Method and system for predicting energy consumption of a vehicle through application of a statistical model utilizing environmental and road condition information

Patent metadata

Field	Value
Publication number	US-11719753-B2
Application number	US-201715793390-A
Country	US
Kind code	B2
Filing date	Oct 25, 2017
Priority date	Feb 27, 2014
Publication date	Aug 8, 2023
Grant date	Aug 8, 2023

Abstract

A method for predicting energy consumption of a vehicle using a statistical model. The method includes (i) predicting a set of future input vectors for the vehicle at defined time intervals corresponding to a plurality of future points in time based on a subset of a plurality of reference input vectors previously generated at the defined time intervals at a plurality of previous points in time, (ii) predicting a change in the energy level of the vehicle using a processor and the statistical model, and (iii) providing results corresponding to the predicted change in the energy level to an output unit of the vehicle. Each reference input vector comprises a vehicle input vector and a database input vector associated with each point in time of the plurality of previous points in time. The database input vector for each defined time interval may be based on at least one of a plurality of environmental data and information about a road condition.

First claim

The invention claimed is:

1. A method performed by one or more processors for predicting energy consumption of a vehicle using a statistical model and indicating, one of visually and audibly, results corresponding to a predicted change in energy level to a driver of the vehicle, said method comprising:
predicting a set of future input vectors for said vehicle at defined time intervals corresponding to a plurality of future points in time based on a subset of a plurality of reference input vectors previously generated at the defined time intervals at a plurality of previous points in time, the subset including a most recent reference input vector, wherein an energy level of said vehicle is associated with each reference input vector of said plurality of reference input vectors at each previous point in time of said plurality of previous points in time, each reference input vector comprises a vehicle input vector and a database input vector associated with each point in time of said plurality of previous points in time, said vehicle input vector for each defined time interval is based at least in part on one or more vehicle operating parameters, and said database input vector for each defined time interval is based at least in part on a plurality of environmental data and information about a road condition;
predicting a change in said energy level of said vehicle using a processor and said statistical model, wherein (i) the change in said energy level comprises a function of corresponding future input vectors and an associated weighting vector, said weighting vector having been derived using said plurality of reference input vectors and associated energy levels at each point in time of said plurality of previous points in time, and representing an overall effect of each input vector on energy consumption of said vehicle, and (ii) said change in said energy level is predicted based on a regression analysis of said energy level associated with each of said reference input vectors; and
providing results corresponding to the predicted change in said energy level to an audio-video output unit of said vehicle and providing the results, using the audio-video output unit, to a driver of the vehicle;
wherein each vehicle input vector comprises a plurality of sensor data and each database input vector comprises a plurality of database data;
wherein said statistical model comprises at least one of a linear function and a periodic function of at least one of a stored energy of the vehicle at each point in time, each vehicle input vector, and each database input vector for each defined time interval, and wherein the statistical model is generated using machine learning based on an equation for each input vector; and
wherein said plurality of sensor data is captured for a vehicle location environment, a vehicle equipment profile, and a driver behavior profile at each point in time of said plurality of previous points in time.

2. The method according to claim 1 , wherein said plurality of sensor data is obtained from a plurality of sensors associated with said vehicle, and wherein the statistical model models one of the energy level of the vehicle and the change of the energy level of the vehicle as a function of the sensor data, vehicle performance data, weather data, traffic data, and road condition data.

3. The method according to claim 2 , wherein said plurality of sensors correspond to at least one of a regenerative braking sensor, a solar radiation sensor, a weight sensor, and an identity sensor.

4. The method according to claim 1 , wherein said plurality of database data is obtained for a plurality of vehicles and comprises data related to a vehicle location environment, a vehicle equipment profile, a driver behavior profile, average speed, acceleration, mileage, battery capacity, and fuel consumption.

5. The method according to claim 4 , wherein said plurality of database data is obtained from a database storing previously recorded data for said vehicle location environment, said vehicle equipment profile, and said driver behavior profile corresponding to said plurality of vehicles, and wherein the method facilitates predicting a cost effectiveness of adding one of a power saving feature and a power generating feature to the vehicle based on the driver behavior profile.

6. The method according to claim 1 , wherein said plurality of sensor data corresponds to time data, day data, wind direction data, driving pattern data, driver identity, velocity data, acceleration data, inclination data, and angular momentum data.

7. The method according to claim 1 , wherein said plurality of database data corresponds to weather data, route data, traffic data, and driving pattern data of a plurality of drivers, and wherein the method includes predicting a most energy-efficient route by executing the statistical model over a plurality of possible routes and predicting an amount of power consumed for each possible route.

8. The method of claim 1 , wherein each reference input vector is associated with each of the previous points in time, and wherein the processor includes a plurality of data aggregation modules to receive data and deliver the data to a time series database which stores the data as the reference input vectors along with corresponding energy levels for each previous point in time.

9. The method of claim 1 , further comprising capturing an energy level associated with each reference input vector at each of the previous points in time using an energy meter that captures at least one of a stored battery power and a stored fuel level.

10. The method of claim 1 , wherein the weighting vector is derived using a linear regression.

11. The method of claim 1 , further comprising, using an optimization engine, capturing an actual change in energy level for each of the future points in time based on the energy level associated with each reference input vector, computing a difference between the predicted change in energy level and the actual change in energy level, and refining the weighting vector to minimize the difference between the predicted change in energy level and the actual change in energy level.

12. The method of claim 11 , wherein refining the weighting vector comprises modifying a value of the weighting vector, and wherein the statistical model is refit in response to refining the weighting vector.

13. A system for predicting energy consumption of a vehicle using a statistical model and for displaying results corresponding to a predicted change in energy level to a driver of the vehicle, said system comprising:
an acquisition module that obtains a plurality of reference input vectors at defined time intervals at a plurality of points in time, wherein each input vector comprises a vehicle input vector and a database input vector associated with each point in time of said plurality of points in time, said vehicle input vector is generated for each defined time interval based at least in part on one or more vehicle operating parameters, and said database input vector is generated for each defined time interval based at least in part on a plurality of environmental data and information about a road condition;
a processor that predicts (i) a set of future input vectors for said vehicle at defined time intervals corresponding to a plurality of future points in time based on a subset of a plurality of reference input vectors previously obtained at the defined time intervals at a plurality of previous points in time, wherein an energy level of said vehicle is associated with each reference input vector of said plurality of reference input vectors at each previous point in time of said plur

Assignees

• Invently Automotive Inc

Inventors

• Brown Stephen J

Classifications

• G01R31/3648 Primary

  comprising digital calculation means, e.g. for performing an algorithm · CPC title

• B60W30/0956

  the prediction being responsive to traffic or environmental parameters · CPC title

• B60W50/0097

  Predicting future conditions · CPC title

• G01C21/3469

  Fuel consumption; Energy use; Emission aspects · CPC title

• G01L5/13

  for measuring the tractive or propulsive power of vehicles · CPC title