Method and system for sensor fusion in vehicle

1 . A method for sensor fusion in a vehicle, the method comprising:

generating a track for a target vehicle based on data acquired through a surround view monitoring camera of a vehicle;

acquiring a plurality of pieces of first point data for an object around the vehicle;

identifying a plurality of pieces of second point data associated with the track for the target vehicle from among the plurality of pieces of first point data;

updating a sensor fusion track generated for the target vehicle using the plurality of pieces of second point data; and

control autonomous driving of the target vehicle based on the updated sensor fusion track;

selecting point data at a first location and point data at a second location among the plurality of pieces of second point data based on an angle formed by a straight line connecting each of the plurality of pieces of second point data from an origin and an X-axis in a vehicle coordinate system having a designated location in the vehicle as the origin,

wherein the updating comprises:

updating location information of the sensor fusion track based on information about the first location and the second location;

generating a two-dimensional equation representing a straight line connecting the first location and the second location;

identifying a first coordinate value of a first corner among corners of the sensor fusion track;

determining a second coordinate value of a corner corresponding to the first corner among corners of the target vehicle based on the two-dimensional equation and the first coordinate value;

calculating a third coordinate value corresponding to a location of a midpoint of a rear bumper of the target vehicle, thereby updating location information of the sensor fusin track;

identifying the second coordinate value having the same Y-axis coordinate value as a Y-axis coordinate value of the first coordinate value in the straight line connecting the first location and the second location;

calculating a heading angle of the sensor fusion track to be updated based on a slope of the straight line connecting the first location and the second location; and

calculating the third coordinate value based on a half value of a width of the track for the target vehicle and the calculated heading angle.

2 . The method according to claim 1 , wherein the sensor fusion track is pre-generated for the target vehicle based on sensor data acquired through a sensing device of the vehicle.

3 . The method according to claim 1 , wherein the target vehicle is located on a side or rear side of the vehicle.

4 . The method according to claim 3 , wherein the identifying comprises:

calculating a distance value from a center point of the track for the target vehicle to any one corner of the track for the target vehicle based on a width and a length of the track for the target vehicle; and

identifying the plurality of pieces of second point data associated with the track for the target vehicle from among the plurality of pieces of first point data based on the distance value.

5 . The method according to claim 4 , wherein the identifying comprises:

determining the distance value as a radius; and

identifying the plurality of pieces of second point data located within the radius from the center point of the track for the target vehicle from among the plurality of pieces of first point data.

6 . The method according to claim 1 , wherein:

the point data at the first location comprises point data having a smallest value as the angle among the plurality of pieces of second point data, and

the point data at the second location comprises point data having a largest value as the angle among the plurality of pieces of second point data.

7 . The method according to claim 1 , wherein the first corner comprises:

a corner located on a lower right side of a center of the sensor fusion track;

a corner located on a lower left side of the center of the sensor fusion track; or

a corner at a shortest distance from a straight line connecting the first location and the second location among the corners of the sensor fusion track.

8 . The method according to claim 1 , wherein:

the heading angle is calculated by applying a slope value of the two-dimensional equation as a variable of an arc tangent function,

an X-axis value of the third coordinate value is calculated based on a value calculated by applying the heading angle as a variable of a sine function, an X-axis value of the second coordinate value, and the half value of the width of the track for the target vehicle, and

a Y-axis value of the third coordinate value is calculated based on a value calculated by applying the heading angle as a variable of a cosine function, a Y-axis value of the second coordinate value, and the half value of the width of the track for the target vehicle.

9 . A system for sensor fusion in a vehicle, the system comprising:

a communication module including a transceiver configured to receive data from a sensing device of a vehicle;

a memory configured to store data of a sensor fusion track generated for a target vehicle; and

a processor electrically connected to the interface and the memory, wherein the processor is configured to:

generate a track for the target vehicle based on data of a surround view monitoring camera of the vehicle received through the interface;

acquire a plurality of pieces of first point data for an object around the vehicle;

identify a plurality of pieces of second point data associated with the track for the target vehicle from among the plurality of pieces of first point data; and

update the sensor fusion track using the plurality of pieces of second point data;

select point data at a first location and point data at a second location among the plurality of pieces of second point data based on an angle formed by a straight line connecting each of the plurality of pieces of second point data from an origin and an X-axis in a vehicle coordinate system having a designated location in the vehicle as the origin;

update location information of the sensor fusion track based on information about the first location and the second location;

generate a two-dimensional equation representing a straight line connecting the first location and the second location;

identify a first coordinate value of a first corner among corners of the sensor fusion track;

determine a second coordinate value of a corner corresponding to the first corner among corners of the target vehicle based on the two-dimensional equation and the first coordinate value;

calculate a third coordinate value corresponding to a location of a midpoint of a rear bumper of the target vehicle, thereby updating location information of the sensor fusin track;

identify the second coordinate value having the same Y-axis coordinate value as a Y-axis coordinate value of the first coordinate value in the straight line connecting the first location and the second location;

calculate a heading angle of the sensor fusion track to be updated based on a slope of the straight line connecting the first location and the second location; and

calculate the third coordinate value based on a half value of a width of the track for the target vehicle and the calculated heading angle.

10 . The system according to claim 9 , wherein the target vehicle is located on a side or rear side of the vehicle.

11 . The system according to claim 10 , wherein the processor is configured to:

calculate a distance value from a center point of the track for the target vehicle to any one corner of the track for the target vehicle based on a width and a length of the track for the target vehicle; and

identify the plurality of pieces of second point data associated with the track for the target vehicle from among the plurality of pieces of first point data based on the distance value.

12 . The system according to claim 11 , wherein the processor is configured to:

determine the distance value as a radius; and

identify the plurality of pieces of second point data located within the radius from the center point of the track for the target vehicle from among the plurality of pieces of first point data.

13 . The system according to claim 9 , wherein:

the point data at the first location comprises point data having a smallest value as the angle among the plurality of pieces of second point data, and

the point data at the second location comprises point data having a largest value as the angle among the plurality of pieces of second point data.

14 . The system according to claim 9 , wherein:

the heading angle is calculated by applying a slope value of the two-dimensional equation as a variable of an arc tangent function,

an X-axis value of the third coordinate value is calculated based on a value calculated by applying the heading angle as a variable of a sine function, an X-axis value of the second coordinate value, and the half value of the width of the track for the target vehicle, and

a Y-axis value of the third coordinate value is calculated based on a value calculated by applying the heading angle as a variable of a cosine function, a Y-axis value of the second coordinate value, and the half value of the width of the track for the target vehicle.