Ford is embedding its driverless vehicles with LiDAR sensors, which use lasers to scan the surroundings, to allow them to drive in snowy conditions without slipping.
The LiDAR system is a surveying technology that measures distance by illuminating a target with a laser light.
The driverless car’s in-built computer uses the information from four LiDAR scanners to create a high resolution 3D map of the area around it.
The technology is so accurate that it can detect individual snowflakes and raindrops in the air and the carmaker has created an algorithm that recognises this and filters them out of the car’s vision, so that it does not alter its course based on precipitation.
While mapping their environment, Ford’s autonomous vehicles collect and process a wide set of data about the road and surrounding landmarks including signs, buildings, trees and other features.
The car’s sensors collect about 600 gigabytes of data per hour, which it uses to create the 3D maps.
Ford says the system is needed for the vehicles to find their way because GPS is only accurate to about four metres, which is not enough for accident-free autonomous vehicle operation.
By scanning their environment for landmarks, then comparing that information to the 3D digital maps stored in their databanks, Ford’s autonomous vehicles can precisely locate themselves to within a centimetre.
In addition to LiDAR sensors, the driverless system uses cameras and radar to monitor the environment around the vehicle, with the data generated from all of those sensors fused together to form a complete picture.
This process results in robust 360-degree situational awareness. Sensor fusion means that one inactive sensor, which can be caused by ice, snow, grime or debris build-up on a sensor lens, does not necessarily hinder autonomous driving.
Ford’s system is able to identify the deterioration of sensor performance, which helps keep sensors in ideal working order. Eventually, the carmaker believes its vehicles will be able to handle ice and grime accumulation themselves through self-cleaning or defogging measures.
In January, Ford said it was planning to use a new, lower cost LiDAR sensor made by California-based Velodyne in its new driverless vehicles.
According to industry executives, the high cost of LiDAR sensors is one of the main technical obstacles to widespread commercialisation of self-driving vehicles.
Ford said it would be the first driverless manufacturer to use Velodyne's new solid state ‘hybrid Ultra PUCK Auto’ sensor. The compact device replaces the spinning scanners mounted on the rooftops of some autonomous test vehicles.
Two of the new sensors can reportedly replace four LiDAR sensors and the Ultra PUCK is small enough to mount on a side-view mirror.
Google’s director of its self-driving car project, along with auto industry executives, are set to testify before the American Congress next week on efforts to develop safe and effective autonomous cars.
The committee wants witness views "on the appropriate role of government in promoting innovation including removing unnecessary hurdles, and their strategy to grow consumer adoption of this new technology."
Major automakers and technology companies have complained that state and federal safety rules are impeding testing and ultimate deployment of driverless vehicles.
BMW recently said it was boosting its efforts to develop the technology, with plans to recruit thousands of software developers to combat Google’s advances in the field.