Ford’s Research to Advance Vision of Autonomous VehiclesFord’s Research to Advance Vision of Autonomous Vehicles
Researchers and the automaker have come up with a method to inform an AV which cameras to use during navigation
August 15, 2022
Australian robotics experts working in tandem with Ford have produced a study that could herald significant advances in autonomous driving.
Researchers from the robotics department at Queensland University (QUT) and the automaker have come up with a method to inform an autonomous vehicle (AV) of which cameras to use during navigation.
The findings stemmed from a project that was created to look at how cameras and lidar sensors, which help AVs deliver their automated functionality, can better understand the world around them.
Professor Michael Milford, Joint Director of the QUT Centre for Robotics and Australian Research Council Laureate Fellow, is the senior author of the research and explained its significance. “The key idea here is to learn which cameras to use at different locations in the world, based on previous experience at that location,” he said.
“For example, the system might learn that a particular camera is very useful for tracking the position of the vehicle on a particular stretch of road and choose to use that camera on subsequent visits to that section of road.”
This capability has the potential to bring additional safety benefits to AV users in the future.
“Autonomous vehicles depend heavily on knowing where they are in the world,” said Dr. Punarjay (Jay) Chakravarty, who led the project on behalf of the Ford Autonomous Vehicle Future Tech group. “Knowing where you are helps you leverage map information that is also useful for detecting other dynamic objects in the scene. A particular intersection might have people crossing in a certain way. This can be used as prior information for the neural nets doing object detection and so accurate localization is critical and this research allows us to focus on the best camera at any given time.”
As part of the research, the team had to devise new ways of evaluating the performance of an AV’s positioning system. This was done by not only assessing performance when the system performed well but also investigating worst-case scenarios.
The full study has just been published in the Institute of Electrical and Electronics Engineers Robotics and Automation Letters journal and will also be presented at a conference on Intelligent Robots and Systems in Kyoto, Japan in October.
As part of its research program into driverless tech, Ford’s Autonomous Vehicle Future Tech team has teamed with a number of different partners. Earlier this year, for example, it was involved in a pilot program that saw self-driving shuttles – operated by Ford subsidiary Quantum Signal AI – deliver food in Detroit.
About the Author(s)
You May Also Like