Self-Driving A.I. Could Save Lives By Predicting the Worst

Self-Driving A.I. Could Save Lives By Predicting the Worst

Call it fatalistic, pessimistic, or just really, really, smart, but a new self-driving car algorithm developed by researchers at Germany’s Technical University of Munich (TUM) thrives on thinking about the worst thing that could happen at every moment. And then figuring out how to get out of it without endangering or obstructing traffic.

“Current autonomous driving systems usually incorporate most-likely evolutions of a traffic scenario, [such as] the preceding vehicle will most likely accelerate,” Christian Pek, a researcher in the university’s researcher in the Cyber-Physical Systems Group, told Digital Trends. “However, this design might result in unsafe behaviors if traffic participants behave differently than expected — for example, [if instead] the preceding vehicle decelerates. Our algorithm addresses this problem by computing all possible future evolutions of the scenario by considering all possible motions of other traffic participants that are compliant with traffic rules. As a result, we are able to ensure that decisions are safe regardless of the future legal motion of other traffic participants.”

The algorithm works by evaluating vehicle sensor data every millisecond to extrapolate potential behavior up to six seconds into the future. This is something that good human drivers do almost unconsciously, but which proves difficult for machines to emulate. Based on the scenarios this new self-driving car system comes up with, it then works out what emergency maneuvers it would need to perform so as not to endanger others or cause collisions. Think of it like Asimov’s Laws of Robotics, self-driving car edition.

This traffic situation forecasting has been deemed too time-consuming in the past. But the team at Munich have shown that it can work, using simplified dynamic models and reachability analysis to figure out future positions that cars and pedestrians might take.

“Our software serves as a safety layer for motion planning and verifies whether decisions of the autonomous vehicles are safe during its operation,” Stefanie Manzinger, a Ph.D. student in the Cyber-Physical Systems Group, told Digital Trends. “In emergency situations, our safety layer stops the autonomous vehicle in dedicated safe areas.”

According to Pek, the team demonstrated the safety benefits and performance of its algorithm on real traffic data recorded with a test vehicle in Munich. “Our scenarios correspond to critical situations, for example, turning left at an intersection with oncoming traffic,” Pek said. “Our results show that our algorithm safeguards the autonomous vehicles in these situations without performance loss. Following this proof-of-concept, our next step is to test our algorithm in more situations together with partners.”

A paper describing the work was recently published in the journal Nature Machine Intelligence.

Editors’ Recommendations




Source Article