Energy efficient routing using an impedance factor
View Patent ↗A method and system for calculating an energy efficient route is disclosed. A route calculation application calculates one or more routes from an origin to a destination. For each of the routes, the route calculation application uses impedance factor data associated with each segment in the route. The impedance factor is calculated using probe data when the probe data is available for a road segment. When probe data is unavailable, the impedance factor is calculated using machine learning techniques that analyze the results of the impedance factor classifications for road segments having probe data.
1. A computer-implemented method for calculating road segment cost for energy efficient routing, comprising:
selecting a sample of road segments, wherein the road segments in the sample have been classified with one of a plurality of impedance factors, wherein an impedance factor represents a probability that a vehicle needs to decelerate while traveling on the road segment;
calculating by a processor a probability that a road segment in the sample has been classified with a particular impedance factor;
calculating a conditional probability of each impedance factor for a plurality of map database attributes;
calculating entropy of the plurality of impedance factors;
calculating conditional entropy for each of the plurality of map database attributes;
calculating information gain using the entropy and the conditional entropy calculations;
creating a decision tree based on the information gain calculation; and
classifying the road segment based on the decision tree.
2. The method of claim 1 , further comprising dividing the sample of road segments by functional class of the road segments.
3. The method of claim 1 , wherein the map data attributes include speed category, lane category, controlled access, ramp, and cross streets.
4. The method of claim 1 , further comprising storing data representing the classification of the road segment in a map database.
5. The method of claim 4 , wherein the classification of the road segment stored in the map database is used as a segment cost during route calculation.
6. The method of claim 4 , wherein the classification of the road segment stored in the map database is used to identify errors in the map database.
7. The method of claim 1 , wherein the road segments are classified with the impedance factor of VERY HIGH, HIGH, MEDIUM, LOW, or VERY LOW.
8. An apparatus comprising:
at least one processor; and
at least one memory including computer program code for one or more programs, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,
select a sample of road segments, wherein the road segments in the sample have been classified with one of a plurality of impedance factors, wherein an impedance factor represents a probability that a vehicle decelerates while traveling on the road segment;
calculate a probability that a road segment in the sample has been classified with a particular impedance factor;
calculate a conditional probability of each impedance factor for a plurality of map database attributes;
calculate entropy of the plurality of impedance factors;
calculate conditional entropy for each of the plurality of map database attributes;
calculate information gain using the entropy and the conditional entropy calculations;
create a decision tree based on the information gain calculation; and
classifying the road segment based on the decision tree.
9. The apparatus of claim 8 , wherein the sample of road segments are divided by functional class of the road segments.
10. The apparatus of claim 8 , wherein the map data attributes include speed category, lane category, controlled access, ramp, and cross streets.
11. The apparatus of claim 8 , wherein the classification of the road segment are stored in a map database.
12. The apparatus of claim 11 , wherein the classification of the road segment stored in the map database is used as a segment cost during route calculation.
13. The apparatus of claim 11 , wherein the classification of the road segment stored in the map database is used to identify errors in the map database.
14. The apparatus of claim 8 , wherein the road segments are classified with an impedance factor of VERY HIGH, HIGH, MEDIUM, LOW, or VERY LOW.
15. A method comprising:
selecting a sample of road segments, wherein the road segments in the sample have been classified with one of a plurality of impedance factors, wherein an impedance factor represents a probability that a vehicle decelerates while traveling on the road segment;
calculating by a processor entropy of the plurality of impedance factors;
calculating information gain using the entropy calculation;
creating a decision tree based on the information gain calculation; and
classifying the road segment based on the decision tree.
16. The method of claim 15 , further comprising:
calculating a probability that a road segment in the sample has been classified with a particular impedance factor.
17. The method of claim 16 , further comprising:
calculating a conditional probability of each impedance factor for a plurality of map database attributes; and
calculating conditional entropy for each of the plurality of map database attributes, wherein calculating the information gain also uses the conditional entropy calculation.