Search

This paper addresses road anomaly detection by formulating it as a classification problem and applying deep learning approaches to solve it. Besides conventional road anomalies, additional ones are introduced from the perspective of a vehicle. In order to facilitate the learning process, the paper pays a close attention to pattern representation, and proposes three sets of numeric features for representing road conditions. Also, three deep learning approaches, i.e. Deep Feedforward Network (DFN), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN), are considered to tackle the classification problem. The detectors, with respect to the three deep learning approaches, are trained and evaluated through data collected from a test vehicle driven on various road anomaly conditions. The comparison study on the detection performances is conducted by setting key hyper-parameters to certain sets of fixed values. Also, the comparison study on performances of each detector with respect to different pattern representations is conducted. The results have shown the effectiveness of the proposed approaches and the efficiency of the proposed feature representations in road anomaly detection.

To address the defocusing problem faced by random stepped frequency (RSF) radars, this paper presents a complementary code cancellation (CCC) method to estimate the target velocity and achieve motion compensation. The proposed CCC method is capable of eliminating the coupling effect between range profile and target velocity. In this paper, we first give a block diagram of RSF radar modulated by the M-sequence generated pseudorandom code, and introduce the baseband sampling echo model of moving target. Then, the velocity estimation accuracy is derived to show the sensitivity of high-resolution range profile to target velocity. Subsequently, the CCC method is proposed and also investigated in the application of multi-target scenario. Performance analyses demonstrate that the proposed method can satisfy the estimation accuracy requirement and improve the signal to noise ratio by the velocity accumulation. Finally, simulations show that the method is effective and more computationally efficient than the existing popular methods.