Recurrent neural networks (RNN) are efficient in modeling sequences for generation and classification, but their training is obstructed by the vanishing and exploding gradient issues. In this paper, we reformulate the RNN unit to learn the residual functions with reference to the hidden state instead of conventional gated mechanisms such as long short-term memory (LSTM) and the gated recurrent unit (GRU). The residual structure has two main highlights: firstly, it solves the gradient vanishing and exploding issues for large time-distributed scales; secondly, the residual structure promotes the optimizations for backward updates. In the experiments, we apply language modeling, emotion classification and polyphonic modeling to evaluate our layer compared with LSTM and GRU layers. The results show that our layer gives state-of-the-art performance, outperforms LSTM and GRU layers in terms of speed, and supports an accuracy competitive with that of the other methods.
This paper describes a dynamic trajectory planning method for lane-changing maneuver of connected and automated vehicles (CAVs). The proposed dynamic lane-changing trajectory planning (DLTP) model adopts vehicle-to-vehicle (V2V) communication to generate an automated lane-changing maneuver with avoiding potential collisions and rollovers during the lane-changing process. The novelty of this method is that the DLTP model combines a detailed velocity planning strategy and considers more complete driving environment information. Besides, a lane-changing safety monitoring algorithm and a lane-changing starting-point determination algorithm are presented to guarantee the lane-changing safety, efficiency and stability of automated vehicles. Moreover, a trajectory-tracking controller based on model predictive control (MPC) is introduced to make the automated vehicle travel along the reference trajectory. The field traffic data from NGSIM are selected as the target dataset to simulate a real-world lane-changing driving environment. The simulations are performed in CarSim-Simulink platform and the experimental results show that the proposed method is effective for lane-changing maneuver.
Monocular depth estimation is a foundation task of three-dimensional (3D) reconstruction which is used to improve the accuracy of environment perception. Because of the simpler hardware requirement, it is more suitable than other multi-view methods. In this study, a new monocular depth estimation algorithm based on graph convolution network (GCN) is proposed. The pixel-wise depth relationship is introduced into conventional convolution neural network (CNN) to make up the disadvantage of processing non-Euclidian data. And the remaining depth topological graph information on the spatial latent variables are extracted based on a multi-scale reconstruction strategy. The final results on NYU-v2 depth dataset and KITTI depth dataset demonstrate that our algorithm improves the quality of monocular depth estimation, especially there are several little objects coexisting in the scenes.
A binocular vision system is a common perception component of an intelligent vehicle. Benefiting from the biomimetic structure, the system is simple and effective. Which are extremely snesitive on external factors, especially missing vision signals. In this paper, a virtual view-generation algorithm based on generative adversarial networks (GAN) is proposed to enhance the robustness of binocular vision systems. The proposed model consists of two parts: generative network and discriminator network. To improve the quality of a virtual view, a generative network structure based on 3D convolutional neural networks (3D-CNN) and attentive mechanisms is introduced to extract the time-series features from image sequences. To avoid gradient vanish during training, the dense block structure is utilized to improve the discriminator network. Meanwhile, three kinds of image features, including image edge, depth map and optical flow are extracted to constrain the supervised training of model. The final results on KITTI and Cityscapes datasets demonstrate that our algorithm outperforms conventional methods, and the missing vision signal can be replaced by a generated virtual view.
Wind power ramp events (WPREs) are a common phenomenon in wind power generation. This unavoidable phenomenon poses a great harm to the balance of active power and the stability of frequency in the power supply system, which seriously threatens the safety, stability, and economic operation of the power grid. In order to deal with the impact of ramp events, accurate and rapid detection of ramp events is of great significance for the formulation of response measures. However, some attribute information is ignored in previous studies, and the laws and characteristics of ramp events are difficult to present intuitively. In this paper, we propose a visualization-based ramp event detection model for wind power generation. Firstly, a ramp event detection model is designed considering the multidimensional attributes of ramp events. Then, an uncertainty analysis scheme of ramp events based on the confidence is proposed, enabling users to analyze and judge the detection results of ramp events from different dimensions. In addition, an interactive optimization model is designed, supporting users to update samples interactively, to realize iterative optimization of the detection model. Finally, a set of visual designs and user-friendly interactions are implemented, enabling users to explore WPREs, judge the identification results, and interactively optimize the model. Case studies and expert interviews based on real-world datasets further demonstrate the effectiveness of our system in the WPREs identification, the exploration of the accuracy of identification results, and interactive optimization.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.