Navigation map learning for intelligent hybrid-electric vehicle planning
Engine activation planning in a hybrid electric vehicle (HEV) is disclosed. Historical driving data of the HEV are analyzed to identify recurring driving scenarios and patterns specific to a driver of the HEV. Future driving scenarios are predicted based on the historical driving data. An engine activation policy is generated for the HEV. The engine activation policy optimizes battery charging and usage in response to the future driving scenarios. The engine activation policy is used to control activation of a gasoline engine in the HEV where a control decision is dynamically adjusted based on a comparison of real-time driving conditions with the future driving scenarios.
1. A method for engine activation planning in a hybrid electric vehicle (HEV), the method comprising:
analyzing historical driving data of the HEV to identify recurring driving patterns of behavior specific to a driver of the HEV;
recording data related to one or more vehicle parameters as the HEV drives along a road segment of a predetermined length;
predicting future driving patterns of behavior as to where the driver is likely to go based on the historical driving data;
generating an engine activation policy for the HEV to control total energy consumption, wherein the engine activation policy optimizes charging of a battery and usage of the battery in response to the future driving patterns and the data recorded;
measuring wasted energy as the HEV drives along the road segment, wherein the wasted energy is energy that cannot be stored in the battery;
analyzing the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon selected objectives;
weighting the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon the analyzing of the engine activation policy to update the engine activation policy for the HEV based upon the selected objectives to create a weighted engine activation policy; and
using the weighted engine activation policy to control activation of a gasoline engine in the HEV, wherein a control decision is continuously adjusted, in real time, based on the weighted engine activation policy and real time driving conditions.
2. The method of claim 1 , wherein the historical driving data comprise at least one of a specific driving pattern of behavior, routes followed, or environmental conditions encountered during past trips.
3. The method of claim 1 , wherein the future driving patterns are predicted based on at least one of a specific driving pattern of behavior, anticipated route characteristics, or typical environmental conditions.
4. The method of claim 1 , wherein the weighted engine activation policy further incorporates engine noise levels as criteria for the activation of the gasoline engine.
5. The method of claim 1 , further comprising:
evaluating an efficiency of a regenerative braking system of the HEV by measuring an amount of energy recovered during braking events; and
adjusting the weighted engine activation policy based on the efficiency of the regenerative braking system.
6. The method of claim 5 , wherein adjusting the weighted engine activation policy based on the efficiency of the regenerative braking system comprises:
maximizing the battery charging from regenerative energy by modifying engine activation thresholds, by changing the activation of the gasoline engine, based on a predetermined regenerative braking efficiency.
7. The method of claim 6 , further comprising:
calculating the predetermined regenerative braking efficiency using an average energy recovered during braking events, multiplied by a ratio of a number of times braking occurs on a specific route segment to a total number of times the specific route segment is traversed.
8. Non-transitory computer-readable storage media, comprising executable instructions that, when executed by one or more processors, facilitate performance of operations, comprising operations to:
analyze historical driving data of a hybrid electric vehicle (HEV) to identify recurring driving patterns of behavior specific to a driver of the HEV;
predict future driving patterns of behavior as to where the driver is likely to go based on the historical driving data;
collect parameters that relate to segments from the historical driving data;
aggregate the parameters collected;
measure wasted energy as the HEV drives along a segment of a road, wherein wasted energy is energy that cannot be stored in a battery of the HEV;
generate an engine activation policy for the HEV to control total energy consumption, wherein the engine activation policy optimizes battery charging and usage in response to the future driving patterns, the aggregated parameters, and the wasted energy;
analyze the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon selected objectives;
weigh the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon the analyzing of the engine activation policy to update the engine activation policy for the HEV based upon the selected objectives to create a weighted engine activation policy; and
use the weighted engine activation policy to control activation of a gasoline engine in the HEV,
wherein a control decision is continuously adjusted, in real time, based on the weighted engine activation policy and real time driving conditions.
9. The non-transitory computer-readable storage media of claim 8 , wherein the historical driving data comprise at least one of a specific driving pattern of behavior, routes followed, or environmental conditions encountered during past trips; and
wherein a probability of changing from the segment to another road segment is determined when the HEV is at a junction point based on the aggregate of the parameters collected.
10. The non-transitory computer-readable storage media of claim 8 , wherein the future driving patterns are predicted based on at least one of a specific driving pattern of behavior, anticipated route characteristics, or typical environmental conditions.
11. The non-transitory computer-readable storage media of claim 8 , wherein the weighted engine activation policy further incorporates engine noise levels as criteria for the activation of the gasoline engine.
12. The non-transitory computer-readable storage media of claim 8 , wherein the operations further comprise to:
evaluate an efficiency of a regenerative braking system of the HEV by measuring an amount of energy recovered during braking events; and
adjust the weighted engine activation policy based on the efficiency of the regenerative braking system.
13. The non-transitory computer-readable storage media of claim 12 , wherein to adjust the engine activation policy based on the efficiency of the regenerative braking system comprises to:
maximize the battery charging from regenerative energy by modifying engine activation thresholds, by changing an activation of the gasoline engine, based on a predetermined regenerative braking efficiency calculated using an average energy recovered during braking events, multiplied by a ratio of a number of times braking occurs on a specific route segment to a total number of times the specific route segment is traversed, wherein the predetermined regenerative braking efficiency represents a braking and energy recovery on that segment.
14. A system for engine activation planning in a hybrid electric vehicle (HEV) comprising:
one or more processors configured to:
analyze historical driving data of the HEV to identify recurring driving patterns and patterns specific to a driver of the HEV;
collect samples as the HEV travels along a road;
predict future driving patterns of behavior as to where the driver is likely to go based on the historical driving data and the samples collected;
generate an engine activation policy for the HEV to control total energy consumption, wherein the engine activation policy optimizes charging of a battery and usage of the battery in response to the future driving patterns;
measure wasted energy as the HEV drives along a segment of the road, wherein wasted energy is energy that cannot be stored in the battery;
analyze the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon selected objectives;
weigh the total energy consumption, the wasted energy, and the usage of the battery of the engine activation policy based upon the analyzing of the engine activation policy for the HEV based upon the selected objectives to update the engine activation policy for the HEV to create a weighted engine activation policy; and
use the weighted engine activation policy to control activation of a gasoline engine in the HEV, wherein a control decision is continuously adjusted, in real time, based on the weighted engine activation policy and real time driving conditions.
15. The system of claim 14 , wherein the historical driving data comprise at least one of a specific driving pattern of behavior, routes followed, or environmental conditions encountered during past trips.
16. The system of claim 14 , wherein the future driving patterns are predicted based on at least one of a specific driving pattern of behavior, anticipated route characteristics, or typical environmental conditions.
17. The system of claim 14 , wherein the weighted engine activation policy further incorporates engine noise levels as criteria for the activation of the gasoline engine.
18. The system of claim 14 , wherein the one or more processors are further configured to:
evaluate an efficiency of a regenerative braking system of the HEV by measuring an amount of energy recovered during braking events; and
adjust the weighted engine activation policy based on the efficiency of the regenerative braking system.
19. The system of claim 18 , wherein to adjust the weighted engine activation policy based on the efficiency of the regenerative braking system comprises to:
maximize the battery charging from regenerative energy by modifying engine activation thresholds, by changing an activation of the gasoline engine, based on a predetermined regenerative braking efficiency.
20. The system of claim 19 , wherein the one or more processors are further configured to:
calculate the predetermined regenerative braking efficiency using an average energy recovered during braking events, multiplied by a ratio of a number of times braking occurs on a specific route segment to a total number of times the specific route segment is traversed, wherein the predetermined regenerative braking efficiency represents braking and energy recovery on that segment.