Multi-feature weighted nearest neighbor data association and tracking algorithm based on Kalman Filter

Clc Number:

Fund Project:

The National Natural Science Foundation of China (General Program, Key Program, Major Research Plan)

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments

    Aiming at the problem that the traditional nearest neighbor data association algorithm has low accuracy and is prone to missing the association, this paper proposes a multi-feature weighted nearest neighbor association algorithm. This algorithm defines a similarity function based on the obstacle data obtained by the intelligent vehicle environment perception system, and proposes a method to calculate the effective correlation degree based on the life cycle, so as to determine whether the objects are related. Then, based on Kalman filter, the associated target is updated iteratively to realize the tracking of the target. Finally, the tracking trajectory of stationary target, low-speed moving target without interaction and low-speed moving target with interaction are compared through experiments. Experimental results show that compared with the conventional nearest neighbor data association algorithm, the improved algorithm proposed in this paper, can realize accurate continuous connection of the low-speed moving target tracking, there will be no target lost or position mutation phenomenon, and the interaction and hiding between the targets have little effect on the tracking performance, so it has high validity and practicality.

    Cited by
Get Citation
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
  • Received:March 24,2020
  • Revised:May 20,2020
  • Adopted:June 02,2020
  • Online:
  • Published: