Autonomous Aerial Pollination Using AI Plant Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional pollination methods for large plants, such as trees, are inefficient, labor-intensive, and inaccurate, requiring manual intervention or continuous drone operation, which leads to pollen wastage and resource inefficiency.
Innovation Solution
An autonomous aircraft system equipped with an image capture device, pollen dispensing system, and flight control system, utilizing computer vision and AI to identify and center on plants, dispense pollen efficiently, and navigate around obstacles, allowing for automated pollination of multiple plants in a plot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional drone pollination continuously dispenses pollen while flying over plants, then pollination coverage is maintained, but pollen waste increases and battery charge depletes faster
Solution Approach 1:
The system performs preliminary actions by capturing images of plants before pollination, processing these images to identify plant locations and characteristics, and planning an optimized flight route in advance. This allows the drone to dispense pollen only when needed and only in the required quantities, rather than continuously dispensing pollen throughout the entire flight area.
Solution Approach 2:
The system uses feedback by capturing real-time images of plants during flight, processing these images to identify plant locations, and adjusting the pollen dispensing accordingly. The processed image data feeds back to the control system to determine precise dispensing timing and quantity, ensuring pollen is released only when the drone is positioned over target plants.
2Measurement precision
If manual hand-pollination of individual flowers is performed, then pollination accuracy is high, but labor time and cost increase significantly
Solution Approach 1:
The system replaces the mechanical manual action of hand-pollination with an automated drone system equipped with image capture and processing capabilities. The drone autonomously identifies plants through image capture and processing, then dispenses pollen automatically, eliminating the need for manual labor while maintaining pollination accuracy through precise plant identification and targeted dispensing.
Solution Approach 2:
The system enables self-service by equipping the drone with autonomous capabilities to capture images, process plant identification, determine pollination targets, and execute pollen dispensing without human intervention. The drone serves itself by making autonomous decisions about where and when to dispense pollen based on processed image data.
3Ease of manufacture
If mobile blower operates at base of tree to blow pollen upward, then equipment cost is reduced, but pollination efficiency and accuracy remain low
Solution Approach 1:
The system inverts the conventional approach by dispensing pollen from above (aerial) rather than from below (ground level). Instead of blowing pollen upward from the base of trees, the drone flies over the plants and dispenses pollen directly onto the flowers from above, reversing the direction of pollen delivery to improve efficiency and accuracy.
Solution Approach 2:
The system transitions from ground-level operation to aerial operation, adding the vertical dimension to pollination delivery. By flying over plants at optimized altitudes and dispensing pollen from above, the system overcomes the limitations of ground-based blowers that struggle to reach high canopy flowers, thereby improving pollination efficiency for tall plants.
Data Source
AI summary
The systems and methods described herein relate to fully or partially autonomous or remotely operated aerial pollination vehicles that use computer vision and artificial intelligence to automatically detect plants, orient the vehicle to a pollen dispensing position above each plant, and pollinate the individual plants.