Problem Description
Weed management is an essential aspect of tree nursery operations due to the competition between weeds and the target crops for resources such as nutrients, water, and light. Additionally, weeds can harbor parasites like insects, nematodes, and mites that may harm the crops. To combat weeds, nurseries often rely on herbicide spraying, a method that is both costly and environmentally damaging. The broad-spectrum application of herbicides leads to excessive use and potential damage to trees, which can render them unsellable. Current precision weed management techniques used in smaller crop farming are not directly applicable to tree nurseries due to the physical layout and the height of trees, presenting a challenge for autonomous solutions.
To address these issues, the project aims to develop an autonomous system capable of precision weeding using lasers within tree nurseries. This system aims to reduce the financial and environmental costs associated with traditional weed management methods. By precisely targeting weeds, the system could lower the total volume of herbicides used or even eliminate the need for them. However, implementing such a system requires overcoming challenges in accurately detecting and localizing weeds for effective targeting and manipulation in varied environmental conditions. The project will build on the existing moss© system for mapping and navigating tree nurseries, indicating a move towards innovative, environmentally friendly, and cost-effective weed management solutions.
Use Case
Ms. Picea operates a Christmas tree farm near Pittsburgh, PA, managing over 100,000 trees across several hundred acres. The trees are organized into blocks, with each block containing a dozen rows of about a hundred trees each, spaced roughly 1.5 meters apart and rows about 2 meters apart. To maintain the farm, Ms. Picea uses a ZAAPP autonomous weeding robot (ZAAPP-bot), which she controls via the moss© app. On a particular day, she decides to weed a single block.
The ZAAPP-bot is designed to autonomously navigate to the selected block and weed each row in turn. As it moves, it detects weeds, determines their positions relative to itself, and, using the moss© system, assigns a geotag to each weed, effectively mapping their locations within the larger farm. This data is compiled into a report.
For weeding, Ms. Picea has equipped the ZAAPP-bot with a laser mechanism, preferring its minimal environmental impact over traditional herbicides. The bot “zaapps” each detected weed with precision, avoiding damage to the trees and eliminating the need for chemical refueling.
Upon completing a block, the ZAAPP-bot sends a report to Ms. Picea’s cellphone, highlighting any weeds it couldn’t neutralize. With this detailed report, Ms. Picea assigns an employee to manually remove the remaining weeds, significantly reducing the time and effort needed by providing exact locations. This process exemplifies how technology can enhance efficiency in agricultural practices, especially in sustainable farming.