Boston Consulting Group has studied self driving cars in depth, but was left with one question: can a self driving car in snow still operate in bad weather as it does on a dry, sunny day? In places such as Canada the weather conditions can vary from region to region, from fog and rain to whiteouts and snowstorms, which is why it is imperative that if they are to be mass-marketed that they be able to withstand such conditions.
Can Today’s Driverless Car Technology Handle Adverse Weather Conditions?
Tests of a self driving car in snow are difficult to find. Ontario has recently allowed for the testing of driverless cars, which may allow for more testing in snowy conditions.
Utilizing Hyundai’s test car as an example, the self driving car was considered to have performed relatively well in rainy, wet conditions during the manufacturer’s 2014 competition. However, there were a few minor incidents. One incident was a wide turn at the intersection that almost caused a fender bender. This was due to the fact that the car drove up onto the curb. Hyundai claims that this failure was caused by the camera settings, which allowed for the car to hit the curb. The camera settings also caused for a simple corner to be turned too sharply. Other issues included difficulty staying on course, pedestrian detection and road signs, which were all accounted for during the dry, sunny weather.
Another example of a self driving car in snow is a 2015 Kia Sedona equipped with driverless features. Senior analyst for Navigant Research, Sam Abuelsamid, tested this in the snow in Michigan. Features such as Adaptive Cruise Control, which uses radar sensing technology, performed well during both dry and snowy conditions. The same radar sensor is utilized for rear cross traffic detection, which monitors the vehicle’s surroundings for pedestrians and other vehicles when reversing. Blind-spot monitoring also uses radar sensors. Although the 2015 Kia Sedona was equipped with many advanced autonomous driving systems, Abuelsamid found that many of those advanced features were neutralized by the weather. The radar sensor that is featured in front of the vehicle and rear camera were covered in snow. Abuelsamid also noted that he had to turn off the vehicle’s Park Assist technology due to the snow triggering the sensors, which continued to beep. The vehicle’s optical cameras which account for features such as lane departure warning and reversing were also compromised.
Self Driving Cars in Snow: the Future
Bringing self driving car technology to a country like Canada will be one of the most difficult hurdles. Despite the advancements of Google’s driverless car, the self driving car in snow we can assume would have been equally as compromised as the Sedona, as it relies on LIDAR sensors. Ford has recently teamed up with Google to help combat adverse weather conditions. Announced at the Detroit Auto Show, Ford confirmed it has launched Project Snowtonomy in Michigan. Project Snowtonomy consists of a 32-acre fake city (dubbed Mcity) to test self driving cars in snow. Currently, Ford is using LIDAR sensors to detect its surroundings, which send out short pulses of laser light. These impulses create a real time, high definition 3D map of the area around it. These sensors can read and interpret lane markings; however, not if they are covered by snow.
Google’s December 2015 report states that “to explore even more challenging environments, we’re beginning to collect data in all sorts of rainy and snowy conditions as we work toward the goal of a self-driving car what will be able to drive come rain, hail, snow or shine.”
One solution in curving the performance of a self driving car in snow is through wireless communication between vehicles and infrastructure. This can be seen in vehicle-to-everything (V2X) communication technology, which combines vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication technology. The system uses Dedicated Short-Range Communications (DSRC) devices, which is a 2-way wireless communication system that works with short ranges. The system was specifically designed for the automotive industry. It is essentially a wi-fi system for vehicles to communicate with one another.
In order for self driving cars to be well received across the globe, it will have to be able to withstand adverse weather conditions. If a self driving car in snow can still function the same way that it does in ideal conditions, then there is no doubt that it is the way of the future.