Computing device and method for facilitating modification of the route of a vehicle

A computing device configured to facilitate modification of a route of a vehicle, the computing device comprising processing circuitry configured to:

• receive an input image of at least one surface extending along the route of the vehicle
• construct a stylized image of the input image, wherein the stylized image is an enhanced version of the input image representing different surfaces found in the input image as geometric objects with a texture associated with each geometric object, wherein the stylized image distinguishes between surfaces having different textures that are present in the input image;
• based upon the stylized image, determine a frictional coefficient associated with at least some of the surfaces having different textures;
• and
• determine whether the route of the vehicle is to be altered from a predefined route based upon the frictional coefficient that has been determined to be associated with a surface over which the predefined route of the vehicle extends.

A method for training machine learning models to facilitate modification of a route of a vehicle, the method comprising:

• based upon a plurality of training images depicting surfaces having different, known textures, training a first machine learning model to distinguish between surfaces having different textures and to construct a stylized image from a respective training image that distinguishes between surfaces having different textures that are present in the respective training image;
• storing the first machine learning model;
• based upon a plurality of stylized images that distinguish between surfaces having different textures and also based upon predefined frictional coefficients associated with surfaces having different textures, training a second machine learning model to determine a frictional coefficient associated with at least some of the surfaces having different textures;
• and
• storing the second machine learning model.