- Given GPS tracking data of many trips belong to different drivers, different type of vehicles
- Each GPC point recorded after 10 seconds
- I can map matching the GPS data back to the roads sections
- Right now I use simple calculation of GPS points density and calculate avg speed for a road section belong to what time of the day, what day of the week
But some road section, same time (say 18:00), same day but maybe very different speed, some road section just have few trips data. And I'm looking for a better way to calculate and predict speed of a (road section, datetime).
- Can AI help to calculate the speed of road section that I already have data better?
- Can AI help to predict the speed of road section I don't have historical data? Given assumption that speed of related road section can be related with the other.
I'm a total newbie and have zero knowledge of AI/ML, I hope you can give me more details. For example what kind of algorithm fit my data, if any. (And maybe a Google Colab with simple script for starting point).