Autonomous Robotic Platforms for Greenhouses

Why it matters?

The protected cropping industry is the fastest-growing food-producing industry in Australia. It is valued at $1.5 billion farm-gate value per annum which is equivalent to 15% of the industry. The current horticulture industry is facing challenges with the high cost of crop production requiring a large amount of labour. By developing robotic crop management systems, which can support the crop production systems, the agricultural industry could benefit economically but also through the reduction of risks caused by pests, diseases, or food safety from contamination.

Currently, glasshouse-based hydroponic growing systems control the cropping environment via temperature, humidity and nutrient inputs. These control systems utilise feedback sensors for these specific states of the environment including temperature, humidity, pH, and reduction of nutrient levels in the hydroponic system. However, these systems fail to directly measure the state of the crop such as health, the current growth stage of the crop, pests or diseases, yield, and quality of the crop. These systems also rely heavily on labour-intensive tasks such as planting, pruning, harvesting, and cleaning. To be efficient in time and resources a new type of robotic crop management system is required which can autonomously make optimal decisions and actions to maximise yield while minimising cost.

Project Overview

The goal of this project is to research and develop a fully autonomous robotic crop management system for protected cropping systems that will help the horticulture industry reduce the risk of crop loss and improve productivity by having the right information to make better decisions.

A new crop monitoring and automation robot is currently being developed that will help farmers effectively monitor their crops, reduce crop losses which contribute to food waste and improve the quality of produce. This new concept aims to achieve this by having the capability of autonomously navigating within a typical greenhouse environment. This new concept is building on the successful Harvey robotic platform, an agricultural robot prototype that was designed to robotically identify and harvest capsicums. The new platform recognises the potential that robotic platforms can bring to the industry by not only offering harvesting capability but lower hanging fruit such as autonomous crop monitoring through state-of-the-art navigation and vision systems.

This new robot platform will have a modular design for different applications and robot arms and monitoring systems to fit on top and a lightweight body for easy transport and deployment. It will also be easily replicated into different lengths and sizes to be used for varying applications. We are designing this robot from the ground up so it can navigate on both concrete floorings and autonomously transition to a pipe rail system to move down a row. We are focused not just on the mechanical design but also on the software needed to execute the behaviour to better support our agricultural industry.


Real-world impact

Growers have expressed a need for better ways to monitor their crops, respond quickly to problems and get support for making decisions in managing their crops. This robotic platform aims to support a variety of activities like crop monitoring and decision-making before acting such as harvesting and pruning. We are focusing on an automation platform that makes it easy to add new features as we go. Autonomous crop monitoring would be a key feature as it has value for the grower in being able to inform the grower about whether their crop is growing well or if there are nutrient deficiencies or diseases that need to be addressed immediately.

Other Team Members

  • Anthony Hellwig
  • Jerry Lo
  • Cheng Yan Ding