LiDAR Crop Row Detection for GNSS-Denied Auto Steering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In environments with high-density tree distribution, such as vineyards or orchards, the use of GNSS for precise positioning is hindered by tree canopies, and SLAM systems face challenges due to seasonal changes in leaf distribution, making it difficult for work vehicles to navigate accurately.
Innovation Solution
A crop row detection system using LiDAR sensors to scan and detect adjacent crop rows, creating a vehicle-fixed coordinate system map, and updating coordinate points to align with crop row centers, enabling automatic steering between rows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GNSS positioning is used for automatic steering, then positioning accuracy is improved, but it becomes inapplicable in environments with high-density tree distribution such as orchards and forests where tree canopies block satellite signals
Solution Approach 1:
The patent introduces LiDAR sensors as an intermediary measurement tool to detect crop row positions and generate maps when GNSS is unavailable. The system uses these sensor-based maps as a mediator to enable automatic steering in environments where satellite positioning cannot penetrate the tree canopy, thus resolving the contradiction between positioning accuracy and environmental adaptability.
Solution Approach 2:
The system dynamically switches between different positioning parameters and methods based on environmental conditions. When GNSS signals are blocked, the system transitions to using LiDAR-based spatial parameters and map coordinates for positioning, allowing automatic steering to function across diverse environments including dense orchards and forests.
2Adaptability or versatility
If SLAM is used for localization in environments where GNSS cannot be used, then automatic navigation becomes possible, but seasonal changes in leaf distribution make it impossible to continue using maps created in the past
Solution Approach 1:
The patent implements dynamic map updating using LiDAR sensors that continuously scan the environment and detect current crop row positions. The system adapts to seasonal changes by regenerating maps based on real-time sensor data rather than relying on static historical maps, ensuring reliable navigation regardless of leaf distribution changes throughout the year.
Solution Approach 2:
The system incorporates feedback mechanisms where LiDAR sensors continuously monitor the actual positions of crop rows and compare them with the map data. This feedback loop allows the system to detect discrepancies caused by seasonal changes and update the map accordingly, maintaining reliable navigation performance across different growing seasons.
3Adaptability or versatility
If LiDAR sensors are used to detect crop rows and create maps, then automatic steering becomes possible in GNSS-denied environments, but the system complexity increases due to sensor integration and processing requirements
Solution Approach 1:
The patent designs the LiDAR-based positioning system to serve multiple functions: it creates environmental maps for navigation, detects crop row positions for steering, and provides spatial reference data for various agricultural operations. This multi-functionality reduces the need for separate specialized systems, thereby limiting the increase in overall system complexity while maintaining high environmental adaptability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables automatic steering of agricultural machines in environments where GNSS positioning is difficult, such as orchards or forests, by accurately detecting and navigating between crop rows.
Implementation Method 1
a sensor attached to a vehicle configured to, when in operation, scan surrounding environment including crop rows and output sensor data containing position information of an object existing in the environment
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A crop row detection system of the present invention includes: a sensor that outputs sensor data including position information of an object; and a processing device that detects adjacent crop rows based on the sensor data. The processing device executes: creating a map of crop rows in a vehicle coordinate system based on the sensor data; using the map to start an object detection search parallel to a second coordinate axis from a first coordinate point in the vehicle coordinate system to detect a left row and a right row of the adjacent crop rows. The processing device is configured to, when updating the first coordinate point, increase or decrease the first coordinate on the first coordinate axis of the first coordinate point, and align a second coordinate on the second coordinate axis of the first coordinate point with a second coordinate on the second coordinate axis of a crop row center point, where distances to the left row and the right row of the adjacent crop rows are equal.