As critical pollinators, bees keep our agricultural systems going—but human-caused changes to the planet heavily impact their foraging options. To help protect our food security, we need more information about bees’ own dietary requirements. Scientists writing in Frontiers in Sustainable Food Systems have studied the nutritional value of 57 types of pollen and found that bees need to forage from a variety of plants to balance their diet between fatty acids and essential amino acids.

“Despite public interest and a rise in pollinator plantings, little is known about which plant species are best suited for bee health,” said Dr. Sandra Rehan of York University, senior author. “This study aimed to better understand the nutritional value of plant species. Based on their ideal protein to lipid ratios for wild bee nutrition, we recommend that pollen species from roses, clovers, red raspberry, and tall buttercup should be emphasized in wildflower restoration projects.”

The bees’ needs

Pollen and bees are heavily interdependent: Plants need bees to spread their pollen to reproduce, and bees need pollen to eat. While bees get their carbohydrates from nectar, pollen provides proteins, lipids, and other critical nutrients. Anthropic changes to the environment which alter the availability and the properties of pollen risk malnourished bees.

Bees especially need to consume high-quality foods containing non-esterified fatty acids like omega-6 and omega-3. Without these nutrients, bees live shorter lives, have weaker immune systems, and are less able to cope with environmental stressors—but if bees consume them in the wrong ratio, they experience cognitive problems. Bees also need essential amino acids, which are necessary for cognitive health and reproduction—but if they eat too much, they may be more susceptible to certain parasites.

To understand which plants are best for bees, the scientists collected pollen samples from 57 species found in North America, either from fresh flowers in the wild or from flowers dried in the lab. They chose the plant species based on their importance to species of wild northeastern bee and their prevalence. The pollen was processed and analyzed for levels of different amino acids, non-esterified fatty acids, and protein to lipid and omega-6:3 ratios, to determine which plants were best for bees.

The scientists also investigated whether closely related species of plant provided similar nutritional benefits, and whether species that had been introduced to the area where they were collected were less nutritious than endemic species.

Healthy eating habits

In general, plants from the same family offered bees quite different nutrients, with the exception of essential amino acids. Plants from the cabbage family, the legume family, and the daisy family all had similar levels of essential amino acids compared to other plants within the same family. Daisies, a very important plant for foraging bees, boasted particularly high levels of essential amino acids. Interestingly, plants that were high in essential amino acids were relatively low in non-esterified fatty acids, and vice versa.

“There is a potential tradeoff between fatty acid and amino acid content within pollen, suggesting that a diverse floral diet may benefit bees more than a single pollen source,” said Rehan. “No one plant species is optimal for generalized wild bee health.”

The scientists’ results indicated that feeding from many different flowers is best for most bees, and that feeding from endemic species of plant offers no nutritional advantage. Most pollen species contain most of the necessary nutrients, but to get the optimal levels of nutrients in their diets, bees would need to forage from several different plant species. The scientists suggested that this diversity of nutritional content reflects the diverse needs of different species of bees, especially the specialist species that favor particular plants. A wide variety of sources of nutrition with different properties means that all bees can forage on the plants that feed them best.

“We hope this work will help inform flowering plant selections for pollinator gardens,” said Rehan. “But here we examined only 57 plant species, and there are thousands to examine to understand nutritional profiles. We hope this will inspire future similar research as well as follow-up studies on the preference and survival of bees on different diets.”

While roboticists have introduced increasingly sophisticated systems over the past decades, teaching these systems to successfully and reliably tackle new tasks has often proved challenging. Part of this training entails mapping high-dimensional data, such as images collected by on-board RGB cameras, to goal-oriented robotic actions.

Researchers at Imperial College London and the Dyson Robot Learning Lab recently introduced Render and Diffuse (R&D), a method that unifies low-level robot actions and RBG images using virtual 3D renders of a robotic system. This method, introduced in a paper published on the arXiv preprint server, could ultimately facilitate the process of teaching robots new skills, reducing the vast amount of human demonstrations required by many existing approaches.

“Our recent paper was driven by the goal of enabling humans to teach robots new skills efficiently, without the need for extensive demonstrations,” said Vitalis Vosylius, final year Ph.D. student at Imperial College London and lead author. “Existing techniques are data-intensive and struggle with spatial generalization, performing poorly when objects are positioned differently from the demonstrations. This is because predicting precise actions as a sequence of numbers from RGB images is extremely challenging when data is limited.”

During an internship at Dyson Robot Learning, Vosylius worked on a project that culminated in the development of R&D. This project aimed to simplify the learning problem for robots, enabling them to more efficiently predict actions that will allow them to complete various tasks.

In contrast with most robotic systems, while learning new manual skills, humans do not perform extensive calculations to determine how much they should move their limbs. Instead, they typically try to imagine how their hands should move to tackle a specific task effectively.

“Our method, Render and Diffuse, allows robots to do something similar: ‘imagine’ their actions within the image using virtual renders of their own embodiment,” Vosylius explained. “Representing robot actions and observations together as RGB images enables us to teach robots various tasks with fewer demonstrations and do so with improved spatial generalization capabilities.”

For a robot to learn to complete a new task, it first needs to predict the actions it should perform based on the images captured by its sensors. The R&D method essentially allows robots to learn this mapping between images and actions more efficiently.

“As hinted by its name, our method has two main components,” Vosylius said. “First, we use virtual renders of the robot, allowing the robot to ‘imagine’ its actions in the same way it sees the environment. We do so by rendering the robot in the configuration it would end up in if it were to take certain actions.

“Second, we use a learned diffusion process that iteratively refines these imagined actions, ultimately resulting in a sequence of actions the robot needs to take to complete the task.”

Using widely available 3D models of robots and rendering techniques, R&D can greatly simplify the acquisition of new skills while also significantly reducing training data requirements. The researchers evaluated their method in a series of simulations and found that it improved the generalization capabilities of robotic policies.

They also showcased their method’s capabilities in effectively tackling six everyday tasks using a real robot. These tasks included putting down the toilet seat, sweeping a cupboard, opening a box, placing an apple in a drawer, and opening and closing a drawer.

“The fact that using virtual renders of the robot to represent its actions leads to increased data efficiency is really exciting,” Vosylius said. “This means that by cleverly representing robot actions, we can significantly reduce the data required to train robots, ultimately reducing the labor-intensive need to collect extensive amounts of demonstrations.”

In the future, the method introduced by this team of researchers could be tested further and applied to other tasks that robots could tackle. In addition, the researchers’ promising results could inspire the development of similar approaches to simplify the training of algorithms for robotics applications.

“The ability to represent robot actions within images opens exciting possibilities for future research,” Vosylius added. “I am particularly excited about combining this approach with powerful image foundation models trained on massive internet data. This could allow robots to leverage the general knowledge captured by these models while still being able to reason about low-level robot actions.”

© 2024 Science X Network