The information of the road grade is an important input for advanced driver assistance systems to improve the vehicle ride comfort, safety, and fuel consumption. Current approaches for road grade estimation in the literature have various disadvantages, e.g., they lack in resolution and sample rate or use data from a lot of sensors, often not from series production. This paper presents methods, which are based on filters that combine the measurements from an inexpensive gyroscope, accelerometer, and magnetometer to estimate the orientation of the sensor relative to the earth’s surface. The methods are evaluated using high-resolution road grade data as reference, which were acquired with an aircraft and the light detection and ranging technique. The road grade information, calculated in the vehicle in real-time, could be transmitted to a central database and merged with information of other vehicles. Therefore, digital maps for advanced driver assistance systems could be kept updated in very short intervals with high-resolution road grade information.