Grapevine is a distinctive crop of the Mediterranean Basin. In Italy, viticulture has a relevant economic impact in all regions, and the wine industry ranks first in the agro-food chain with over 5.5 billion euro of export in 2017. Despite widespread Vertical Shoot Positioning systems recently demonstrated positive features in several wine districts, modern vineyards still require expensive selective manual operations. Winter pruning impacts for the 20-35% of the yearly labor demand that can be lowered by introducing mechanical operations when spur-pruning is adopted. However selective manual follow-up is still required and cannot be performed in the absence of direct human cognition. Robots can represent a revolution upscaling in forthcoming innovative farming systems providing automated solutions combining intelligent robot vision and manipulation. Despite agricultural solutions have been developed in recent years, robotics in viticulture is still bursting, and only a few prototypes have been described for performing winter spur-pruning; at present engineering processes seem to be ongoing whilst interactions with grapevine physiology were never assessed. Moreover, the new generation of legged robots appears as a more flexible solution in challenging terrains as compared to traditional tracked and wheeled platforms. The main challenge of the project is developing and testing innovative robotic technologies for grapevine spur-pruning automation.
In a multidisciplinary perspective, the project aims at:
- providing an original flexible and robust legged pruning robot operating in field destructured surfaces and allowing a quality of work comparable to traditional methods;
- achieving significant knowledge on still unexplored interactions between robotized pruning, plant growth, productivity and enological potential of grapes.
To fulfill the above targets, the research activity will be developed over five years and structured in four actions (a.k.n., WPs) leading to the following deliverables:
- setting up of a specific vision system
- adaptation to pruning operations of an already existing arm
- control of a legged platform in a vineyard scenario
- integration of the vision system and the arm onto the legged locomotion platform resulting in a pruning-robot prototype
The first action will provide hierarchical guidelines for a translation of the human cognition process into algorithms for automated plant recognition, and robotized pruning. Multiple combinations of integrated general rules driving winter pruning process with specifically conceived case studies will compose a dedicated database for developing multi-modal sensing system for recognition of grapevine architecture and identification of target regions and cutting points.
The second action will integrate off-the-shelf electric shears into a commercial robot manipulator controller, to autonomously perform the cutting operation from the output of the first action. The designed controller should not disrupt the condition of the grapevine, and the cuts quality must be comparable to the human one.
The third action aims at advancing the state of the art of legged robots. Theoretical and practical researches will be pursued to allow a legged platform to move within a vineyard autonomously. Outcomes from previous activities will be available in the second year of the project when will be integrated together and the resulting prototype will be tested in the fourth action, under real scenarios and compared with human operators. So, the impact of robotized pruning on cutting-pose will be initially tested on mature potted-vines for assessing operation quality. Moreover, the effect of robotized pruning on grapevine physiology will be evaluated by monitoring whole-canopy real-time gas-exchange on a seasonal base.
By the end of the project, the robot prototype will be demonstrated in a commercial spur-pruned vineyard where a comparison with human operators will be carried out under different conditions. Therefore, robot and plant performance will be characterized in several parcels varying in terms of soil conditions (bare soil vs grassed), slope (flat vs steep versant), plant growth (high vs low vigor), and cultural facilities (pre-pruning vs control vines).
Due to its innovative nature, the project has relevant application potentialities impacting on different domains. The project targets significant advancement of knowledge and technological goals aiming to jump from the initial Technology Readiness Level TRL3 up to TRL6. Because novel services to growers, technicians, and scientists will foster research and innovation, companies adopting these technologies will become more competitive in the market. The upcoming robotics in viticulture might reduce labor demand and costs, will improve growers’ living conditions and reduce accidents’ frequency among workers.