Method and system for dynamically predicting driving range of vehicles

A system for dynamically predicting driving range of vehicles, the device comprising:

• a processor;
• and
• a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, causes the processor to:
• dynamically determine a set of real-time values corresponding to a plurality of in-transit parameters associated with a vehicle, wherein the plurality of in-transit parameters comprises a set of external parameters and a set of vehicle parameters;
• train a ML model based on a set of pre-defined parameters associated with a target vehicle, wherein training the ML model comprises:
• operating the target vehicle on a given path with a pre-determined value for each of the set of pre-defined parameters;
• predicting, by the ML model, a driving range for the target vehicle based on the pre-determined value of each of the set of pre-defined parameters;
• determining an accuracy factor of the driving range predicted for the target vehicle by comparing the predicted driving range with an actual driving range of the target vehicle;
• and
• optimizing the ML model to predict the actual driving range for the target vehicle based on the determined accuracy factor;
• determine, via a trained Machine Learning (ML) model, a variance in the plurality of in-transit parameters when compared with the set of pre-defined parameters, wherein determining the variance comprises:
• identifying an overlapping subset of parameters between the plurality of in-transit parameters and the set of pre-defined parameters;
• identifying a non-overlapping subset of parameters between the plurality of in-transit parameters and the set of pre-defined parameters;
• determining a difference in each of the real-time values determined for the overlapping subset of parameters with a corresponding optimal value;
• and
• computing the variance based on the non-overlapping subset of parameters and the difference in real-time values determined for the overlapping subset of parameters;
• determine, by the range prediction device via the trained ML model, a percentage deviation from an absolute driving range associated with the target vehicle based on the determined variance;
• and
• predict, by the range prediction device via the trained ML model, a current driving range for the vehicle based on the identified percentage deviation.

A non-transitory computer-readable medium storing computer-executable instructions for dynamically predicting driving range of vehicles, the computer-executable instructions configured for:

• dynamically determining a set of real-time values corresponding to a plurality of in-transit parameters associated with a vehicle, wherein the plurality of in-transit parameters comprises a set of external parameters and a set of vehicle parameters;
• training a ML model based on a set of pre-defined parameters associated with a target vehicle, wherein training the ML model comprises:
• operating the target vehicle on a given path with a pre-determined value for each of the set of pre-defined parameters;
• predicting, by the ML model, a driving range for the target vehicle based on the pre-determined value of each of the set of pre-defined parameters;
• determining an accuracy factor of the driving range predicted for the target vehicle by comparing the predicted driving range with an actual driving range of the target vehicle;
• and
• optimizing the ML model to predict the actual driving range for the target vehicle based on the determined accuracy factor;
• determining, via a trained Machine Learning (ML) model, a variance in the plurality of in-transit parameters when compared with the set of pre-defined parameters, wherein determining the variance comprises:
• identifying an overlapping subset of parameters between the plurality of in-transit parameters and the set of pre-defined parameters;
• identifying a non-overlapping subset of parameters between the plurality of in-transit parameters and the set of pre-defined parameters;
• determining a difference in each of the real-time values determined for the overlapping subset of parameters with a corresponding optimal value;
• and
• computing the variance based on the non-overlapping subset of parameters and the difference in real-time values determined for the overlapping subset of parameters;
• determining, via the trained ML model, a percentage deviation from an absolute driving range associated with the target vehicle based on the determined variance;
• and
• predicting, via the trained ML model, a current driving range for the vehicle based on the identified percentage deviation.