Perception sensors based fusion system for vehicle control and method thereof

1. A system comprising:

one or more radar sensors configured to detect one or more targets around the host vehicle and associating a radar target identifier with each of said detected one or more targets;

one or more image sensors configured to detect one or more targets in the field of view of the host vehicle and associating a vision target identifier with each of said detected one or more targets; and

a processing unit comprising a processor coupled with a memory, the memory storing the instructions executable by the processor to:

receive information corresponding to detected one or more targets from the one or more radar sensors and the one or more image sensors, wherein the information comprises lateral distance, longitudinal distance and velocity components of the detected one or more targets from the host vehicle;

match each of the one or more targets detected by the one or more radar sensors with one or more targets detected by the one or more image sensors to identify a target as a matched target when said target is detected by both the one or more radar sensors and the one or more image sensors;

categorize the matched target as a locked target when said target is iteratively identified as the matched target for a pre-defined duration; and

track the locked target using information received from the one or more radar sensors, wherein said tracking is performed even in the event when said locked target is subsequently not detected by the one or more image sensors.

2. The system of claim 1 , wherein the target is identified as the matched target when location of said target detected by the one or more image sensors is within one or more of a lateral error band or a longitudinal error band of location of said target detected by the one or more radar sensors.

3. The system of claim 2 , wherein the longitudinal error band is larger than the lateral error band.

4. The system of claim 1 , wherein the processing unit maintains a record of radar target identifiers and vision target identifiers of each of a set of targets, which are identified as matched targets.

5. The system of claim 1 , wherein the processing unit further classifies each target of the set of targets into one or more classes based on attributes of the target.

6. The system of claim 5 , wherein the processing unit further identifies a single target from a set of targets, which are identified as matched targets, by:

separating each of the targets of the set of targets into either of lane targets or path targets based on class information of each of the targets;

selecting a dominant target from targets associated with each class of the one or more classes; and

selecting a single target from the selected dominant targets based on a priority level associated with each class of the one or more classes.

7. The system of claim 1 , wherein the processing unit provides control signals for controlling the host vehicle to an electronic control unit (ECU) operatively coupled with the processing unit and the host vehicle.

8. The system of claim 7 , wherein the processing unit provides the control signals based on attributes of the selected single target.

9. A method comprising:

receiving information corresponding to detected one or more targets from each of a radar based detection unit and a vision based detection unit, wherein the information comprises lateral distance, longitudinal distance velocity components of the detected one or more targets from the host vehicle, and wherein the radar based detection unit comprises one or more radar sensors to detect one or more targets around the host vehicle and the vision based detection unit comprises one or more image sensors to detect one or more targets in the field of view of the host vehicle;

matching each of the one or more targets detected by the radar based detection unit with one or more targets detected by vision based detection unit to identify a target as a matched target when said target is detected by both the radar based detection unit and the vision based detection unit;

categorizing the matched target as a locked target when said target is iteratively identified as the matched target for a pre-defined duration; and

tracking the locked target using information received from the radar based detection unit, wherein said tracking is performed even in the event when said locked target is subsequently not detected by the vision based detection unit.

10. The method of claim 9 , wherein each target of a set of targets is classified into one or more classes based on attributes of the target.

11. The method of claim 9 , wherein the target is identified as the matched target when location of said target detected by the vision based detection unit is within one or more of a lateral error band or a longitudinal error band of location of said target detected by radar based detection unit.

12. The method of claim 11 , wherein the longitudinal error band is larger than the lateral error band.

13. The method of claim 9 , further comprising maintaining a record of radar target identifiers and vision target identifiers of each of a set of targets, which are identified as matched targets.

14. The method of claim 9 , further comprising classifying each target of the set of targets into one or more classes based on attributes of the target.

15. The system of claim 14 , further comprising identifying a single target from a set of targets, which are identified as matched targets, by:

separating each of the targets of the set of targets into either of lane targets or path targets based on class information of each of the targets;

selecting a dominant target from targets associated with each class of the one or more classes; and

selecting a single target from the selected dominant targets based on a priority level associated with each class of the one or more classes.

16. The method of claim 9 , further comprising providing control signals for controlling the host vehicle to an electronic control unit (ECU) operatively coupled with the processing unit and the host vehicle.

17. The method of claim 16 , wherein providing the control signals is based on attributes of the selected single target.

18. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform a method, the method comprising:

receiving information corresponding to detected one or more targets from each of a radar based detection unit and a vision based detection unit, wherein the information comprises lateral distance, longitudinal distance velocity components of the detected one or more targets from the host vehicle, and wherein the radar based detection unit comprises one or more radar sensors to detect one or more targets around the host vehicle and the vision based detection unit comprises one or more image sensors to detect one or more targets in the field of view of the host vehicle;

matching each of the one or more targets detected by the radar based detection unit with one or more targets detected by vision based detection unit to identify a target as a matched target when said target is detected by both the radar based detection unit and the vision based detection unit;

categorizing the matched target as a locked target when said target is iteratively identified as the matched target for a pre-defined duration; and

tracking the locked target using information received from the radar based detection unit, wherein said tracking is performed even in the event when said locked target is subsequently not detected by the vision based detection unit.

19. The non-transitory computer-readable medium of claim 18 , wherein the target is identified as the matched target when location of said target detected by the vision based detection unit is within one or more of a lateral error band or a longitudinal error band of location of said target detected by radar based detection unit, wherein the longitudinal error band is larger than the lateral error band.

20. The non-transitory computer-readable medium of claim 18 , wherein the method further comprises identifying a single target from a set of targets, which are identified as matched targets, by:

separating each of the targets of the set of targets into either of lane targets or path targets based on class information of each of the targets;

selecting a dominant target from targets associated with each class of the one or more classes; and

selecting a single target from the selected dominant targets based on a priority level associated with each class of the one or more classes.