Connect with us



Safety boost for self-driving cars

Researchers in America have developed a method for testing the perception algorithms that enable autonomous cars to ‘understand’ what they see.

A team from the University of Southern California (USC) and Arizona State University has pioneered a ‘mathematical method’ that picks up ‘bugs in the systems’ in advance of the cars being put on the road. This is a vital step in getting the autonomous car revolution back on track, after some well publicised fatalities and accidents.

“Making perception algorithms robust is one of the foremost challenges for autonomous systems,” said the study’s lead author Anand Balakrishnan, a USC computer science PhD student. “Using this method, developers can narrow in on errors in the perception algorithms much faster and use this information to further train the system. The same way cars have to go through crash tests to ensure safety, this method offers a pre-emptive test to catch errors in autonomous systems.”

Newsletter Signup

Staff reporter
Written By


coronavirus attacking planet coronavirus attacking planet

Coronavirus ‘cure’ a step closer


ventilator to treat COVID-19 ventilator to treat COVID-19

COVID-19 ventilator challenge


MIT COVID-19 virtual hackathon


COVID-19 global online hackathon


Newsletter Signup