The computing power of edge devices is difficult to keep up with the development of modern computer technology, and the computing power is not improved enough. In practical application environments, so there is the birth of lightweight models. Lightweight models specifically refer to some models with simple model architecture and low computational load. Although the lightweight model is fast and the model is relatively simple, the detection effect is not very good. This paper proposes a parallel convolution module, performs feature fusion through parallel processing of multi-scale convolution kernels, and then integrates the spatial channel attention mechanism into the module to implement a multi-scale target detection module on a single feature layer. Model fusion proposes anchor boxes to generate heads and become a single-stage object detection model. © 2022 IEEE.

Speed bumps and expansion joints widely available on pavements cause pulse-like vibrations that can be measured by an in-vehicle accelerator. Such vibration patterns are called road-pulses in this paper, and we proposed a sliding Gaussian model (SGM) for detection. The detected road-pulses are associated with accurate positions and image features for road-pulse map construction. In the localization step, road-pulse patterns are detected and matched with the road-pulse map. And the corresponding image features and positions stored in the map allow accurate vehicle localization afterward. The proposed road-pulse-based localization is integrated into a downward pavement visual odometry with a Kalman filter based on the second-order Markov model (KF-MM2) for enhancing vehicle localization. The proposed method was validated on two different pavement conditions. Experimental results demonstrate that the proposed method can greatly improve vehicle localization with low cost and high feasibility. © 2021 IEEE.

The Yangtze River is the first of China and the third-longest river in the world. It is the most developed inland water transportation system in China. Chongqing, located in the upstream of the Three Gorges Project, is not only the gateway of the industrial and commercial center of China but also the most abundant water and intermodal inland hub and material distribution center in the southwest. Frequent water traffic accidents pose severe threats to the safety of human life and property and the water environment as well as bring adverse effects on social stability. Therefore, based on the ten years of statistical data of ship accidents in 2009-2018 from the Chongqing Maritime Safety Administration (MSA), this paper summarizes the characteristics of the spatiotemporal distribution of accidents through statistical analysis of the historical data. Moreover, it proposes accident prevention and supervision methods to provide decision support for maritime safety. This research is of great significance to the prevention and control of water traffic accidents in this region. © 2020 IEEE.

In this paper we propose a method of millimeter wave radar installation correction. The two main processes of installation correction are the acquisition of the true position of the target and the establishment and solution of the relationship between the true position and the position detected by the millimeter wave radar. First, we designed and made a 10∗30m calibration site to get the real position of the target. Second, based on the principle of stereo calibration in computer vision, the relationship between the millimeter wave radar and the car front bumper center coordinate system is constructed, and then an algorithm based on the Levenberg-Marquardt optimization is used to solve the objective function. Finally, a random sampling method is used to collect data to verify the effectiveness of the proposed method. The experimental results show that the proposed method can effectively correct the installation deviation of millimeter wave radar. © 2020 IEEE.

Intelligent vehicle formation can improve the driving status of vehicle platoon, increase road traffic flow, reduce traffic accidents, reduce environmental pollution and save energy. As an important part of cooperative platoon control system, cooperative platoon longitudinal speed following control has an important influence on the overall performance of vehicle platoon. To solve the longitudinal speed following control problem of cooperative vehicle platoon, a longitudinal following control strategy based on nonlinear proportion, integration, differentiation (PID) is proposed on the premise of fully considering the delay of vehicle actuator, the delay of vehicle communication and vehicle platoon stability. According to the communication topology structure of Predecessor-Leader Following, the information of platoon state is obtained, and the strategy of constant spacing is adopted. A layered vehicle platoon longitudinal controller is designed based on nonlinear PID control. The expected speed and expected acceleration are obtained by calculation, and the throttle opening or brake pressure is obtained according to the vehicle inverse dynamics model and PID feedback control to realize the longitudinal following control of the vehicle platoon. To verify the validity of the controller, co-simulation with MATLAB/Simulink and CarSim. The simulation results show that compared with the traditional PID control, the speed extremum, acceleration extremum and vehicle spacing extremum of the nonlinear PID controller designed in this paper can be improved by 0.14 m/s and 0.28 m/s, 0.05 m/s2 and 0.09 m/s2, 0.26 m and 0.22 m under two different conditions, which can effectively improve the longitudinal following state of the platoon, realize the speed cruise and formation maintenance of the vehicle platoon. © 2020 SAE International.

As an important role in autonomous vehicles or advanced driving assistance systems, lane detection uses the onboard camera high up on the windshield to provide the vehicle’s lateral offset within its own lane in a real-time, low-cost way. In this paper, we propose an efficient, robust lane detection method based on histogram of oriented vanishing points. First, the lane features are extracted by symmetrical local threshold. Then, the lines are generated from oriented vanishing points. The lines crossing most features are selected and oriented vanishing points are updated by the overlap between features and selected lines. Last step will be repeated for getting stable oriented vanishing points. Therefore, the last selected lines are most likely to be lane lines. Finally, Validate and select the best lane lines. The proposed method has been tested on a public dataset. The experimental results show that the method can improve robustness under real-time automated driving. © 2020, Springer Nature Singapore Pte Ltd.

Cooperative adaptive cruise control (CACC) has important significance for the development of the connected and automated vehicle (CAV) industry. In this paper, a learning control method combined Deep Deterministic Policy Gradient and Proportional-Integral-Derivative (DDPG-PID) controller is proposed. The main contribution of this study is automating the PID weight tuning process by formulating this objective as a deep reinforcement learning (DRL) problem. Based on the Hardware-in-the-Loop (HIL) simulation platform, the DDPG-PID controller is compared with the conventional PID controller under the test condition. Experiment results indicate that on 38.95% stability time in vehicular platooning system is decreased by utilizing the proposed method. The performance of maximum distance error is also improved efficiently, which is reduced by 60.94%. The research in this paper is a further development of learning control method and provides a new idea for the practical application of DRL algorithm in industrial field. © 2020 IEEE.

The Yangtze River is the main artery of the inland water transportation of China, connecting the resource-rich southwest hinterland, the Central China region and the economically prosperous southeast coast of East China. Chongqing, located in the fluctuating backwater of the Three Gorges Project, is the mainstream of the Yangtze River and the Jialing River. It is not only the industrial and commercial center of China but also the most abundant water and intermodal inland hub and material distribution center in the southwest. Based on the ten years of statistical data of ship accidents from 2009 to 2018 of the Chongqing Maritime Safety Administration (MSA), this paper takes the navigation section of the fluctuating backwater area of the Three Gorges Reservoir as the research object, and comprehensively uses statistical analysis, comparative analysis, and data visualization methods. The temporal and spatial distribution characteristics of ship accidents in the fluctuating backwater area and the economic losses in this area are studied. The research results are helpful for strengthening maritime management, improving navigation conditions, and reducing water traffic accidents in the Three Gorges Reservoir area and other similar reservoir areas all over the world. In addition, the research gives detailed guidance for stakeholders and mangers for the prevention of water traffic accidents. © ESREL2020-PSAM15 Organizers.Published by Research Publishing, Singapore.

Intelligent vehicle formation can improve the driving status of vehicle platoon, increase road traffic flow, reduce traffic accidents, reduce environmental pollution and save energy. As an important part of cooperative platoon control system, cooperative platoon longitudinal speed following control has an important influence on the overall performance of vehicle platoon. To solve the longitudinal speed following control problem of cooperative vehicle platoon, a longitudinal following control strategy based on nonlinear proportion, integration, differentiation (PID) is proposed on the premise of fully considering the delay of vehicle actuator, the delay of vehicle communication and vehicle platoon stability. According to the communication topology structure of Predecessor-Leader Following, the information of platoon state is obtained, and the strategy of constant spacing is adopted. A layered vehicle platoon longitudinal controller is designed based on nonlinear PID control. The expected speed and expected acceleration are obtained by calculation, and the throttle opening or brake pressure is obtained according to the vehicle inverse dynamics model and PID feedback control to realize the longitudinal following control of the vehicle platoon. To verify the validity of the controller, co-simulation with MATLAB/Simulink and CarSim. The simulation results show that compared with the traditional PID control, the speed extremum, acceleration extremum and vehicle spacing extremum of the nonlinear PID controller designed in this paper can be improved by 0.14 m/s and 0.28 m/s, 0.05 m/s2 and 0.09 m/s2, 0.26 m and 0.22 m under two different conditions, which can effectively improve the longitudinal following state of the platoon, realize the speed cruise and formation maintenance of the vehicle platoon. © 2020 SAE International.