Robotic technology for alternative soil management

Alternative soil management is a crucial aspect of sustainable agriculture, focusing on practices that maintain or improve soil health while reducing environmental impact. Robotic technology has emerged as a vital tool in this domain, offering precise and efficient solutions for various soil management tasks.

In the current state of the art, we have autonomous soil monitoring systems, mostly at TRL6. These robots are designed to gather real-time data on soil conditions, including moisture, temperature, pH levels, and nutrient content. Equipped with advanced sensors and IoT connectivity, these robots enable continuous monitoring and data collection. The biggest challenge in developing these systems is researching and finding soil sensors that are cost-effective, robust, and accurate, which can work without the need to collect soil content in depth. However, soil monitoring inevitably requires soil sampling. In the state of the art, robotic systems automate the process of collecting soil samples for laboratory analysis. These robots can navigate fields autonomously, using GNSS and computer vision to locate sampling points. The main challenge here is developing soil sampling tools that are fast, cost-effective, and capable of working in rocky and inclined terrain.

Precision agriculture robotics includes robots designed for high-accuracy soil management tasks. These tasks include soil tillage, vegetation grow controlcontol, weeding, and planting, which require precise control and adaptability. Innovations in this field include: Autonomous Tillage Robots: Machines that prepare the soil by tilling, ensuring optimal soil structure for planting;  Weeding Robots: Robots that identify and remove weeds, or just control the vegetation grow, reducing the need for chemical herbicides; and, Variable Rate Technology (VRT): Systems that adjust the application of inputs (fertilisers, water) based on real-time soil data.

Robotic technology can also manage and mitigate soil erosion, a significant challenge in sustainable agriculture. Solutions in this area focus on erosion monitoring robots that can monitor erosion-prone areas and provide data for preventive measures. Examples include, reforestation and vegetation robots: Systems that plant or seed vegetation to stabilise soil and prevent erosion.

Example from NOVATERRA

In the NOVATERRA project, the TRIBE LAB from INESC TEC has developed the Mowit tool. This tool extends the Modular-E ( see pictures) with capacity for vegetation control both intra-row and in the row. It also includes – from other ongoing projects – a soil sensing module that measures electrical conductivity, a soil fertiliser that can fertilise each plant according to a prescription map, and a VRT sprayer for precision spraying to reduce losses and increase efficiency levels.

These robots navigate autonomously using the Simultaneous Localisation and Mapping  framework VINESlam, which is capable of localising the robot even when GNSS is unavailable by detecting natural features using cameras and/or LiDAR. The path can be manually designed by the user or generated by AgRobPP, a path planning approach for agricultural terrains that considers the robot’s center of mass to ensure its safety. Additionally, the robots are equipped with supervisors to ensure the safety of both the robot and its surroundings, aiming to achieve conformity with all European ISO standards for autonomous machinery in agriculture.

Robotic technology for alternative soil management is rapidly evolving, offering innovative solutions to enhance soil health and agricultural sustainability.