OBJECT FUSION SYSTEM OF MULTIPLE RADAR IMAGING SENSORS

• US 20150198711A1
• Filed: 01/16/2014
• Published: 07/16/2015
• Est. Priority Date: 01/16/2014
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: ZENG, SHUQING, SALINGER, JEREMY A., O'DEA, KEVIN A., MURAD, MOHANNAD, NICKOLAOU, JAMES N., LITKOUHI, BAKHTIAR B., PAZHAYAMPALLIL, JOEL
• Granted Patent: US 9,255,988 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01S 13/66 Radar-tracking systems; Analogous systems
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/878 Combination of several spaced transmitters or receivers of known location for determining the position of a transponder or a reflector G01S13/874 takes precedence
• CPC Class Codes: G01S 13/9029 specially adapted for moving target detection within a single SAR image or within multiple SAR images taken at the same time
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 2013/93271 in the front of the vehicles
• CPC Class Codes: G01S 2013/93272 in the back of the vehicles

Abstract

A method of detecting and tracking objects using multiple radar sensors. Objects relative to a host vehicle are detected from radar data generated by a sensing device. The radar data includes Doppler measurement data. Clusters are formed, by a processor, as a function of the radar data. Each cluster represents a respective object. Each respective object is classified, by the processor, as stationary or non-stationary based on the Doppler measurement data of each object and a vehicle speed of the host vehicle. Target tracking is applied, by the processor, on an object using Doppler measurement data over time in response to the object classified as a non-stationary object; otherwise, updating an occupancy grid in response to classifying the object as a stationary object.

Claims

1. A method of detecting and tracking objects using multiple radar sensors comprising:

detecting objects relative to a host vehicle from radar data generated by a sensing device, the radar data including Doppler measurement data;

forming clusters, by a processor, as a function of the radar data, each cluster representing a respective object;

classifying, by the processor, each respective object as stationary or non-stationary based on the Doppler measurement data of each object and a vehicle speed of the host vehicle; and

applying target tracking, by the processor, on an object using Doppler measurement data over time in response to the object classified as a non-stationary object, otherwise, updating an occupancy grid in response to classifying the object as a stationary object.

2. The method of claim 1 wherein forming clusters comprises the following steps:

identifying each point detected by the first sensing device, each point including a location position and an associated range relative to the host vehicle;

assigning each point as a separate cluster;

comparing neighboring points and merging neighboring points into a same cluster if similarity metrics between the neighboring points is within a similarity threshold.

3. The method of claim 2 further comprising the step of indexing each point using a k-d tree.

4. The method of claim 3 further comprising the step of indexing each point using a hash look-up tree.

5. The method of claim 3 wherein the similarity metrics includes Doppler measurement data.

6. The method of claim 1 wherein classifying each respective object as stationary or non-stationary comprises the following steps:

identifying a velocity for each point within a cluster;

identifying a unit vector for each point with the cluster;

determining a range rate residue value for each point within the cluster;

determining that the cluster is a stationary cluster if a predetermined percent of the range rate residue values are within a residue threshold;

otherwise determining that the cluster is a dynamic cluster.

7. The method of claim 6 wherein the velocity of each point within the cluster is determined by the following equation:

ν

_xi=y_iω

_H−

ν

_H

ν

_yi=−

x_iφ

_H

where ν

_xi is a lateral velocity of an i^th point, ν

_yi is a longitudinal velocity of the i^th point, y_i is a longitudinal coordinate relative to the vehicle of the i^th point, x_i is a latitudinal coordinate relative to the vehicle of the i^th point, ω

_H is yaw rate, and, ν

_H is a speed.

8. The method of claim 7 wherein the range rate residue value is determined by the following equation:

ε

_i=|{right arrow over (v)}_i·

{right arrow over (n)}_i−

d_i|,

where ν

_i is the determined speed, and n_i is a unit vector.

9. The method of claim 6 wherein the predetermined percentage is substantially 80%.

10. The method of claim 1 wherein applying tracking comprises the following steps:

determining an orientation and position of the cluster at a previous instance of time using radar data;

determining an orientation and position of the cluster at a current instance of time using radar data;

determining a translation velocity in response to the orientation and position of the cluster at the previous and next instance of time;

updating an object tracking model of the cluster at the current instance of time utilizing the translation velocity.

11. The method of claim 10 wherein the transformation velocity is determined by the following equation:

12. The method of claim 11 wherein the related velocity for the model is determined by the following formula:

ν

_j=ν

_t+(m_j−

o)×

ω

where (ν

_t, ω

) are translation and angular velocities in ν

.

13. The method of claim 12 wherein the object tracking model for the current instance of time is updated as a function of the transformation velocity determined for a current instance of time.

14. The method of claim 13 wherein an estimated motion of the cluster at a next instance of time is determined as a function of the updated object tracking model for the current instance of time.

15. The method of claim 14 wherein a Bayesian-based technique is applied to recursively estimate each object tracking model and each estimated motion.

16. The method of claim 10 wherein the orientation and position of the cluster is determined by a wheel hub detection technique comprising the steps of:

determining a baseline width between a front wheel hub and a rear wheel hub or a same side of a target vehicle;

computing an orientation of target vehicle relative to the host vehicle, the orientation based on the an alignment of the first object detection device and a baseline vector between the front and rear wheel hubs;

updating a position, orientation, and velocity of the target vehicle by tracking orientation and baseline of the target vehicle relative to the host vehicle.

17. The method of claim 1 wherein a plurality of sensing devices are used for detecting objects, wherein the data from the plurality of sensing devices are fused for updating an object model and tracking targets.