Driver assistance system and control method for the same
It is an aspect of the present disclosure to provide a driver assistance system and control method thereof in which perform a data association of various data to be collected, thereby matching a plurality of features and performing precise position measurement.
1. A driver assistance system (DAS) comprising:
a camera provided in a vehicle to have a field of view (FOV), and configured to acquire image data;
a radar sensor provided in the vehicle to have a field of sensing (FOS) including the region forward or a side of the vehicle, and configured to acquire radar data; a storage provided in the vehicle and storing an High Definition (HD) Map;
a controller including at least of one processor configured to process the image data acquired by the camera and the radar data acquired by the radar sensor,
wherein the controller extracts a feature included in the HD map based on the image data and the radar data, and modifies a Global Positioning System (GPS) data including the HD map according to a feature determined based on a confidence score of extracted feature and a probability distribution, and
wherein the controller extracts a plurality of features based on the image data and the HD map and measure a plurality of offsets by matching the plurality of features with the radar data.
2. The driver assistance system (DAS) according to claim 1 , wherein the controller obtains the confidence score and the probability distribution based on Integrated Probabilistic Data Association (IPDA), and calculates a weight for the plurality of feature information determined based on the confidence score and the probability distribution.
3. The driver assistance system (DAS) according to claim 1 , wherein the controller updates an offset estimation value included in the weight based on a Kalman filter.
4. The driver assistance system (DAS) according to claim 3 , wherein the controller matches the updated offset estimation value with the HD map, and stops the modification of the GPS data based on the matching result.
5. The driver assistance system (DAS) according to claim 3 , wherein the controller matches the updated offset estimation value with the HD map, and predicts the offset estimation value with an offset of the current time point at an offset of a current point based on the matching result and a dynamics model.
6. The driver assistance system (DAS) according to claim 5 , wherein the controller matches the offset of the current point with the HD map, and stops the modification of the GPS data based on the matching result.
7. The driver assistance system (DAS) according to claim 5 , wherein the controller matches the offset of the current point with the HD map, and modifies the GPS data based on the matching result and the offset of the current point.
8. The driver assistance system (DAS) according to claim 7 , wherein the controller determines a location of the vehicle in the HD map based on the modified GPS data.
9. A controlling method of a driver assistance system comprising,
acquiring image data by a camera provided in a vehicle to have a field of view (FOV);
acquiring radar data by a radar sensor provided in the vehicle to have a field of sensing (FOS) including the region forward or a side of the vehicle;
storing a High Definition (HD) map by a storage after receiving the HD map from outside;
extracting a feature included in the HD map based on the image data and the radar data; and
modifying a Global Positioning System (GPS) data including the HD map according to a feature determined based on a confidence score of extracted feature and a probability distribution,
wherein the extracting comprises: extracting a plurality of features based on the image data and the HD map, and
wherein the modifying comprises: measuring a plurality of offsets by matching the plurality of features with the radar data.
10. The controlling method of a driver assistance system according to claim 9 , wherein the modifying comprises:
obtaining the confidence score and the probability distribution based on Integrated Probabilistic Data Association (IPDA); and
calculating a weight for the plurality of feature information determined based on the confidence score and the probability distribution.
11. The controlling method of a driver assistance system according to claim 10 , wherein the modifying comprises:
updating an offset estimation value included in the weight based on a Kalman filter.
12. The controlling method of a driver assistance system according to claim 11 , wherein the modifying comprises:
matching the updated offset estimation value with the HD map; and
stopping the modification of the GPS data based on the matching result.
13. The controlling method of a driver assistance system according to claim 11 , wherein the modifying comprises:
matching the updated offset estimation value with the HD map, and
predicting the offset estimation value with an offset of the current time point at an offset of a current point based on the matching result and a dynamics model.
14. The controlling method of a driver assistance system according to claim 13 , wherein the modifying comprises:
matching the offset of the current point with the HD map; and
stopping the modification of the GPS data based on the matching result.
15. The controlling method of a driver assistance system according to claim 13 , wherein the modifying comprises:
matching the offset of the current point with the HD map; and
modifying the GPS data based on the matching result and the offset of the current point.
16. The controlling method of a driver assistance system according to claim 15 , further comprises:
determining a location of the vehicle in the HD map based on the modified GPS data.