Improved fuzzy clustering method based on millimeter wave radar

Abstract

The invention provides an improved fuzzy clustering method based on a millimeter wave radar, and the method comprises the following steps: obtaining the distance, angle and speed information of a sample point according to the echo data received by the radar; calculating a self-adaptive ellipse distance and local density of the sample points; drawing a two-dimensional decision diagram, and determining an initial cluster center; substituting into an FCM algorithm to perform primary fuzzy clustering processing, and updating a membership degree and a cluster center coordinate; repeating the steps until a primary fuzzy clustering iteration termination condition is met, and outputting a primary fuzzy clustering result; obtaining speed information of each cluster sample point; performing secondary fuzzy clustering according to an FCM algorithm, and updating a membership matrix and a speed median; judging whether a secondary fuzzy clustering result meets a secondary clustering iteration termination condition or not; and obtaining a final fuzzy clustering result. The invention is high in precision and good in robustness, adjacent vehicle targets can be correctly clustered, reliable and accurate target information is provided for processing such as vehicle tracking and traffic state estimation in a road, and the calculated amount of subsequent engineering is greatly reduced.

Description

technical field [0001] The invention relates to the field of radar signal processing, in particular to a clustering calculation method based on millimeter-wave radar data points of road moving vehicles. Background technique [0002] With the development of Intelligent Transportation System (ITS), multi-objective traffic radar system has become a current research hotspot. As a road traffic auxiliary tool, the traffic flow meter based on millimeter wave radar can monitor the movement status information of all vehicles in the detection area, and timely grasp the road vehicle information, which is beneficial to the traffic control department to control the road, reduce the occurrence of traffic accidents, and improve the safety index of the road. . [0003] Multi-target clustering is a key step in radar echo data processing. Quickly obtaining accurate clustering results can reduce the impact of clustering errors in subsequent tracking, ensure that the tracked is an effective ta...

AI Technical Summary

Problems solved by technology

However, the FCM algorithm needs to set the initial cluster center. If the parameter initialization is not appropriate, it may affect the correctness of the final clustering result. Secondly, when the data sample set is large, the real-time performance of the algorithm is not good.

Combined with the urban road traffic scene, when the vehicles on the road are very close to each other and the radar data sampling points are densely distributed, the problem that the traditional FCM algorithm cannot identify adjacent vehicles well

Images