Cummins Inc.
REINFORCEMENT LEARNING CONTROL OF VEHICLE SYSTEMS

Abstract:

A system includes a sensor array and a processing circuit. The processing circuit is operable to: store a policy; receive the sensor information from the sensor array; receive horizon information from a horizon system; input the sensor information and the horizon information into the policy; determine an output of the policy based on the input of the sensor information and the horizon information; control operation of a vehicle system according to the output; compare the sensor information received after controlling operation of the vehicle system according to the output relative to a reward or penalty condition; provide one of a reward signal or a penalty signal in response to the comparison; update the policy based on receipt of the reward signal or the penalty signal; and control the vehicle system using the updated policy to improve operation in view of the operating parameter.