The Free Press Journal

Soon, self-driving cars can guess pedestrian movements

-

Scientists are using humans' gait, body symmetry and foot placement to teach self-driving cars to recognise and predict pedestrian movements with greater precision than current technologi­es.

Data collected by vehicles through cameras, LiDAR and global positionin­g system (GPS) allowed the researcher­s at the University of Michigan in the US to capture video snippets of humans in motion and then recreate them in three-dimensiona­l (3D) computer simulation. With that, they have created a ‘biomechani­cally inspired recurrent neural network’ that catalogs human movements.

The network can help predict poses and future locations for one or several pedestrian­s up to about 50 yards from the vehicle, at about the scale of a city intersecti­on. LiDAR is a surveying method that measures distance to a target by illuminati­ng the target with pulsed laser light and measuring the reflected pulses with a sensor.

“Prior work in this area has typically only looked at still images. It wasn’t really concerned with how people move in three dimensions,” said Ram Vasudevan, an assistant professor at the University of Michigan. “But if these vehicles are going to operate and interact in the real world, we need to make sure our prediction­s of where a pedestrian is going does not coincide with where the vehicle is going next,” said Vasudevan.

Equipping vehicles with the necessary predictive power requires the network to dive into the minutiae of human movement: the pace of a human’s gait (periodicit­y), the mirror symmetry of limbs, and the way in which foot placement affects stability during walking. Much of the machine learning used to bring autonomous technology to its current level has dealt with two dimensiona­l images — still photos. A computer shown several million photos of a stop sign will eventually come to recognise stop signs in the real world and in real time.

Newspapers in English

Newspapers from India