Method, apparatus, and system for detecting and characterizing parking events based on sensor data
An approach is provided for determining a vehicle parking event and respective characteristics using sensor data. The approach, for example, involves receiving sensor data from at least one sensor associated with a mobile device in a vehicle. The approach also involves processing the sensor data to determine a sequence of semantic events. The semantic events respectively indicate a maneuver performed by the vehicle. The approach further involves processing the sensor data to determine a distance estimation over which at least one of the semantic events is performed. The approach further involves detecting a parking event of the vehicle, a characterization of the parking event, or a combination thereof based on the sequence of semantic events and the distance estimation. The approach further involves providing the parking event, the characterization of the parking event, or a combination thereof as an output.
1. A method comprising:
receiving sensor data from at least one sensor associated with a mobile device in a vehicle, wherein the at least one sensor comprises an accelerometer, a gyroscope, a magnetometer, or a combination thereof that is configured to collect the sensor data for a designated distance or time associated with a parking maneuver performed by the vehicle to minimize an error accumulation in the sensor data over the designated distance or time;
processing the sensor data to determine a sequence of semantic events, wherein the semantic events respectively indicate the parking maneuver performed by the vehicle;
processing the sensor data to determine a distance estimation over which at least one of the semantic events is performed;
detecting a parking event of the vehicle, a characterization of the parking event, or a combination thereof based on the sequence of semantic events and the distance estimation; and
providing the parking event, the characterization of the parking event, or a combination thereof as an output.
2. The method of claim 1 , further comprising:
extracting accelerometer data from the sensor data; and
integrating over the accelerometer data to determine the distance estimation.
3. The method of claim 1 , wherein the detecting of the parking event, the characterization of the parking event, or a combination thereof is based on determining that the distance estimation is below a distance threshold associated with a parking maneuver.
4. The method of claim 3 , wherein the distance threshold is based on a length of the vehicle.
5. The method of claim 3 , wherein the distance threshold is based on a maneuver distance associated with the parking maneuver corresponding to at least one of the semantic events.
6. The method of claim 5 , wherein the parking maneuver corresponding to the at least one of the semantic events includes a forward or reverse maneuver, a left or right turn maneuver, a stopping maneuver, or a combination thereof.
7. The method of claim 1 , wherein the characterization indicates a parking type of the parking event, and wherein the parking type includes a perpendicular parking type, a parallel parking type, an angular parking type, or a combination thereof.
8. The method of claim 1 , further comprising:
extracting gear selection data from the sensor data,
wherein the gear selection data indicates whether the vehicle is in a forward drive gear or a reverse drive gear; and
wherein the semantic events, the sequence of the semantic events, or a combination thereof are further based on the gear selection data.
9. The method of claim 1 , further comprising:
extracting engine activation data from the sensor data,
wherein the engine activation data indicates whether an engine status of the vehicle is on or off; and
wherein the semantic events, the sequence of the semantic events, or a combination thereof are further based on the engine status.
10. The method of claim 1 , further comprising:
extracting air pressure data from the sensor data,
wherein the air pressure data indicates whether a door status of the vehicle is open or close, and
wherein the semantic events, the sequence of the semantic events, or a combination thereof are further based on the door status.
11. The method of claim 1 , wherein the at least one sensor is an inertial measurement unit associated with a mobile device in which a satellite-based location sensor is unavailable or unused.
12. 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,
receive sensor data from a plurality of sensors associated with at least one mobile device in at least one vehicle, wherein the at least one sensor comprises an accelerometer, a gyroscope, a magnetometer, or a combination thereof that is configured to collect the sensor data for a designated distance or time associated with a parking maneuver performed by the vehicle to minimize an error accumulation in the sensor data over the designated distance or time;
process the sensor data to determine a plurality of parking modes, wherein each of the parking mode includes a time-sequenced semantic events, wherein each of the time-sequenced semantic events indicates the parking maneuver performed by the at least one vehicle;
determine a parking mode of a vehicle based on comparing a time-sequenced semantic events of the vehicle with time-sequenced semantic events of the plurality of parking modes; and
provide the parking mode as an output.
13. The apparatus of claim 12 , wherein the apparatus is further caused to:
extract accelerometer data from the sensor data; and
integrate over the accelerometer data to determine the distance estimation,
wherein the parking mode is further determined based on the distance estimation.
14. The apparatus of claim 13 , wherein the parking mode is based on determining that the distance estimation is below a distance threshold associated with a parking maneuver.
15. The apparatus of claim 14 , wherein the distance threshold is based on a length of the vehicle.
16. The apparatus of claim 12 , wherein the apparatus is further caused to:
extract data of one or more of an accelerometer, a gyroscope, a magnetometer, or a barometer from the sensor data; and
integrate over the data to determine a vehicle idle state, an estimated turning angle, a lane change, a direction of motion, a door state, or a combination thereof of at least one of the time-sequenced semantic events,
wherein the parking mode is further determined based on the vehicle idle state, the estimated turning angle, the lane change, the direction of motion, the door state, or a combination thereof.
17. The apparatus of claim 12 , wherein the plurality of parking modes include one or more of left forward bay parking, right forward bay parking, left reverse bay parking, right reverse bay parking, left forward parallel parking, right forward parallel parking, left reverse parallel parking, right reverse parallel parking, left forward angular parking, right forward angular parking, left reverse angular parking, or right reverse angular parking.
18. A non-transitory computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to perform:
receiving sensor data from at least one sensor associated with a mobile device in a vehicle, wherein the at least one sensor comprises an accelerometer, a gyroscope, a magnetometer, or a combination thereof that is configured to collect the sensor data for a designated distance or time associated with a parking maneuver performed by the vehicle to minimize an error accumulation in the sensor data over the designated distance or time;
processing the sensor data to determine a sequence of semantic events, wherein the semantic events respectively indicate the parking maneuver performed by the vehicle;
processing the sensor data to determine a distance estimation over which at least one of the semantic events is performed;
detecting a parking event of the vehicle, a characterization of the parking event, or a combination thereof based on the sequence of semantic events and the distance estimation; and
providing the parking event, the characterization of the parking event, or a combination thereof as an output.
19. The non-transitory computer-readable storage medium of claim 18 , wherein the apparatus is caused to further perform:
extracting accelerometer data from the sensor data; and
integrating over the accelerometer data to determine the distance estimation.
20. The non-transitory computer-readable storage medium of claim 18 , wherein the detecting of the parking event, the characterization of the parking event, or a combination thereof is based on determining that the distance estimation is below a distance threshold associated with a parking maneuver.