Real-time vehicle state trajectory prediction for vehicle energy management and autonomous drive
Systems and methods described herein are directed to predicting velocity and trajectory of a vehicle based on a route to be navigated for a vehicle, thereby estimating the vehicle drive cycle, for one or more segments of the route. The predicted velocity trajectory is determined from a machine learning function that takes into account historical information from vehicle routes traveled by other vehicles.
1. A vehicle, comprising:
a processor, configured to:
determine an expected velocity for the vehicle for one or more portions of a route, the expected velocity determined from a velocity trajectory estimation function constructed from a machine learning process and configured to determine the expected velocity for the vehicle for the one or more portions of the route;
modify the expected velocity for the one or more portions of the route based on data received from one or more sensors to correct error in the expected velocity determined from the velocity trajectory estimation function; and
control an after-treatment system of the vehicle based on the modified expected velocity for the one or more portions of the route; and
provide the data received from the one or more sensors to correct error in the expected velocity to the machine learning process.
2. The vehicle of claim 1 , wherein the velocity trajectory estimation function is configured to determine the expected velocity for the vehicle based on a global positioning satellite route (GPS) input to the vehicle and historical trip information, wherein the data received from the one or more sensors comprises traffic information along the route.
3. The vehicle of claim 1 , wherein the processor is configured to:
construct the velocity trajectory estimation function to determine the expected velocity for the vehicle for the one or more portions of the route from the machine learning process, the machine learning process configured to:
process as input, historical global positioning satellite (GPS) information and historical traffic information for one or more historical routes; and
generate the velocity trajectory estimation function to be configured to process a route of the vehicle as input to determine the expected velocity.
4. The vehicle of claim 1 , wherein the processor is configured to:
receive vehicle platooning instructions for a portion of the one or more portions of the route based on the modified expected velocity; and
control the vehicle according to the vehicle platooning instructions for the portion of the one or more portions of the route based on the modified expected velocity and a trajectory determined for the portion of the one or more portions of the route.
5. The vehicle of claim 1 , wherein the processor is configured to:
determine exhaust gas production for an after-treatment system for the one or more portions of the route based on the modified expected velocity; and
control the after-treatment system of the vehicle to meet vehicle driving performance parameters and minimize fuel consumption while maintaining the exhaust gas production for the one or more portions of the route.
6. The vehicle of claim 1 , wherein the processor is configured to:
control at least one of a gear shift and a gear ratio of the vehicle for the one or more portions of the route based on the modified expected velocity.