Many great minds at companies like Google and Tesla have been laboring towards creating a self-driving car system. Millions of dollars have gone into the research, resulting in expensive prototypes and costly components. Nineteen-year-old Romanian student Ionut Budisteanu took the top prize at the Intel International Science and Engineering Fair for developing an autonomous system that would only cost $4,000 and safely guide a car through the streets.
Budisteanu’s goal was to use artificial intelligence to do away with Google’s high-resolution 3D radar. His cheaper low-resolution radar is able to recognize large objects while webcams mounted on the car can identify lane markings and curbs in real-time. The information is processed by artificial intelligence technology with on-board computers that an determine the safest route for the vehicle. He states that he has run his system without incident on 47 out 50 trials, although 3 tests failed to identify figures at a distance of 20-30 meters. He says that utilizing a slightly higher resolution radar would improve his rates of success and still cost much less than Google’s system.
Budisteanu won the top prize at the Intel International Science and Engineering Fair, an award that granted him $75,000. He has also attracted funding from a Romanian company that will work with him to test his prototype this year. With more than 2 million people killed in car accidents annually, 87 percent of which are caused by human error, autonomous driving technology could be the next big advancement in vehicle safety. A low-cost system like Budisteanu’s could bring driverless technology to the mass market and greatly expand its potential.