirements, thereby resolving the contradiction between on-vehicle processing capability and processing speed.
4Ease of operation
If mobile devices are used for automated operation, then commercial availability and convenience are improved, but processing and power constraints still cause delays
Solution Approach 1:
The patent applies merging by combining the computational resources of the mobile device with cloud-based or external processing capabilities. The mobile device handles data acquisition and preliminary processing, while more computationally intensive tasks are offloaded to external systems, creating a hybrid architecture that leverages the convenience of mobile devices while overcoming their processing limitations through resource consolidation.
Data Source
AI summary
A row vision system modifies automated operation of a vehicle based on edges detected between surfaces in the environment in which the vehicle travels. The vehicle may be a farming vehicle (e.g., a tractor) that operates using automated steering to perform farming operations that track an edge formed by a row of field work completed next to the unworked field area. A row vision system may access images of the field ahead of the tractor and apply models that identify surface types and detect edges between the identified surfaces (e.g., between worked and unworked ground). Using the detected edges, the system determines navigation instructions that modify the automated steering (e.g., direction) to minimize the error between current and desired headings of the vehicle, enabling the tractor to track the row of crops, edge of field, or edge of field work completed.