Online trajectory planning of bistatic SAR based on regularised extreme learning machine
Zhifeng Luo, Zhichao Sun, Huarui Sun, Junjie Wu, Jianyu Yang · IET conference proceedings. · 2024
As a special bistatic configuration, the maneuvering platform bistatic SAR using GEO transmitter (GEO-BiSAR) can provide long-duration and wide beam coverage over the target scene. This bistatic configuration can continuously observe targets during flight and obtain target recognition and tracking information from images. The imaging performance of bistatic SAR depends on the observation geometry, which is determined by the trajectory of the maneuvering platform receiver. Therefore, it is necessary to design the flight trajectory of the receiver. However, during the flight, the trajectory of the receiver is vulnerable to external disturbances and generates certain offsets. The original trajectory can't meet the actual imaging requirements, and a new trajectory needs to be re-planned in real time. To address the above issues, this paper models the problem as a multi-objective optimisation problem and proposes a new framework for online trajectory planning using Regularised Extreme Learning Machine (RELM). The framework first obtains the planned trajectory as a dataset through multi-objective optimisation. Then train the dataset into RELM to achieve online trajectory planning, and the results meet the requirements of real-time and effectiveness. Experimental studies demonstrate the rapidity and effectiveness of the proposed method.