Underride car crashes are among the most deadly accidents in Turkey, according to a report by the country's General Directorate of Highways. The accidents occur when a car collides with the rear or side of a large truck and gets stuck underneath it.
After learning about how fatal the crashes can be from recent news reports, Kerem Bayhan was inspired to develop a technology to help reduce them. Bayhan, who lives in İzmit, Turkey, is set to study medicine this semester at Hacettepe University, in Ankara.
He says he sees science through a "solution-based lens": He isn't looking for new problems to solve; he is seeking simple solutions to current problems such as underride collisions.
Commercial trucks already have guard bars on the back to prevent such accidents, so instead of looking to improve trucks, Bayhan decided to build an attachment for cars. He came up with a mechanical barrier system that attaches to the hood and connects to the car's electronic control unit. The outer layer is made of energy-absorbing material, and the second layer is made of metal. Airbags are placed between the first and second layers to propel the hood attachment forward. When the car's active safety system senses an impact is about to occur, the device extends before the crash happens to protect the driver and passengers.
The invention won Bayhan the US $10,000 IEEE Presidents' Scholarship, which is payable over four years of undergraduate university study. He also received a complimentary IEEE student membership. The award was presented during the Regeneron International Science and Engineering Fair. Established by the IEEE Foundation and administered by IEEE Educational Activities, the award is given annually at the ISEF to one student for an outstanding project that demonstrates an understanding of an IEEE field of interest. Recipients are selected by a team of IEEE members.
Bayhan plans to pitch his device to the Turkish government and car manufacturers. He hopes to talk with Turkey's science minister and the TOGG [Turkish Automobile Joint Venture Group], a new company formed by the government. The TOGG is working on designing electric vehicles.
A SELF-REPLICATING 3D PRINTER
The second-place winner was Brian Minnick, who developed a self-replicating 3D printer. The incoming freshman at MIT won $600. He plans to study material science and engineering.
The concept of a 3D-printed 3D printer was a theory before Minnick began his project three years ago. Several research articles he came across said it was impossible to create such a machine because it would be unable to print motors and other electronic parts. Although 3D printers require conductive materials, Minnick found during his research that commercially available filaments were not conductive enough to print the necessary components.
That didn't stop him, however. In fact, the very idea of it seeming impossible is what motivated him to pursue the project, he says.
"I'm a very competitive person," he says, "so I was motivated to try and disprove these articles and actually build this device."
His solution was to create a sintered solder paste material that was conductive enough and 3D printable. The material is made of a high-metal-load, low-melting-point paste, which is sintered into a thermoplastic scaffold. It was important for the metal in the paste to ensure reliability while its melting point would prevent damaging the scaffolding. The material is distributed from a 3D-printed paste extruder that consists of an outer case, a syringe, and a motor. This development allowed the printer to take shape.
Minnick says the opportunities the printer creates are plentiful. The main application he sees is the creation of a self-replicating spaceship. He says that the unsustainable cost of building spaceships is what prevents long-term space travel. Sending out several spacecraft is far more expensive than sending out one that could self-replicate.
Biopolymers—materials that can be grown rather than manufactured—could be used to build self-replicating factories, Minnick says. Constructing such facilities in underdeveloped countries would lower the cost of industrialization while speeding up processes.
RIP-CURRENT DETECTION SYSTEM
Boglarka Ecsedi, an incoming Georgia Tech freshman studying computer science, created a rip-current detection system that incorporates machine learning. The invention earned her $400.
Rip currents are dangerous phenomena that can pull swimmers into deep water. Because the currents generally can't be seen by the naked eye, it's hard to detect one, let alone avoid getting caught up in one. According to the U.S. Lifesaving Association, there are more than 100 rip current deaths each year.
Ecsedi says she hopes her project will raise awareness of rip currents among people who don't live near an ocean but vacation at one.
"A lot of people can be exposed to this danger," she says, "especially from landlocked countries, like Hungary," where she lives.
Ecsedi determined that using machine learning could help detect the treacherous currents. Her system uses an image-processing algorithm with a real-time deep neural network that filters through images and video of the water to detect rip currents. Her method has the ability to evolve with new data.
"A machine-learning approach is effective because we can build algorithms that are able to recognize this phenomenon on images and video recordings," she says.
She began working on the project while she was a research intern at the University of California, Santa Cruz. The city was an ideal place to tackle the project because of its beaches, she says.
After returning to Hungary, she had to turn to other methods to finish the model, however. Using Google Earth, she was able to gather and annotate more than 1,000 images of beaches in Africa, Australia, and North America.
Ecsedi says she hopes to develop a mobile application that will allow people to detect rip currents just by using their smartphone's camera.