• quarter car model
• Mahony filter
• Kalman filter
• roughness estimation
• real time application
• international roughness index

Road roughness measurement is an important procedure for transport authorities.
Knowledge of the road roughness also provides useful information for active suspension control, that could have the capability to adapt their parameters depending on the road the vehicle is driving on.
The aim of this Thesis was to develop a procedure to identify the road surface roughness, with a real time algorithm, associating the road roughness with a simple numerical index.
Minimising the number of sensors was a constraint.
A model based approach was used, therefore the vertical accelerations were the data of interest, and they were measured by a cheap IMU.
If the sensor is mounted with an unknown orientation, the measured accelerations must be suitably rotated.
For this reason the estimation of the unknown IMU orientation has been investigated.
A passive Mahony filter was used to deal with the unknown IMU orientation.
A Kalman filter has been developed to estimate the required variables for the IRI computation.
Actually, two different Kalman filter have been developed to estimate the required variables for the IRI computation.
They differ in the number of IMU required.
The whole problem was first analysed in a simulated environment.
After satisfying results obtained in simulations, the develop of a real time application has been made.
Several real experiments were carried out, testing the algorithm with a single IMU.
Accelerations with different road surface were registered and analyzed.