导航

ACTA AERONAUTICAET ASTRONAUTICA SINICA ›› 2022, Vol. 43 ›› Issue (S1): 726904-726904.doi: 10.7527/S1000-6893.2022.26904

• Swarm Intelligence and Cooperative Control • Previous Articles     Next Articles

Learning-based multi-rate multi-sensor fusion localization method

CHEN Bo, YUE Kai, WANG Rusheng, HU Mingnan   

  1. 1. College of Information Engineering, Zhejiang University of Technology, Hangzhou 310014, China
  • Received:2022-01-06 Revised:2022-01-20 Published:2022-03-11
  • Supported by:
    Key Research and Development Program of Zhejiang Province (2022C03029);Zhejiang Provincial Natural Science Foundation of China (LR20F030004)

Abstract: This paper concerned with a class of problems of moving target tracking with unknown target motion model and multi-sensor asynchronous sampling. A data-driven target tracking algorithm is proposed, which only relies on measurement information. To solve the problem of the unknown motion model, a distributed neural network structure is designed based on the measurement model and measurement range, then the mapping relationship between the observation data and the state variables is established based on the designed neural network. A compensation strategy based on the measurement data at the last sampling instant is introduced to solve the multi-rate and multi-sensor sampling asynchronous problem. We also construct a weight network model with time difference as the input feature to estimate the real target position by iterative learning. An experiment is given to show the superiority and effectiveness of the proposed method.

Key words: target tracking, multi-sensor fusion, neural network, multi-rate sampled-data, data driven

CLC Number: