Work Vehicle Obstacle Detection Using Field Coordinate Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automatic traveling systems for work vehicles in agricultural fields face inefficiencies due to erroneous obstacle recognition, leading to reduced work efficiency as they struggle to accurately determine obstacles while navigating crop rows.
Innovation Solution
The system employs a detection unit on the work vehicle to acquire coordinate positions of detection targets in a sensor coordinate system, converts these positions to a global coordinate system referencing the travel region, and determines obstacles based on these converted positions, thereby enhancing accuracy and preventing unnecessary travel restrictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the work vehicle automatically travels along the crop row route using GNSS positioning, then the work vehicle can navigate autonomously through the field, but the work vehicle may erroneously detect crops as obstacles due to posture changes and position deviations
Solution Approach 1:
The patent introduces an intermediary coordinate conversion process that transforms detection target positions from the vehicle's local coordinate system to the global field coordinate system. This intermediary step acts as a mediator between the detection unit and obstacle determination, resolving the contradiction by enabling automated travel while improving obstacle detection reliability through proper coordinate reference frame transformation.
2Object-affected harmful factors
If the work vehicle executes travel restriction process upon detecting obstacles, then collision avoidance is achieved, but work efficiency deteriorates due to erroneous obstacle recognition
Solution Approach 1:
The patent implements a feedback mechanism where detection target positions are continuously converted to global coordinate system and compared against the target route. This feedback loop allows the system to distinguish between actual obstacles and crops by referencing their positions relative to the predetermined route, thereby avoiding erroneous travel restrictions while maintaining collision avoidance capability.
3Speed
If the work vehicle detects detection targets in a coordinate system with the detection unit as reference, then real-time detection is achieved, but accurate obstacle determination becomes difficult due to coordinate system misalignment
Solution Approach 1:
The patent applies preliminary action by pre-establishing the global coordinate system of the field and the target route before the work vehicle begins operation. This preliminary setup enables real-time detection in the vehicle's local coordinate system to be quickly and accurately converted to the global coordinate system, maintaining both detection speed and position accuracy through pre-computed coordinate transformations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
[Problem] To provide an automatic traveling method, an automatic traveling system, and an automatic traveling program, by which it is possible to increase the determination accuracy of an obstacle to improve the work efficiency of a work vehicle. [Solution] The automatic traveling method includes causing a work vehicle 10 to automatically travel according to a target route R previously set in a farm field F, detecting a detection target in the farm field F by a detection unit provided in the work vehicle 10, acquiring a first coordinate position representing a position of the detection target in a sensor coordinate system having the detection unit as a reference, converting the first coordinate position into a second coordinate position representing a position of the detection target in an NED coordinate system having the farm field F as a reference, and determining whether the detection target is an obstacle, based on the second coordinate position.