Corso di Laurea Magistrale Interateneo in Geografia e Scienze Territoriali (LM-80)

Given a series of GPS (timestamp, latitude and longitude) tuples, accurately detect multiple transportation modes, i.e., walking, running, bicycling, car, metro or train, boat, airplane, or being stationary. Available in most modern smartphones, accelerometers and GPS receivers can be used for identify transportation modes. However, because of the high sampling rate required to accurately recognize transportation modes using accelerometers, the required data quickly becomes too big to store and transfer through mobile networks. Compared to accelerometers, GPS devices require a much lower sample rate and become a better choice when the recognition doesn’t need to happen in real time. The designed algorithm will first process the GPS dataset to:

With the above data, it will then segment the dataset and identify the transportation modes used in each segment (probably using Machine Learning techniques).

 

Tesi per studenti della LM in Geografia con specializzazione in Geoinformatica.