A Roomba for Rivers
The Big Picture features technology through the lens of photographers.
Every month, IEEE Spectrum selects the most stunning technology images recently captured by photographers around the world. We choose images that reflect an important advance, or a trend, or that are just mesmerizing to look at. We feature all images on our site, and one also appears on our monthly print edition.
Enjoy the latest images, and if you have suggestions, leave a comment below.
Roomba for Rivers
Humankind is enamored with water. The beauty and utility of the Earth's oceans, rivers, lakes, and streams explain why 40 percent of us live within 100 kilometers of the planet's coastlines. But we don't always respect and properly care for the things we love. As with many of our habitats, the world's waterways have become dumping grounds for our trash. Picking up the litter that fouls these otherwise picturesque areas is a full-time job. But few localities have the resources or political will to pay for cleanup costs. That might change now that French robotics company Interactive Autonomous Dynamic Systems (IADYS) has introduced the Jellyfishbot. The machine, which can run autonomously or at the direction of a remote operator, goes around collecting the junk and gunk (like plastic bottles, oil spills, and algae) that float on the water, as well as detritus located up to 10 meters below the surface. The Jellyfishbot is studded with sensors that not only allow it to navigate autonomously, but also measure the quality of the water in terms of salinity, temperature, turbidity, and the proliferation of organisms, including cyanobacteria and phytoplankton. Hooray for robot labor!
Sebastian Gollnow/Getty Images
Blowin' in the Wind
Anyone who has ever been responsible for tending to a lawn or garden knows that dandelions are an indomitable foe. As with the mythological hydra, you pluck one from among the blades of grass only to watch helplessly as several others take its place. At long last, scientists have studied the dandelion's winning ways to adopt the weed's traits for their own purposes. A team of researchers at the University of Washington has designed tiny sensors that mimic the shape and aerodynamic capability of the dandelion's fuzzy, seed-bearing spores. Nature has crafted the spores so that they can catch a gust of wind and travel as far as a kilometer before landing and putting down roots that will eventually yield a new flower. By taking cues from the strategy that has allowed the dandelion to flourish despite our herbicidal efforts, the scientists have successfully spread their solar-powered sensors across a large area for environmental monitoring, without the time, effort, and cost that manual placement would require.
Mark Stone/University of Washington
AI on the Fusion Menu
At first glance, the item pictured here looks like a motorcycle tire. But it's actually a tokamak fusion reactor. Researchers from DeepMind and the Swiss Federal Institute of Technology (EPFL), in Lausanne, Switzerland, are using this particular one to get us closer to harnessing fusion power for generating electricity. In order to pull that off, the tokamak will need a carefully calibrated arrangement of the magnetic coils that generate the device's magnetic fields. Those fields control fusion reactions in plasma and contain the material as it reaches temperatures approximating those inside the sun. In their quest to get those coils perfectly situated, the team has been relying on a type of AI called deep reinforcement learning.
DeepMind
AI's Green Thumb
What isn't AI good for? Scientists have known for many years that the oils and sugars in algae could be refined and turned into a renewable replacement for the petroleum products that have kicked off global climate change since the start of the Industrial Revolution. They've also been aware that we would need a whole heck of a lot of it if it is to make a dent in our reliance on fossil fuels. Many schemes for ramping up the production of algae blooms have been devised; thus far, all have fallen way short of meeting the massive demand for energy to provide light and heat (or cooling) in our homes and businesses, mechanical propulsion in vehicles, and labor-saving work done by machines of myriad shapes and sizes. Into that gap has stepped AI. A team of researchers at Texas A &M University is using artificial intelligence to more reliably industrialize the cultivation of algae so that it can live up to its initial promise. They have created two machine learning models that boost algae cultivation. One predicts how light will propagate through an algae bloom; the other predicts the point at which the algae's growth will become self-limiting because the parts of the bloom closest to the light source begin to block the rays from reaching the rest of the organism. Harvesting some of the algae just before it reaches that counterproductive concentration keeps the blue-green stuff growing at rates that were heretofore unsustainable.
Texas A&M University