Winter airport snow removal is critical for ensuring operational safety and efficiency. Although autonomous snowplows have the potential to perform reliably in harsh conditions, excessive wheel slip on icy surfaces can lead to tracking errors and safety hazards. This paper proposes a practical terminal sliding mode control strategy, augmented by a nonlinear disturbance observer, to simultaneously constrain the slip ratio and achieve precise trajectory tracking. The approach introduces a comprehensive tracking error model that incorporates tire–road adhesion dynamics and uses feedback linearization for model decoupling. A robust disturbance observer estimates unknown snow removal resistance, and a high-order terminal sliding mode controller guarantees finite-time convergence of tracking errors while keeping the slip ratio within safe limits. Simulations in representative scenarios show that the proposed method outperforms two baseline algorithms, effectively suppressing slip and improving tracking performance. Real-world experiments in a vehicle prototype further validate the strategy, reducing the slip ratio from 7.4% to 1.8% and from 9.2% to 2.6% under different conditions compared with conventional methods. The results confirm that the proposed control scheme enhances both safety and efficiency for autonomous snow removal on variable road surfaces.

Recently many deep neural networks have been utilized to learn and extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (A-TR). However, in actual applications the types and amount of data that can be obtained are limited and difficult, which makes it hard to train the networks effectively. In this paper, we propose a multi-view deep learning framework combined with deformable convolution for SAR ATR. The scattering distribution characteristics and morphological characteristics of the target will be learned by the special structure of the deformable convolution, providing more sufficient information for subsequent fusion of features from the distinct views. Experimental results have shown the superiority of the proposed network based on the Moving and Stationary Target Acquisition and Recognition data set and the better recognition performance in the condition of a small number of raw SAR images.

Designing radar waveforms with desired autocorrelation properties is a key point in the development of cognitive radar. To solve the problem of concealing weak targets by strong targets in detection, we consider minimizing the weighted integrated sidelobe level (WISL) metric in frequency domain where the weak targets are located. In order to directly solve the complex non-convex optimization problem, an iteration algorithm based on the general framework of the iterative sequential quartic optimization (ISQO) algorithm that can guarantee fast convergence to a static point is developed. Numerical simulations are provided to assess the effectiveness of the proposed algorithm.