Today, a week after its cars on public roads and days after it raised in capital, Waymo took the wraps off of an AI model it claims “significantly” improved its driverless systems’ ability to predict pedestrians’, cyclists’, and drivers’ behaviors. It’s called VectorNet, and it ostensibly provides more accurate projections while requiring less compute compared with previous approaches.

Anticipating road agents’ future positions is table stakes for driverless cars, which by definition must navigate challenging environments without any human supervision. As by the March 2018 collision involving an autonomous Uber vehicle and a bicyclist, perception is critical. Without it, self-driving cars can’t reliably make decisions about how to respond in familiar — or unfamiliar — scenarios.

VectorNet aims to help predict the movements of road

Read More At Article Source | Article Attribution