Actions prediction for hypothetical driving conditions

• US 9,434,389 B2
• Filed: 03/06/2014
• Issued: 09/06/2016
• Est. Priority Date: 11/18/2013
• Status: Active Grant

• Current Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Harsham, Bret, Hershey, John R., Le Roux, Jonathan, Nikovski, Daniel Nikolaev, Esenther, Alan W.
• Primary Examiner(s): Cheung, Calvin
• Assistant Examiner(s): Martinez Borrero, Luis A
• Application Number:
  US14/198,742
• Publication Number:
  US 20150142205A1
• Time in Patent Office: 915 Days
• Field of Search: 701/1, 701/22, 701/117, 701/2, 701/3, 701/36, 701/102, 701/101, 701/104, 701/108, 701/112, 701/400, 701/103, 701/100, 701/54, 340/441, 340/901, 340/436, 340/961
• US Class Current: 1/1
• CPC Class Codes: B60W 2050/0075 Automatic parameter input, automatic initialising or calibrating means
• CPC Class Codes: B60W 2556/10 Historical data
• CPC Class Codes: B60W 50/0097 Predicting future conditions
• CPC Class Codes: B60W 50/085 Changing the parameters of the control units, e.g. changing limit values, working points by control input
• CPC Class Codes: B60W 50/10 Interpretation of driver requests or demands
• CPC Class Codes: G01C 21/3484 Personalized, e.g. from learned user behaviour or user-defined profiles

Abstract

An information system includes a prediction engine for predicting an action based on a set of driving state parameters, and a driving history, and a simulation engine for generating a hypothetical scenario by simulating one or a combination of at least one driving state parameter and at least part of the driving history, such that the prediction engine predicts the action for the hypothetical scenario.

Claims

1. An information system installed in a vehicle, comprising:

a persistent storage device to store a driving history of a driver of the vehicle;

a set of internal sensors to determine internal measurements;

a receiver to receive, over a network, external measurements of a set of external sensors;

a processor to determine, using one or combination of the internal and the external measurements, a set of driving state parameters and to update the driving history according to the set of driving state parameters; and

a human-machine-interface (HMI) including input and output interfaces to modify a current or a historical value of the driving state parameter with a simulated value;

wherein the processor further executes modules of the information system including a prediction engine and a simulation engine,

wherein the prediction engine executed by the processor predicts an action based on the set of driving state parameters and the driving history,

wherein the simulation engine executed by the processor generates a hypothetical scenario, the prediction engine predicts the action for the hypothetical scenario, and the processor renders the predicted action on an output device, and

wherein the simulation engine executed by the processor generates the hypothetical scenario by initializing a set of driving state parameters based on the driving history, such that each driving state parameter is initialized with a corresponding current or historical value, rendering a subset of the set of driving state parameters to the interface, receiving, in response to the rendering, the simulated value forming a modified subset of driving state parameters, and merging the modified subset of driving state parameters with the set of driving state parameters for predicting the action.

2. The system of claim 1, further comprising:

a driving history module for collecting the driving history of a vehicle or actions of the driver of the vehicle and for creating a prediction model for the prediction engine; and

at least one sensor for determining the driving state parameter having a value independent from the driving history, wherein the simulation engine modifies the value of the driving state parameter.

3. The system of claim 2, further comprising:

a system module using the driving state parameter, wherein the simulation engine modifies only a copy of the driving state parameter used by the prediction engine without modifying the driving state parameter used by the system module.

4. The system of claim 3, wherein the driving state parameter is a current location of a vehicle, wherein the system module is a navigation system of the vehicle, and the simulation engine changes the current location of the vehicle as used by the prediction engine without changing the current location of the vehicle used by the navigation system.

5. The system of claim 3, wherein the driving state parameter is a current time or a calendar date, wherein the system module is a clock or a calendar system of the vehicle, and the simulation engine changes the current time or the calendar date for the prediction engine without changing the current time or the calendar date used by the clock or the calendar system.

6. The system of claim 2, wherein the simulation engine modifies the prediction model without changing the driving history.

7. The system of claim 1, wherein the set of driving state parameters includes an external parameter received from an external sensor, and wherein the simulation engine modifies a received value of the external parameter.

8. The system of claim 1, further comprising:

an output interface for rendering the set of driving state parameters used in a current prediction; and

an input interface for modifying values of at least some driving state parameters in the set.

9. The system of claim 1, wherein the processor further executes

an analytical module for performing an analysis of a set of actions predicted for a set of hypothetical scenarios and for determining a subset of driving state parameters based on the analysis.

10. The system of claim 9, wherein the output interface displays concurrently multiple actions and multiple corresponding hypothetical scenarios.

11. A method for using an information system installed in a vehicle, comprising:

storing a driving history of a driver of the vehicle a persistent storage device of the vehicle;

determining internal measurements using a set of internal sensors of the vehicle;

receiving, over a network, external measurements of a set of external sensors;

determining, using one or combination of the internal and the external measurements, a set of driving state parameters;

updating the driving history according to the set of driving state parameters;

modifying a current or a historical value of the driving state parameter with a simulated value received from a human-machine-interface (HMI) of the vehicle;

generating a hypothetical scenario by initializing a set of driving state parameters based on the driving history, such that each driving state parameter is initialized with a corresponding current or historical value, rendering a subset of the set of driving state parameters to the interface, receiving, in response to the rendering, the simulated value forming a modified subset of driving state parameters, and merging the modified subset of driving state parameters with the set of driving state parameters;

predicting an action for the hypothetical scenario; and

rendering the predicted action on an output device, wherein at least some steps of the method are performed by a processor.