Spatio-temporal track density shaping

A non-transitory machine-readable medium comprising instructions that when executed by a processor executes a process comprising:

• receiving into a computer processor a first dataset of vehicle trajectories comprising a plurality of individual vehicle tracks forming a first track density;
• receiving into the computer processor a vehicle track density profile;
• receiving into the computer processor a target spatio-temporal coverage metric that is an indication of a user-specified vehicle track density profile;
• receiving into the computer processor a track model comprising a plurality of Heaviside functions encoding track time origins and durations for the plurality of individual vehicle tracks and comprising locations for the plurality of individual vehicle tracks;
• minimizing an approximation of the Heaviside functions as a function of the target spatio-temporal coverage metric;
• optimizing the first dataset of vehicle trajectories as a function of the vehicle track density profile and the minimized approximation of the Heaviside functions;
• reshaping each individual vehicle track start time and start location in the first dataset of vehicle trajectories as a function of the optimizing, thereby generating a second dataset of vehicle trajectories having a second track density;
• and
• training a vehicle resources machine learning algorithm using the second dataset of vehicle trajectories.

A system comprising:

• a computer processor;
• and
• a computer memory coupled to the computer processor;
• wherein the computer processor and the computer memory are operable for:
• receiving into the computer processor a first dataset of vehicle trajectories comprising a plurality of individual vehicle tracks forming a first track density;
• receiving into the computer processor a vehicle track density profile;
• receiving into the computer processor a target spatio-temporal coverage metric that is an indication of a user-specified vehicle track density profile:
• receiving into the computer processor a track model comprising a plurality of Heaviside functions encoding track time origins and durations for the plurality of individual vehicle tracks and comprising locations for the plurality of individual vehicle tracks;
• minimizing an approximation of the Heaviside functions as a function of the target spatio-temporal coverage metric;
• optimizing the first dataset of vehicle trajectories as a function of the vehicle track density profile and the minimized approximation of the Heaviside functions;
• reshaping each individual vehicle track start time and start location in the first dataset of vehicle trajectories as a function of the optimizing, thereby generating a second dataset of vehicle trajectories having a second track density;
• and
• training a vehicle resources machine learning algorithm using the second dataset of vehicle trajectories.