Agricultural Road Map Generation Using LiDAR Feature Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing agricultural machines lack efficient map generation systems for self-traveling on both fields and roads, particularly agricultural roads, which are crucial for autonomous navigation and path planning.
Innovation Solution
A map generation system that utilizes a storage to store feature block images and a processor to acquire position distribution data from LiDAR sensors and imagers, aligning these images to generate map data for agricultural machines, enabling them to navigate fields and roads effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If map data is generated using LiDAR sensor data and imager data collected while moving, then the map accuracy and feature representation are improved, but the data processing time and computational complexity increase
Solution Approach 1:
The patent segments the map generation process into distinct stages: data collection by LiDAR and imager, feature extraction to identify road boundaries and characteristics, and map data construction by integrating these features. This segmentation allows parallel processing of different data streams and enables intermediate storage of processed features, reducing overall processing time while maintaining map accuracy.
Solution Approach 2:
The patent performs preliminary processing of LiDAR point cloud data and imager data during the data collection phase, including noise filtering, feature detection, and preliminary classification. By preparing and pre-processing the raw sensor data in advance, the system reduces the computational burden during final map generation, thereby decreasing processing time without compromising measurement precision.
2Manufacturing precision
If feature block images are aligned according to position distribution data, then the map representation accuracy is improved, but the computational complexity and processing requirements increase
Solution Approach 1:
The patent introduces position distribution data as an intermediary element that bridges the LiDAR sensor data and imager data. This position distribution data contains pre-calculated spatial relationships and transformation parameters that simplify the alignment process. By using this intermediary, the system achieves accurate feature block image alignment without requiring complex real-time coordinate transformations, thus reducing processing complexity while maintaining high map representation accuracy.
3Area of stationary object
If the system collects sensor data and image data while the movable body is moving, then the map coverage area is improved, but the data synchronization and integration difficulty increase
Solution Approach 1:
The patent implements a feedback mechanism where the position and orientation of the movable body are continuously monitored using GNSS and inertial measurement units during data collection. This real-time feedback information is used to dynamically adjust the coordinate transformations and synchronize the LiDAR and imager data streams. By incorporating this feedback, the system achieves accurate data integration from moving platforms, enabling large map coverage areas while managing integration complexity through continuous position-based correction.
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
Efficiently generates map data for agricultural machines to travel autonomously on agricultural roads and fields, ensuring accurate representation of road positions and features, facilitating appropriate path planning and navigation.
Implementation Method 1
sensor data from a LiDAR sensor, the sensor data being output while a movable body including at least one of the LiDAR sensor and the imager is moving along the road
Data Source
AI summary
A map generation system generates map data for an agricultural machine to automatically travel on a road around a field, and includes a storage to store feature block images associated with different types of road features, and a processor configured or programmed to acquire position distribution data on one or more types of road features, the position distribution data being generated based on at least one of sensor data from a LiDAR sensor and image data from an imager output while a movable body including at least one of the LiDAR sensor and the imager is moving along the road; read from the storage one or more types of feature block images associated with the one or more types of features; and align the feature block images in accordance with the position distribution data to generate map data on a region including the road.