METHODS AND APPARATUS TO AUTONOMOUSLY NAVIGATE A VEHICLE BY SELECTING SENSORS FROM WHICH TO OBTAIN MEASUREMENTS FOR NAVIGATION

• US 20170131716A1
• Filed: 11/06/2015
• Published: 05/11/2017
• Est. Priority Date: 11/06/2015
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Brekke, Darin William, McBride, Robert Stewart
• Granted Patent: US 10,048,686 B2
• Time in Patent Office: Days
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G01C 21/3415 Dynamic re-routing, e.g. recalculating the route when the user deviates from calculated route or after detecting real-time traffic data or accidents
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00

Abstract

Methods and apparatus to autonomously navigate a vehicle by selecting sensors from which to obtain measurements for navigation are disclosed. An example method to navigate a vehicle includes determining environmental data associated with an area in which the vehicle is to navigate; based on the environmental data, automatically ranking a plurality of sensors that are available to the vehicle by respective costs of using the sensors to generate a navigation solution; automatically determining a subset of the sensors from which to obtain measurements based on the ranking and based on comparisons of a) first importance values of the sensors for navigating the vehicle with b) second importance values of the sensors for performing a non-navigation task; obtaining measurements from the subset of the sensors; calculating a navigation command to navigate the vehicle; and executing a first navigation action to implement the navigation command.

Claims

1. A method to navigate a vehicle, comprising:

determining environmental data associated with an area in which the vehicle is to navigate;

based on the environmental data, automatically ranking a plurality of sensors that are available to the vehicle by respective costs of using the sensors to generate a navigation solution;

automatically determining a subset of the sensors from which to obtain measurements based on the ranking and based on comparisons of a) first importance values of the sensors for navigating the vehicle with b) second importance values of the sensors for performing a non-navigation task;

obtaining measurements from the subset of the sensors;

calculating a navigation command to navigate the vehicle; and

executing a first navigation action to implement the navigation command.

2. The method defined in claim 1, further including:

calculating observation noise metrics corresponding to the sensors in the subset based on the environmental data;

predicting a covariance of a navigation state of the vehicle based on the observation noise metrics; and

in response to determining that the predicted covariance exceeds a threshold variance, requesting a second navigation action.

3. The method defined in claim 2, wherein the second navigation action is at least one of:

executing a maneuver using the vehicle;

following a modified route via the vehicle;

or

forcing a modification of the ranking of the sensors.

4. The method defined in claim 1, wherein the determining of the subset of the sensors includes:

selecting a first one of the sensors based on a first rank of the first one of the sensors;

comparing a first one of the first importance values for the first one of the sensors to a first one of the second importance values for the first one of the sensors; and

when the first one of the first importance values exceeds the first one of the second importance values, including the first one of the sensors in the subset.

5. The method defined in claim 4, wherein the including of the first one of the sensors in the subset is based further on determining that, prior to adding the first one of the sensors to the subset, the subset of the sensors would not generate a navigation state to have at least a threshold accuracy.

6. The method defined in claim 4, wherein the determining of the subset of the sensors further includes:

selecting a second one of the sensors based on a second rank of the second one of the sensors;

determining whether the second one of the sensors is available for use in generating the navigation solution; and

when the second one of the sensors is not available, omitting the second ones of the sensors from the subset of the sensors.

7. The method defined in claim 6, wherein the determining of whether the second one of the sensors is available includes at least one of:

a) identifying a restriction on using the second one of the sensors, the restriction being defined by the non-navigation task or a second task;

b) identifying a vehicle maneuver that is required to use the second one of the sensors for navigation as being outside of bounds placed by the second task or the third task;

or c) identifying interference with the non-navigation task or the second task as occurring based on using the second one of the sensors for navigation.

8. The method defined in claim 1, further including determining the costs based on at least one of:

a) predicted respective uncertainties in the navigation solution that would result from using the sensors;

b) relative preferences for the sensors based on the navigating of the vehicle and the non-navigation task;

c) power requirements of the sensors;

or d) bandwidth usage for communicating measurements from the sensors.

9. The method defined in claim 1, further including:

detecting a change in at least one of the environmental data or a navigation state of the vehicle;

automatically updating the ranking of the plurality of sensors based on updated environmental data and an updated navigation state of the vehicle;

automatically updating the first importance values of the sensors and the second importance values of the sensors;

automatically updating the subset of the sensors from which to obtain measurements based on the updated ranking, the updated first importance values, and the updated second importance values;

obtaining second measurements from the updated subset of the sensors;

calculating a second navigation command to navigate the vehicle; and

executing a second navigation action to implement the second navigation command.

23. The method of claim 1, further including predicting uncertainties introduced by the respective sensors into the navigation solution based on a covariance when the respective sensors are used as a source of data to update a navigation state, and wherein the costs of using the respective sensors are based on the uncertainties introduced by the respective sensors.

10. A vehicle, comprising:

a propulsion system to provide thrust to the vehicle;

a control system to provide directional control to the vehicle;

a plurality of sensors to collect data;

a sensor ranker to;

receive environmental data associated with an area in which the vehicle is to navigate;

based on the environmental data, automatically rank the sensors by respective costs of using the sensors to generate a navigation solution;

a sensor negotiator to automatically determine a subset of the sensors from which to obtain measurements based on the ranking and based on comparisons of a) first importance values of the sensors for navigating the vehicle with b) second importance values of the sensors for performing a non-navigation task; and

a navigation filter to;

obtain measurements from the subset of the sensors;

calculate a navigation command to navigate the vehicle; and

execute a first navigation action to implement the navigation command.

11. The vehicle as defined in claim 10, further including:

a variance matrix calculator to calculate observation noise metrics corresponding to the sensors in the subset based on the environmental data, the navigation filter to predict a covariance of a navigation state of the vehicle based on the observation noise metrics; and

a state error monitor to, in response to determining that the predicted covariance exceeds a threshold variance, request a second navigation action.

12. The vehicle as defined in claim 11, wherein the second navigation action is at least one of:

executing a maneuver via at least one of the control system or the propulsion system;

following a modified route via at least one of the control system or the propulsion;

or

forcing the sensor ranker to modify the ranking of the sensors.

13. The vehicle as defined in claim 10, wherein the sensor negotiator is to determine of the subset of the sensors by:

selecting a first one of the sensors based on a first rank of the first one of the sensors;

comparing a first one of the first importance values for the first one of the sensors to a first one of the second importance values for the first one of the sensors; and

when the first one of the first importance values exceeds the first one of the second importance values, including the first one of the sensors in the subset.

14. The vehicle as defined in claim 13, wherein the sensor negotiator is to include the first one of the sensors in the subset based on determining that, prior to adding the first one of the sensors to the subset, the subset of the sensors would not generate a navigation state to have at least a threshold accuracy.

15. The vehicle as defined in claim 13, wherein the sensor negotiator is to determine of the subset of the sensors by:

selecting a second one of the sensors based on a second rank of the second one of the sensors;

determining whether the second one of the sensors is available for use in generating the navigation solution; and

when the second one of the sensors is not available, omitting the second one of the sensors from the subset of the sensors.

16. The vehicle as defined in claim 10, wherein the sensors include a navigation sensor having a primary task of providing navigation information, the sensor ranker to include the navigation sensor in the ranking of the sensors.

17. A tangible machine readable storage medium comprising machine executable instructions which, when executed, cause a processing circuit of a vehicle to at least:

determine environmental data associated with an area in which the vehicle is to navigate;

based on the environmental data, automatically rank a plurality of sensors that are available to the vehicle by respective costs of using the sensors to generate a navigation solution;

automatically determine a subset of the sensors from which to obtain measurements based on the ranking and based on comparisons of a) first importance values of the sensors for navigating the vehicle with b) second importance values of the sensors for performing a non-navigation task;

access measurements from the subset of the sensors;

calculate a navigation command to navigate the vehicle; and

execute a first navigation action to implement the navigation command.

18. The tangible machine readable storage medium of claim 17, wherein the instructions are further to cause the processing circuit to:

calculate observation noise metrics corresponding to the sensors in the subset based on the environmental data;

predict a covariance of a navigation state of the vehicle based on the observation noise metrics; and

in response to determining that the predicted covariance exceeds a threshold variance, request a second navigation action.

19. The tangible machine readable storage medium of claim 17, wherein the instructions are to cause the processing circuit to determine of the subset of the sensors by:

selecting a first one of the sensors based on a first rank of the first one of the sensors;

comparing a first one of the first importance values for the first one of the sensors to a first one of the second importance values for the first one of the sensors; and

when the first one of the first importance values exceeds the first one of the second importance values, including the first one of the sensors in the subset.

20. The tangible machine readable storage medium of claim 19, wherein the instructions are to cause the processing circuit to include the first one of the sensors in the subset based further on determining that, prior to adding the first one of the sensors to the subset, the subset of the sensors would not generate a navigation state to have at least a threshold accuracy.

21. The tangible machine readable storage medium of claim 19, wherein the instructions are to cause the processing circuit to determine the subset of the sensors by:

selecting a second one of the sensors based on a second rank of the second one of the sensors;

determining whether the second one of the sensors is available for use in generating the navigation solution; and

when the second one of the sensors is not available, omitting the second sensor from the subset of the sensors.

22. The tangible machine readable storage medium of claim 17, wherein the instructions are further to cause the processing circuit to:

detect a change in at least one of the environmental data or a navigation state of the vehicle;

automatically update the ranking of the plurality of sensors based on updated environmental data and an updated navigation state of the vehicle;

automatically update the first importance values of the sensors and the second importance values of the sensors;

automatically update the subset of the sensors from which to obtain measurements based on the updated ranking, the updated first importance values, and the updated second importance values;

access second measurements from the updated subset of the sensors;

calculate a second navigation command to navigate the vehicle; and

execute a second navigation action to implement the second navigation command.