Engineers are turning trees into power plants, using sound waves to hear the footsteps of disease, and building an AI that can warn you when you are talking too much. Enough said.
MIT researchers are developing a wearable artificial intelligence system to detect when conversations are going south. In its early stages, the device can detect from a conversation’s tone whether speakers are happy, sad or neutral from the wearer’s vital signs and audio analysis of participants. Eventually the team wants to improve the system so that it can be a social coach, helping people with anxiety or Asperger’s syndrome navigate the complexities of human interaction. “As far as we know, this is the first experiment that collects both physical data and speech data in a passive but robust way, even while subjects are having natural, unstructured interactions,” says Mohammad Ghassemi, a PhD candidate at MIT’s Computer Science and Artificial Intelligence Laboratory. “Our results show that it’s possible to classify the emotional tone of conversations in real-time.”
This is what we call a real power plant! Iowa State University researchers have developed a power generator that mimics how tree leaves interact with wind. When the wind blows, it moves special plastic attached to the leaves that creates electricity through what’s known as the piezoelectric effect. The first prototype is a simple proof-of-concept device. The team calculated that a real cottonwood tree catches enough energy carried by a breeze blowing at 10 miles per hour to generate 80 watts of power. They believe that more research could lead to a tree-inspired power plant capable of generating half that power or more. It could be enough to recharge batteries or power household appliances.
There’s something oddly calming about watching a soft and nearly invisible grabber robot made of hydrogel catch and release a fish. But that’s not why MIT engineers are building it. The team is hoping to adapt their water-based machines for surgical applications, where they could gently manipulate tissues and organs. That’s where the fish grabbing comes in. “When you release the fish, it’s quite happy because [the robot] is soft and doesn’t damage the fish,” says Xuanhe Zhao, associate professor of mechanical engineering and civil and environmental engineering. “Imagine a hard robotic hand would probably squash the fish.”
Early diagnosis improves the odds of recovery, but many deadly diseases can be hard to spot, especially in the earliest stages. Duke University researchers are developing a tool that could help doctors listen for the body’s distress signals. The tool creates a small acoustic whirlpool “that can concentrate nanoparticles using nothing but sound,” according to a press release. The whirlpool can capture “proteins and other biological structures from blood, urine or saliva samples” for further analysis. “My goal is to create a small diagnostic device about the size of a cell phone that can autonomously separate biomarkers from samples,” said Tony Huang, professor of mechanical engineering at Duke. “With this vortex technology, the biomarkers could then be concentrated enough to see with a simple camera like the ones found in today’s cellular phones.”
A piece of paper could one day transform the costly, inefficient process of purifying polluted water or saltwater for consumption. Engineers at the University of Buffalo in New York are using a porous paper coated with carbon black to rapidly vaporize water in a still powered by the sun. The whole device is about the size of a dorm fridge and loses only about 12 percent of the energy that goes into the system as waste heat. Materials for their still cost just $1.60 per square meter and the still can produce up to 10 liters of water a day, according to the team. Current systems that use pricey solar concentrators can cost hundreds of dollars per square meter and produce up to around five liters of water a day. “Using extremely low-cost materials, we have been able to create a system that makes near maximum use of the solar energy during evaporation,” says Qiaoqiang Gan, a University at Buffalo associate professor of electrical engineering. “At the same time, we are minimizing the amount of heat loss during this process.”