January 13, 2016 at 13:07Ford teaches unmanned cars to ride in the snow
At the Detroit auto show, Ford talked about starting testing autonomous cars in snow. Difficult weather conditions such as heavy rains, snowfall and hail are now one of the main unresolved technical problems that prevent unmanned vehicles from entering the global market.
Autonomous cars use the technology for obtaining and processing information about LIDAR objects - Light Identification Detection and Ranging. To do this, special sensors are installed on the car, they can be seen on the roof of the vehicle - they are constantly spinning and make an environment map in high resolution.
Snow is an obstacle for these sensors. Ford also had to work on this issue because the test sitelocated in Michigan, where for several months a year autonomous cars that are not adapted to bad weather conditions cannot be on the road.
Ford in the car used not one, but 4 lidars at once. They work simultaneously to provide data to a powerful computer that maps the environment. But this is not enough: the computer works with an already compiled map, created before all the road markings and signs were not covered with snow. Ford's goal is to train the car in poor conditions, just like an experienced driver does.
LIDAR technology has an alternative. Volkswagen and GM partnered with Mobileye to work together on technology using the mobile Internet and sensors that are already installed on many cars. Tesla, which has released a beta version of autopilot , does something similar. Mobileye's customers already include Honda, BMW, Huyndai and Tesla.
Google also began testing unmanned cars in bad weather, including developers installing miniature wipers on sensors located on the roof of the car.
Autonomous cars use the technology for obtaining and processing information about LIDAR objects - Light Identification Detection and Ranging. To do this, special sensors are installed on the car, they can be seen on the roof of the vehicle - they are constantly spinning and make an environment map in high resolution.
Snow is an obstacle for these sensors. Ford also had to work on this issue because the test sitelocated in Michigan, where for several months a year autonomous cars that are not adapted to bad weather conditions cannot be on the road.
Ford in the car used not one, but 4 lidars at once. They work simultaneously to provide data to a powerful computer that maps the environment. But this is not enough: the computer works with an already compiled map, created before all the road markings and signs were not covered with snow. Ford's goal is to train the car in poor conditions, just like an experienced driver does.
LIDAR technology has an alternative. Volkswagen and GM partnered with Mobileye to work together on technology using the mobile Internet and sensors that are already installed on many cars. Tesla, which has released a beta version of autopilot , does something similar. Mobileye's customers already include Honda, BMW, Huyndai and Tesla.
Google also began testing unmanned cars in bad weather, including developers installing miniature wipers on sensors located on the roof of the car.