System and method for detecting active road work zones
The disclosure provides a system, a method, and a computer program product for detecting active road work zones. The system obtains first sensor data associated with a detection of a road work zone, using a first sensor type and obtains second sensor data associated with presence of at least one individual in the road work zone, using a second sensor type. Further, the system determines a confidence level associated with the detection of the road work zone based on fusing of the first sensor data and the second sensor data. The system classifies the detected road work zone as at least one of the active road work zone or a non-active road work zone, based on the confidence level. The system updates map data associated with a map of a region based on the classification.
1 . A system for detecting active road work zone, the system comprising:
a memory configured to store computer-executable instructions; and
at least one processor configured to execute the computer-executable instructions to:
obtain first sensor data associated with a detection of a road work zone, using a first sensor type, the first sensor data comprising first real-time sensor data;
obtain second sensor data associated with a presence of at least one individual in the road work zone, using a second sensor type, the first sensor data comprising second real-time sensor data;
determine a confidence level associated with the detection of the road work zone based on fusing of the first sensor data obtained from the first sensor type and the second sensor data obtained from the second sensor type;
classify the detected road work zone as at least one of the active road work zone or a non-active road work zone, based on the confidence level, wherein the detected road work zone is classified as the active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data indicating the presence of the at least one individual in the road work zone, and wherein the detected road work zone is classified as a non-active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data not indicating the presence of the at least one individual; and
update map data associated with a map of a region in real-time based on the classification.
2 . The system of claim 1 , wherein the detected road work zone is classified as at least one of:
the active road work zone when the confidence level is above a predefined threshold; or
the non-active road work zone when the confidence level is below the predefined threshold.
3 . The system of claim 1 , wherein the first sensor data and the second sensor data are obtained from a plurality of sensors comprised of the first sensor type and the second sensor type associated with one or more vehicles, wherein the first sensor type comprises a first camera and the second sensor type comprises a second camera.
4 . The system of claim 3 , wherein the first sensor data comprises data associated with at least one of distance between the one or more vehicles and objects on one or more roads, geographical location of the one or more roads, images of road objects, 3-dimensional surrounding images of the one or more vehicles, and construction related signs on the one or more roads.
5 . The system of claim 1 , wherein fusion of the first sensor data obtained from the first sensor type and the second sensor data obtained from the second sensor type comprises causing the at least one processor to execute computer-executable instructions to:
determine a first region associated with the detection of the road work zone using the first sensor data of the first sensor type;
extract the second sensor data of the second sensor type for the determined first region; and
fuse the first sensor data of the first sensor type and the extracted second sensor data of the second sensor type based on a spatio-temporal constraint criterion.
6 . The system of claim 5 , wherein the spatio-temporal constraint criterion comprises causing the at least one processor to execute computer-executable instructions to:
determine a first road link, a first travel direction, and a first timestamp associated with the first sensor data of the first sensor type;
determine a second road link, a second travel direction, and a second timestamp associated with the second sensor data of the second sensor type; and
fuse the first sensor data of the first sensor type and the extracted second sensor data of the second sensor type when the first road link, the first travel direction, and the first timestamp associated with the first sensor data of the first sensor type overlap respectively with the second road link, the second travel direction, and the second timestamp associated with the second sensor data of the second sensor type.
7 . The system of claim 1 , wherein the second sensor data associated with at least the presence of at least one individual in the road work zone using the second sensor type comprises at least:
vulnerable road user (VRU) data associated with the second sensor type, wherein the VRU data further comprises at least: a number value, a timestamp value, and position of the at least one individual.
8 . The system of claim 7 , wherein the confidence level is incremented by a predetermined value when presence of the VRU is determined in the detected road work zone.
9 . The system of claim 1 , wherein the at least one processor is configured to calculate a longitudinal distance and a lateral distance between the at least one individual and each of the one or more vehicles.
10 . The system of claim 1 , wherein the at least one processor is further configured to:
apply a computer vision algorithm on the obtained first sensor data and the second sensor data associated with the detected road work zone.
11 . The system of claim 10 , wherein the at least one processor is configured to send alerts associated with the active road work zone, based on the determined confidence level.
12 . The system of claim 1 , wherein the at least one processor is configured to aggregate a number of observations of the active road work zone to determine the confidence level.
13 . The system of claim 12 , wherein the confidence level associated with the active road work zone increases as the number of observations of the active road work zone increases.
14 . The system of claim 12 , wherein the at least one processor is configured to create pattern data for active road work zones based on the number of observations of the active road work zone.
15 . A method for detecting active road work zones, the method comprising:
obtaining first sensor data associated with a detection of a road work zone, using a first sensor type, the first sensor data comprising first real-time sensor data;
obtaining second sensor data associated with presence of at least one individual in the road work zone, using a second sensor type, the first sensor data comprising second real-time sensor data;
determining a confidence level associated with the detection of the road work zone based on fusing of the first sensor data obtained from the first sensor type and the second sensor data obtained from the second sensor type;
classifying the detected road work zone as at least one of the active road work zone or a non-active road work zone, based on the determined confidence level, wherein the detected road work zone is classified as the active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data indicating the presence of the at least one individual in the road work zone, and wherein the detected road work zone is classified as a non-active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data not indicating the presence of the at least one individual; and
updating map data associated with a map of a region in real-time based on the classification.
16 . The method of claim 15 , wherein classifying the detected road work zone based on the determined confidence level comprises at least one of:
the active road work zone when the confidence level is above a predefined threshold for a number of observations of the active road work zone; or
the non-active road work zone when the confidence level is below the threshold for the number of observations of the active road work zone.
17 . The method of claim 15 , further comprising:
creating pattern data for active road work zones based on the number of observations of the active road work zone.
18 . The method of claim 15 , wherein fusing the first sensor data obtained from the first sensor type and the second sensor data obtained from the second sensor type comprises:
determining a first region associated with the detection of the road work zone using the first sensor data of the first sensor type;
extracting the second sensor data of the second sensor type for the determined first region; and
fusing the first sensor data of the first sensor type and the extracted second sensor data of the second sensor type based on a spatio-temporal constraint criterion.
19 . The method of claim 18 , wherein the spatio-temporal constraint criterion comprises:
determining a first road link, a first travel direction, and a first timestamp associated with the first sensor data of the first sensor type;
determining a second road link, a second travel direction, and a second timestamp associated with the second sensor data of the second sensor type; and
fusing the first sensor data of the first sensor type and the extracted second sensor data of the second sensor when the first road link, the first travel direction, and the first timestamp associated with the first sensor data of the first sensor type overlap respectively with the second road link, the second travel direction, and the second timestamp associated with the second sensor data of the second sensor type.
20 . A computer program product comprising a non-transitory computer readable medium having stored thereon computer executable instructions which when executed by at least one processor, cause the processor to carry out operations for detecting active road work zones, the operations comprising:
obtaining first sensor data associated with a detection of a road work zone, using a first sensor type, the first sensor data comprising first real-time sensor data;
obtaining second sensor data associated with presence of at least one individual in the road work zone, using a second sensor type, the first sensor data comprising second real-time sensor data;
determining a confidence level associated with the detection of the road work zone based on fusing of the first sensor data and the second sensor data;
classifying the detected road work zone as at least one of the active road work zone or a non-active road work zone, based on the confidence level, wherein the detected road work zone is classified as the active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data indicating the presence of the at least one individual in the road work zone, and wherein the detected road work zone is classified as a non-active road work zone based on the first sensor data indicating the detection of the road work zone and the second sensor data not indicating the presence of the at least one individual; and
updating map data associated with a map of a region in real-time based on the classification.