Method and system for predicting energy consumption of a vehicle using a statistical model

• US 9,841,463 B2
• Filed: 02/27/2014
• Issued: 12/12/2017
• Est. Priority Date: 02/27/2014
• Status: Active Grant

• Current Assignee: iQar Inc.
• Original Assignee: Invently Automotive Inc. (iQar Inc.)
• Inventors: Brown, Stephen J.
• Primary Examiner(s): Nguyen, Lam
• Application Number:
  US14/192,645
• Publication Number:
  US 20150241310A1
• Time in Patent Office: 1,384 Days
• Field of Search: 701 3, 701123
• CPC Class Codes: B60W 30/0956 the prediction being responsive to traffic or environmental parameters
• CPC Class Codes: B60W 50/0097 Predicting future conditions
• CPC Class Codes: G01C 21/3469 Fuel consumption; Energy use; Emission aspects
• CPC Class Codes: G01L 17/00 Devices or apparatus for measuring tyre pressure or the pressure in other inflated bodies
• CPC Class Codes: G01L 3/26 Devices for measuring efficiency, i.e. the ratio of power output to power input
• CPC Class Codes: G01L 5/13 for measuring the tractive or propulsive power of vehicles
• CPC Class Codes: G01R 31/3648 comprising digital calculation means, e.g. for performing an algorithm
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05D 1/0214 in accordance with safety or protection criteria, e.g. avoiding hazardous areas monitoring the location of vehicles within a certain area, e.g. forbidden or allowed areas, in traffic control systems for road vehicles G08G1/13
• CPC Class Codes: G05D 1/0221 involving a learning process
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G08G 1/0104 Measuring and analyzing of parameters relative to traffic conditions
• CPC Class Codes: G08G 1/0125 Traffic data processing
• CPC Class Codes: G08G 1/0129 for creating historical data or processing based on historical data

Abstract

A method and system includes predicting energy consumption of a vehicle using a statistical model. The method includes obtaining a plurality of input vectors for plurality of points in time, wherein each input vector includes a plurality of variables with a weight vector. Thereafter, the energy level for each input vector is captured for each point in time. Subsequent to capturing the energy level, the method includes predicting a change in energy level of the vehicle using the statistical model.

Claims

1. A method for predicting energy consumption of a vehicle using a statistical model, said method comprising:

obtaining a plurality of input vectors for said vehicle at defined time intervals at a plurality of points in time, wherein each input vector is associated with each point in time of said plurality of points in time;

capturing an energy level associated with each input vector of said plurality of input vectors at each point in time for said vehicle, wherein said energy level corresponds to at least one of a stored battery power and a stored fuel level of said vehicle;

predicting a change in said energy level using a processor and said statistical model, wherein (i) the change in said energy level comprises a function of corresponding input vectors and an associated weight vector, (ii) said weight vector is derived using said plurality of input vectors and associated energy levels at each point in time of said plurality of points in time, and represents an overall effect of each said input vector on energy consumption of said vehicle, and (iii) said change in said energy level is predicted through a regression analysis of said energy level associated with each said input vector; and

providing results corresponding to the predicted change in said energy level to an audio-video output unit of said vehicle.

2. The method of claim 1, wherein said weight vector associated with said input vector is derived using a linear regression that derives said weight vector based on said plurality of input vectors and respective energy levels at said plurality of points in time.

3. The method of claim 2, further comprising:

predicting a set of input vectors at defined time intervals at a plurality of future points in time based on a subset of said plurality of input vectors generated at said defined time intervals, at said plurality of points in time, wherein said subset of said plurality of input vectors represents the most recent input vectors of said vehicle;

deriving a change in said energy level for said plurality of future points in time using said statistical model, wherein said change in said energy level is derived by adding a change in energy level for each defined time interval;

capturing an actual change in energy level for each point in time of said plurality of future points in time, wherein said actual change in said energy level is based on said energy level of said vehicle associated with each input vector corresponding to each point in time;

computing a difference between the derived change in said energy level and said actual change in said energy level; and

refining said weight vector for minimizing the difference between the derived change in said energy level and said actual change in said energy level, wherein refining said weight vector comprises modifying the value of the weight vector to minimize the difference, wherein said statistical model is refit in response to the refined weight vector.

4. The method of claim 1, wherein said each input vector comprises a plurality of sensor data and a plurality of database data, wherein said plurality of sensor data is captured for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile at each point in time of said plurality of points in time, wherein said plurality of sensor data is obtained from a plurality of sensors coupled to said vehicle, wherein said plurality of database data is obtained for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile for a plurality of vehicles, and wherein said plurality of database data is obtained from a database storing previously recorded data for at least one of said vehicle location environment, said vehicle equipment profile, and said driver behavior profile corresponding to said plurality of vehicles.

5. The method of claim 4, wherein said plurality of sensor data correspond to at least one of location data, time data, day data, solar radiation data, temperature data, humidity data, barometric pressure data, wind speed data, wind direction data, fuel level data, driving pattern data, and driver identity data associated with said vehicle and an environment around said vehicle.

6. The method of claim 4, wherein said plurality of sensors correspond to at least one of a tire pressure sensor, a regenerative braking sensor, a battery capacity sensor, a battery charge sensor, a solar radiation sensor, a humidity sensor, a temperature sensor, a barometric pressure sensor, a motor temperature sensor, a lubrication level sensor, a wind resistance sensor, a proximity sensor, a weight sensor, an identity sensor, and a set of environmental sensors.

7. The method of claim 4, wherein said plurality of database data corresponds to at least one of weather data, route data, traffic data, and driving pattern data of a plurality of drivers.

8. The method of claim 1, wherein said statistical model comprises at least one of a linear function, a quadratic function, a periodic function, and a rule based 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, wherein said database input vector is generated based on at least one of a plurality of environmental data and a road condition information.

9. A system for predicting energy consumption of a vehicle using a statistical model, said system comprising:

an acquisition module that obtains a plurality of input vectors at defined time intervals at a plurality of points in time;

an energy meter that captures an energy level associated with each input vector of said plurality of input vectors at each point in time for said vehicle, wherein said energy meter captures said energy level by capturing at least one of a stored battery power and a stored fuel level of said vehicle;

a processor that predicts a change in energy level using said statistical model, wherein (i) said change in energy comprises a function of corresponding input vectors and an associated weight vector, wherein (ii) said weight vector is derived using said plurality of input vectors and associated energy level at each point in time of said plurality of points in time, and represents an overall effect of each said input vector on energy consumption of the vehicle, and (iii) said change in energy level is predicted through a regression analysis of said energy level associated with said each input vector; and

an output unit that displays results corresponding to the predicted change in said energy level of said vehicle.

10. The system of claim 9, wherein said processor derives said weight vector associated with said input vector using linear regression of said energy level associated with each input vector at each point in time.

11. The system of claim 9, wherein said processor:

predicts a set of input vectors at defined time intervals at a plurality of future points in time based on a subset of said plurality of input vectors generated at said defined time intervals;

captures an actual change in energy level for each point in time of said plurality of future points in time, wherein said actual change in said energy level is based on said energy level of said vehicle associated with each input vector corresponding to each point in time;

computes a difference between a derived change in said energy level and said actual change in said energy level; and

refines said weight vector for minimizing a difference between said derived change in said energy level and said actual change in said energy level, wherein refining said weight vector comprises modifying a value of said weight vector to minimize a difference, wherein said statistical model is refit in response to the refined weight vector.

12. The system of claim 9, wherein said acquisition module acquires a plurality of sensor data for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile at each point in time of said plurality of points in time, wherein said plurality of sensor data is obtained from a plurality of sensors coupled to said vehicle, wherein said plurality of database data is obtained for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile for a plurality of vehicles, and wherein said plurality of database data is obtained from a database storing previously recorded data for at least one of said vehicle location environment, said vehicle equipment profile, and said driver behavior profile corresponding to said plurality of vehicles.

13. The system of claim 12, wherein said acquisition module acquires said plurality of sensor data from a plurality of sensors that correspond to at least one of a tire pressure sensor, a regenerative braking sensor, a battery capacity sensor, a battery charge sensor, a solar radiation sensor, a humidity sensor, a temperature sensor, a barometric pressure sensor, a motor temperature sensor, a lubrication level sensor, a wind resistance sensor, a proximity sensor, a weight sensor, an identity sensor, and a set of environmental sensors.

14. The system of claim 9, wherein said acquisition module acquires said plurality of database data corresponding to at least one of weather data, route data, traffic data, and driving pattern data of a plurality of drivers.

15. The system of claim 9, wherein said processor utilizes said statistical model comprising at least one of a linear function, a quadratic function, a periodic function, and a rule based function of at least one of a stored energy of said vehicle at each point in time, each vehicle input vector, and each database input vector for each defined time interval, and wherein said database input vector is generated based on at least one of a plurality of environmental data and road condition information.

16. A non-transitory program storage device readable by a computer, and comprising a program of instructions executable by said computer to perform a method for predicting energy consumption of a vehicle using a statistical model, said method comprising:

obtaining a plurality of input vectors for said vehicle at defined time intervals at a plurality of points in time, wherein each input vector is associated with each point in time of said plurality of points in time;

capturing an energy level associated with each input vector of said plurality of input vectors at each point in time for said vehicle, wherein said energy level corresponds to at least one of a stored battery power and a stored fuel level of said vehicle;

predicting a change in said energy level using said statistical model, wherein (i) the change in said energy level comprises a function of corresponding input vectors and an associated weight vector, (ii) said weight vector is derived using said plurality of input vectors and associated energy level at each point in time of said plurality of points in time, and represents an overall effect of each said input vector on energy consumption of said vehicle, and (iii) said change in said energy level is predicted through a regression analysis of said energy level associated with each said input vector; and

providing results corresponding to the predicted change in said energy level to an output unit of said vehicle.

17. The program storage device of claim 16, wherein said weight vector associated with said input vector is derived using a linear regression that derives said weight vector based on said plurality of input vectors and respective energy levels at said plurality of points in time.

18. The program storage device of claim 17, wherein said method further comprises:

predicting a set of input vectors at defined time intervals at a plurality of future points in time based on a subset of said plurality of input vectors generated at said defined time intervals, at said plurality of points in time, wherein said subset of said plurality of input vectors represents the most recent input vectors of said vehicle;

deriving a change in said energy level for said plurality of future points in time using said statistical model, wherein said change in said energy level is derived by adding a change in energy level for each defined time interval;

capturing an actual change in energy level for each point in time of said plurality of future points in time, wherein said actual change in said energy level is based on said energy level of said vehicle associated with each input vector corresponding to each point in time;

computing a difference between the derived change in said energy level and said actual change in said energy level; and

refining said weight vector for minimizing the difference between the derived change in said energy level and said actual change in said energy level, wherein refining said weight vector comprises modifying the value of the weight vector to minimize the difference, wherein said statistical model is refit in response to the refined weight vector.

19. The program storage device of claim 16, wherein said each input vector comprises a plurality of sensor data and a plurality of database data, wherein said plurality of sensor data is captured for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile at each point in time of said plurality of points in time, wherein said plurality of sensor data is obtained from a plurality of sensors coupled to said vehicle, wherein said plurality of database data is obtained for at least one of a vehicle location environment, a vehicle equipment profile, and a driver behavior profile for a plurality of vehicles, and wherein said plurality of database data is obtained from a database storing previously recorded data for at least one of said vehicle location environment, said vehicle equipment profile, and said driver behavior profile corresponding to said plurality of vehicles.

20. The program storage device of claim 16, wherein said statistical model comprises at least one of a linear function, a quadratic function, a periodic function, and a rule based 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, wherein said database input vector is generated based on at least one of a plurality of environmental data and a road condition information.