Work Vehicle Row-Following Control Under GNSS-Denied Canopy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In environments with high-density tree or crop distributions, such as vineyards or orchards, the canopy of leaves creates obstacles and multiple reflectors for radio waves, hindering accurate Global Navigation Satellite System (GNSS) positioning. Additionally, seasonal changes in leaf distribution render pre-created maps obsolete, posing challenges for autonomous work vehicle navigation using Simultaneous Localization and Mapping (SLAM) techniques.
Innovation Solution
A work vehicle equipped with exterior sensors, such as LiDAR, to output sensor data indicating the distribution of geographic features. The vehicle includes a controller programmed to manage self-traveling in inter-row and turning travel modes. In the inter-row mode, the vehicle travels along a target path between detected crop rows, and in the turning mode, it calculates positional and directional deviations to switch from turning to inter-row travel upon satisfying specific threshold conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If SLAM is used for autonomous navigation in high-density tree environments, then the work vehicle can travel automatically without GNSS, but the system cannot adapt to seasonal changes in leaf distribution making pre-created maps obsolete
Solution Approach 1:
The system dynamically adapts by detecting crop rows in real-time using LiDAR and dynamically adjusting the travel path based on current environmental conditions rather than relying on static pre-created maps. The controller continuously updates the target path between detected crop rows, enabling adaptation to seasonal changes in leaf distribution and crop growth stages.
2Ease of operation
If the work vehicle switches between turning travel mode and inter-row travel mode based on multiple conditions, then the vehicle can achieve smooth self-traveling among crop rows, but the control system complexity increases
Solution Approach 1:
The autonomous travel function is segmented into distinct modes: turning travel mode for navigating at intersections or headlands, and inter-row travel mode for moving between crop rows. Each mode has specific switching conditions based on positional deviation and directional deviation thresholds, which simplifies the control logic within each mode while enabling complex overall behavior through mode transitions.
3Reliability
If exterior sensors like LiDAR are used to detect geographic features and crop rows, then the vehicle can navigate without GNSS in canopy environments, but the measurement precision may be affected by complex terrain and vegetation
Solution Approach 1:
The system uses detected crop rows as intermediary reference features to establish the target path. Instead of directly navigating to absolute coordinates which may be imprecise in canopy environments, the vehicle detects crop rows using LiDAR, calculates the target path between these intermediary features, and follows this relative path, thereby improving navigation reliability despite measurement challenges.
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 smooth self-traveling among multiple crop rows in environments where GNSS positioning is difficult, maintaining accuracy and adaptability despite seasonal changes in leaf distribution.
Implementation Method 1
A work vehicle equipped with exterior sensors, such as LiDAR, to output sensor data indicating the distribution of geographic features
Data Source
AI summary
A work vehicle includes an exterior sensor to output sensor data indicating geographic features around the work vehicle, and a controller configured or programmed to control self-traveling of the work vehicle in an inter-row travel mode of travel along a target path between adjacent crop rows detected based on the sensor data, and in a turning travel mode of turning in a headland before and after the inter-row travel mode. In the turning travel mode, the controller calculates amounts of positional and directional deviations with respect to a target path in a next instance of the inter-row travel mode based on the sensor data, and switches from the turning travel mode to the inter-row travel mode upon satisfying a first condition that the amount of positional deviation is smaller than a first threshold and a second condition that the amount of directional deviation is smaller than a second threshold.