Gyroscopy and Navigation
N.V. Panokin et al. Application of Sparse Representation of Complex Data in Railway Positioning and Collision Alert Systems Using Millimeter-Wave Radar
The paper presents the results from the experimental study of a modified artificial neural network MFNN (minimum fuel neural network). Sparse representation of complex data with overcomplete basis and L0/L1 norm optimization are used instead of the classical fast Fourier transform (FFT) algorithm. The results showed a significant enhancement in the ability of obstacle recognition and autonomous railway control systems to distinguish between close objects such as trains on adjacent tracks at marshalling yards.
Keywords: railway transport, obstacle recognition, radar, angular resolution, artificial neural network, MFNN, overcomplete basis, L0 norm, L1 norm.
| About authors |
N.V. Panokin Moscow Polytechnic University, Moscow, Russia ORCID 0000-0001-8680-9510. I. A. Kostin Moscow Polytechnic University, Moscow, Russia ORCID 0000-0002-9069-9198. A. V. Averin Moscow Polytechnic University, Moscow, Russia A. V. Karlovskii Moscow Polytechnic University, Moscow, Russia ORCID 0000-0001-7660-3375. D. I. Orelkina Moscow Polytechnic University, Moscow, Russia A. Yu. Nalivaiko Moscow Polytechnic University, Moscow, Russia ORCID 0000-0003-2475-4811. |
N.V. Panokin, I. A. Kostin, A. V. Averin, A. V. Karlovskii, D. I. Orelkina, and A. Yu. Nalivaiko. Application of Sparse Representation of Complex Data in Railway Positioning and Collision Alert Systems Using Millimeter-Wave Radar / Gyroscopy and Navigation, 2024, Vol. 15, No.1, pp. 61-68
