System and method for ride order dispatching

A system for ride order dispatching, comprising a processor and a non-transitory computer-readable storage medium storing instructions that, when executed by the processor, cause the system to perform a method, the method comprising:

• obtaining information on a location of a vehicle and a time to input into a trained second neural network algorithm;
• and
• based on a policy generated from the trained second neural network algorithm, obtaining action information for the vehicle, the action information comprising:
• staying at a current position of the vehicle, re-positioning the vehicle, or accepting a ride order, wherein:
• the second neural network algorithm is trained with one or more first network weights and with second ride order dispatching data associated with a second region in a second training to obtain one or more second network weights;
• the one or more first network weights are obtained from training a first neural network algorithm with first ride order dispatching data associated with a first region in a first training;
• the second neural network algorithm comprises an input layer, a first sequence of hidden layers, a second sequence of hidden layers, and an output layer;
• each of the hidden layers in the first sequence has an output correspondingly coupled to one of the hidden layers in the second sequence and the output layer;
• and
• for the input layer, the second ride order dispatching data comprises (1) one or more second state inputs for inputting into the second sequence of hidden layers and (2) one or more first state inputs for inputting into the first sequence of hidden layers.

A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform a ride order dispatching method, the method comprising:

• obtaining information on a location of a vehicle and a time to input to a trained second neural network algorithm;
• based on a policy generated from the trained second neural network algorithm, obtaining action information for the vehicle, the action information comprising:
• staying at a current position of the vehicle, re-positioning the vehicle, or accepting a ride order;
• and
• causing the action information to be played on a device associated with a driver of the vehicle, wherein:
• the second neural network algorithm is trained with one or more first network weights and with second ride order dispatching data associated with a second region in a second training to obtain one or more second network weights;
• the one or more first network weights are obtained from training a first neural network algorithm with first ride order dispatching data associated with a first region in a first training;
• the second neural network algorithm comprises an input layer, a first sequence of hidden layers, a second sequence of hidden layers, and an output layer;
• each of the hidden layers in the first sequence has an output correspondingly coupled to one of the hidden layers in the second sequence and the output layer;
• and
• for the input layer, the second ride order dispatching data comprises (1) one or more second state inputs for inputting into the second sequence of hidden layers and (2) one or more first state inputs for inputting into the first sequence of hidden layers.