Navigation map learning for intelligent hybrid-electric vehicle planning

Patent metadata

Field	Value
Publication number	US-11946760-B2
Application number	US-202017130490-A
Country	US
Kind code	B2
Filing date	Dec 22, 2020
Priority date	Dec 22, 2020
Publication date	Apr 2, 2024
Grant date	Apr 2, 2024

Abstract

Activating an engine of a HEV to charge a battery includes obtaining a first GPS trace from a first trip of the HEV along a first route, where the first GPS trace includes first trace metadata; obtaining a second GPS trace from a second trip, where the second GPS trace includes second trace metadata; adding, to a navigation map, an aggregation of the first trace metadata and the second trace metadata for edges of the navigation map; using the navigation map to obtain an activation action of the engine, where the activation action is selected from a set that includes a first activation action of turning the engine on and a second activation action of turning the engine off; and activating the engine according to the activation action, where activating the engine using the first activation action causes the engine to turn on to charge the battery of the HEV.

First claim

What is claimed is:

1. A method for activating an engine of a hybrid electric vehicle (HEV) to charge a battery, comprising:
obtaining a first global positioning system (GPS) trace from a first trip of the HEV along a first route, wherein the first route comprises first route segments represented by first respective edges in a navigation map, wherein the first GPS trace comprises first coordinates along the first route and first trace metadata, wherein the first trace metadata comprise first hybrid-related aspects of the HEV recorded at at least some of the first coordinates, and wherein the first hybrid-related aspects comprise, for the first respective edges, first respective average battery consumption/regeneration (e abcr ), first battery regeneration on stop (e abrs ), first number of times stopped (e nts ), and first average stop time (e ast );
obtaining a second GPS trace from a second trip of the HEV along a second route that at least partially overlaps the first route, wherein the second route comprises second route segments represented by second respective edges in the navigation map, wherein the second GPS trace comprises second coordinates along the second route and second trace metadata, wherein the second trace metadata comprise second hybrid-related aspects of the HEV recorded at at least some of the second coordinates, and wherein the second hybrid-related aspects comprise, for the second respective edges, second respective average battery consumption/regeneration (e abcr ), second battery regeneration on stop (e abrs ), second number of times stopped (e nts ), and second average stop time (e ast );
updating the navigation map to incorporate an aggregation of the first trace metadata and the second trace metadata for edges of the navigation map corresponding to route segments traversed by the HEV along the first route or along the second route;
using, after updating the navigation map, the navigation map to obtain an activation action of the engine of the HEV, wherein the activation action is selected from a set comprising a first activation action of turning the engine on and a second activation action of turning the engine off, wherein the activation action is obtained using a policy that is configured to map states of the HEV to activation actions, wherein a state of the HEV is determined based on a set of edges of the navigation map probabilistically determined to have been traversed, a current battery level of the HEV, and a current status of the engine of the HEV, wherein the current status of the engine is one of an ON state or an OFF state; and
activating the engine according to the activation action, wherein activating the engine using the first activation action causes the engine to turn on to charge the battery of the HEV.

2. The method of claim 1 , wherein the first trace metadata and the second trace metadata comprise respective metadata relating to at least one of a battery charge of the HEV, ambient noise, slope, speed of the HEV, acceleration of the HEV, gas pedal status, or brake pedal status.

3. The method of claim 1 , wherein the aggregation of the first trace metadata and the second trace metadata for the edges comprises a probability of traversal of a first route segment of the first route segments subsequent to a traversal of a second route segment of the second route segments.

4. The method of claim 1 , further comprising:
selecting, based on GPS traces comprising coordinates, nodes to create the edges of the navigation map, wherein the nodes are selected at a predefined threshold distance from each other.

5. The method of claim 1 , wherein using the navigation map to obtain the activation action of the engine of the HEV comprises:
identifying, based on a current position of the HEV, a set of most likely roads the HEV to follow, wherein the current position of the HEV is obtained later in time from the first GPS trace and the second GPS trace; and
selecting the activation action based on the set of most likely roads.

6. The method of claim 5 , wherein identifying, based on the current position of the HEV, the set of most likely roads the HEV to follow comprises:
identifying, using edge traversal probabilities of the navigation map and based on a predetermined distance ahead of the HEV, the set of most likely road segments for the HEV to follow.

7. The method of claim 1 , wherein using the navigation map to obtain the activation action of the engine of the HEV comprises:
obtaining the activation action using a Markov decision process (MDP), wherein a state space of the MDP includes the navigation map.

8. The method of claim 1 , further comprising:
determining, using an objective, a route for the HEV using the navigation map, wherein the objective is related to one of an energy expenditure or a noise level.

9. A system for activating an engine of a target hybrid electric vehicle (HEV) to charge a battery, comprising:
a server configured to:
receive from HEVs respective global positioning system (GPS) traces, wherein the GPS traces comprise respective coordinates and trace metadata, wherein each trace metadata relates to at least one of a battery charge of a respective HEV, ambient noise, slope, speed of the respective HEV, acceleration of the respective HEV, gas pedal status of the respective HEV, brake pedal status of the respective HEV, average battery consumption/regeneration (e abcr ) by of the respective HEV, battery regeneration on stop (e abrs ) of the respective HEV, number of times the respective HEV stopped (e nts ), and average stop time (e ast ) by the respective HEV;
obtain a navigation map using the GPS traces, wherein the navigation map comprises edges and edge metadata obtained using the trace metadata;
obtain a policy using the navigation map to obtain an activation action of the engine of the target HEV, wherein the policy is configured to provide the activation action based on historical road segment traversals, a current battery level of the target HEV, and whether the engine of the target HEV is in an ON state or an OFF state, and wherein the activation action is selected from a set comprising a first activation action of turning the engine on and a second activation action of turning the engine off; and
transmit the policy to the target HEV.

10. The system of claim 9 , wherein the policy reduces an energy expenditure of the battery of the target HEV.

11. The system of claim 9 , wherein the policy is configured to reduce a total noise level associated with activating the engine of the target HEV.

12. A non-transitory computer-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
obtaining a first global positioning system (GPS) trace from a first trip along a first route, wherein the first route comprises first route segments represented by first respective edges in a navigation map, wherein the first GPS trace comprises first coordinates along the first route and first trace metadata, wherein the first trace metadata comprises first hybrid-related aspects of a hybrid electric vehicle (HEV) recorded at at least some of the first coordinates, and wherein the first hybrid-related aspects comprise first respective average battery consumption/regeneration (e abcr ) and first battery regeneration on stop (e abrs ) for the first respective edges;
obtaining a second GPS trace from a second trip along a second route that at least partially overlaps the first route, wherein the second route comprises second route segments represented by second respective edges in the navigation map, wherein the second GPS trace comprises second c

Assignees

• Nissan North America Inc

Inventors

• Wray Kyle Hollins
• Ilstrup David
• Pedersen Liam
• Lui Richard
• Ostafew Christopher

Classifications

• G01C21/3469 Primary

  Fuel consumption; Energy use; Emission aspects · CPC title

• B60W10/06 Primary

  including control of combustion engines · CPC title

• B60W20/14

  in conjunction with braking regeneration · CPC title

• B60W40/08

  related to drivers or passengers · CPC title

• G01C21/3807

  characterised by the type of data · CPC title