Stickbug pollination robot | Trevor Smith
The issue of endangered pollinators has been a major area of concern for the floriculture and horticulture industries. Climate change, the increased use of pesticides, the loss of habitats, and many other reasons are putting the populations of bees, butterflies, and other important pollinators at extreme risk. This, in turn, has raised the risk for food security and shortages in the near future.
This has led to the technological sector and engineers experimenting with ways to properly mimic the effects of pollinators, thereby replacing and/or offsetting the dwindling populations of natural and native pollinating species. While this idea has been around for years, producing such designs of individual robotic bees covered by NPR, the practicality of such devices has often been a major sticking point, running into such issues as cost-effectiveness, imitation ability, and scalability. However, researchers from West Virginia University think they’ve made a breakthrough with their most recent design, a precision pollination robot they’ve called Stickbug.
Stickbug: Designed for Greenhouses
This information comes from a robotics study, straightforwardly dubbed “Design of Stickbug: A Six-Armed Precision Pollination Robot,” which can be found hosted on arXiv. Crucially, what sets Stickbug apart from past robotic pollinators is its multi-agent design, which “combines the accuracy of single-agent systems with swarm parallelization in greenhouses.” In doing so, each of the autonomous robot’s six arms functions independently while not only reducing planning complexity but also increasing task effectiveness and enhancing pollination output.
What also makes Stickbug’s design unique compared to past robotic pollinators is that it has been specifically designed to operate in greenhouse environments. This is achieved through a compact holonomic Kiwi drive, which enables the robot to navigate narrow rows, paired with a detection model and classifier to identify flowers in need of contact-based pollination. The process is then commenced using “a felt-tipped end-effector.”
Putting the Stickbug prototype to the test, the researchers assigned the robot to an artificial bramble plant and tasked it to pollinate as many flowers as possible within the course of 5 minutes. According to the authors of the study, “Initial experimental validation demonstrates that Stickbug can attempt over 1.5 pollinations per minute with a 50% success rate.”
The Future of Robotic Pollinators
While showing promising results, the researchers noted that there’s still a lot of work to be done with Stickbug. They aim to conduct subsequent experiments during the flowering season, this time with live plants. That’s because a primary limitation of the initial study was “the lack of access to real flowers for experimentation due to the short pollination season.” They now hope to “improve flower memory and re-identification through Intersection over Union (IoU) and flower relative graph-based mapping.”
“Additionally, we plan to enhance manipulator capabilities by integrating a search function and flower load balancing, leveraging the referee to generate a global flower map to direct manipulators toward unexplored and flower-dense regions.”
With further rounds of experimentation and improvements, Stickbug could be a strong starting point for a future of completely robotic pollinators securing the world’s food supply.