阅读说明：本技术 用于车道检测的设备 (Apparatus for lane detection ) 是由 尤金·沙弗尔 于 2018-05-03 设计创作，主要内容包括：本发明提供了一种用于机动车辆驾驶员辅助系统的设备。该设备被配置为优化物体群集,其中每个物体群集包括针对车辆附近的至少一个物体的位置测量值序列。最初,在预聚类阶段,所测量的物体位置到物体群集的分配可以基于所测量物体位置的相对接近度。该设备基于第一诊断来识别恶意物体群集,并根据恶意物体群集内的测量值来识别恶意物体轨迹。来自恶意物体轨迹的位置测量值已从群集中移除,并且将恶意物体群集中的其余位置测量值重新分配给其他物体群集。移除恶意物体群集。因此,优化物体群集。(The invention provides an arrangement for a driver assistance system for a motor vehicle. The apparatus is configured to optimize clusters of objects, wherein each cluster of objects comprises a sequence of position measurements for at least one object in the vicinity of the vehicle. Initially, in the pre-aggregation class stage, the assignment of measured object positions to object clusters may be based on the relative proximity of the measured object positions. The device identifies clusters of malicious objects based on the first diagnosis and identifies malicious object trajectories from measurements within the clusters of malicious objects. The position measurements from the malicious object trajectories have been removed from the clusters and the remaining position measurements in the malicious object clusters are reassigned to other object clusters. Clusters of malicious objects are removed. Thus, object clustering is optimized.)

1. An apparatus for a driver assistance system for a motor vehicle, the apparatus being operable to initialize and optimize a plurality of object clusters,

each cluster of objects comprising at least one object trajectory, wherein each object trajectory comprises a plurality of sequential measurements of a position of a respective object located in the vicinity of the motor vehicle, and each object trajectory has a trajectory length,

the device is configured to perform the steps of:

a) assigning the position measurements of the plurality of object trajectories to a set of object clusters, wherein each object cluster comprises at least one position measurement from at least one of the object trajectories,

b) calculating a respective value for a first diagnosis for each cluster of objects in the set;

c) identifying clusters of malicious objects based on the value of the first diagnosis;

d) identifying an object track with the longest length as a malicious object track from object tracks assigned to the malicious object cluster from position measurement values;

e) removing a sequence of position measurements corresponding to the malicious object trajectory from all object clusters;

f) reassigning any remaining position measurements previously assigned to the cluster of malicious objects to other clusters of objects in the set, and;

g) removing the clusters of malicious objects from the set, thereby optimizing the clusters of objects.

2. The device of claim 1, further configured to repeat steps b) through g) until no malicious object clusters are identified.

3. The apparatus of claim 1 or claim 2, further configured to determine in step b) whether each cluster of objects merges with each other cluster of objects.

4. The apparatus according to any of the preceding claims, further configured to determine, for each cluster of objects in the set, whether said each cluster of objects is an internal cluster of objects or an external cluster of objects.

5. The apparatus of claim 4, wherein the value of the first diagnosis for each internal object cluster is the number of other internal object clusters that said each internal object cluster has merged with, and;

wherein the value of the first diagnosis for each cluster of external objects is set to zero.

6. The apparatus according to any of the preceding claims, wherein the cluster of malicious objects is the cluster of objects having the only highest value of the first diagnosis.

7. The apparatus of any of the preceding claims, further configured to calculate a value of a second diagnosis at least for each object cluster having the same value of the first diagnosis, in case both object clusters have the same value of the first diagnosis, and wherein the identification of the malicious object clusters is further based on the value of the second diagnosis.

8. The apparatus of claim 7, wherein the value of the second diagnosis for each object cluster is equal to the number of contacts of said each object cluster with other object clusters.

9. The apparatus of claim 8, wherein the cluster of malicious objects is identified as the cluster of objects having the unique highest value of the second diagnosis.

10. The apparatus of claim 9, further configured to, in the event that two object clusters have the same value of the second diagnosis, calculate a value of a third diagnosis at least for each object cluster having the same value of the second diagnosis, and wherein the identification of the malicious object cluster is further based on the value of the third diagnosis.

11. The apparatus of claim 10, wherein a value of the third diagnosis for each object cluster is equal to a number of parallelisms of the each object cluster with other object clusters.

12. The apparatus of claim 11, wherein the cluster of malicious objects is identified as the cluster of objects having the lowest unique value of the third diagnosis.

13. A method for a driver assistance system of a motor vehicle for initializing and optimizing a plurality of object clusters,

each of the object clusters comprising at least one object trajectory, wherein each object cluster comprises a plurality of sequential measurements of a position of a respective object located in the vicinity of the motor vehicle, and each object trajectory has a trajectory length,

the method comprises the following steps:

a) assigning the position measurements of the plurality of object trajectories to a set of object clusters, wherein each object cluster comprises at least one position measurement from at least one of the object trajectories,

b) calculating a respective value for a first diagnosis for each cluster of objects in the set;

c) identifying clusters of malicious objects based on the value of the first diagnosis;

d) identifying an object track with the longest length as a malicious object track from object tracks assigned to the malicious object cluster from position measurement values;

e) removing a sequence of position measurements corresponding to the malicious object trajectory from all object clusters;

f) reassigning any remaining position measurements included in the cluster of malicious objects to other clusters of objects in the set, and;

g) removing the cluster of malicious objects from the set,

thereby optimizing the object clustering.