Traffic holographic perception refers to the real-time, high-fidelity, and multi-dimensional sensing of traffic states through the fusion of heterogeneous sensors, including cameras, radars, and connected vehicle data. The multi-source perception data obtained thereby can provide a complete digital representation of the road network for the Intelligent Transportation System (ITS). However, sensors are vulnerable to environmental interference, which can result in data loss at specific points or along arterial highways for certain periods, potentially undermining system safety and decision-making reliability. To address these challenges, a deep learning method based on Graph Convolutional Networks (GCN) and Gated Recurrent Units (GRU) is proposed, leveraging Artificial Intelligence (AI) and intelligent connected technologies for real-time acquisition of multi-sensor perception data. A feature-level fusion integrates multi-source perception data. GCN captures spatial dependencies from the road network topology, while GRU extracts temporal features from time series, enabling accurate imputation of missing traffic data. The method is evaluated at intelligent connected intersections in the Beijing High-level Autonomous Driving Demonstration Area. Results show that the accuracy of long-term traffic state completion reaches 89.36%, and the Root Mean Square Error (RMSE) is reduced by 17.2% compared to the Long Short-Term Memory (LSTM) baseline. This framework provides a practical solution for deploying traffic holographic perception technology in secure and trustworthy ITS.
- Article type
- Year
Open Access
Article
Issue
The rapid development of the intelligent transportation industry brings new opportunities and challenges for talent cultivation. The traditional cultivation system and teaching platforms for transportation majors face the following problems: 1) The cultivation modes of talents are not suitable for the rapid development of the industry. 2) The teaching methods are not coordinated with the development of industrial technologies. 3) The construction of a practical teaching platform is not synchronized with industrial development. 4) There is no sustainable and effective cooperation mechanism in the integration of industry and education. To solve the above problems, a teaching reform method based on the digital twin is proposed in this paper.
First, this paper analyzed the demand for transportation innovation talents and the significance of technological innovation capabilities, interdisciplinary comprehensive capabilities, and industry foresight capabilities. Meanwhile, a double-closed-loop innovative talent cultivation system of the integration of industry and education was established. One closed loop was theoretical teaching and professional skills practice, and the other closed loop was industrial demand and school supply. Second, a transportation digital twin method was proposed based on the five-dimensional model of the digital twin. The five dimensions of physical space, cyberspace, service system, digital twin data, and connection were deeply integrated with traffic scenarios. Based on the transportation digital twin method, this paper combined hardware (e.g., intelligent roadside devices, driving simulators, and miniature intelligent vehicles) with software (e.g., PreScan, Matlab, and PyCharm) to build the transportation digital twin experimental platform. The platform comprised three parts: the real traffic road, the cyberspace, and the semiphysical traffic simulation scene. Using this platform, real field tests can be replaced by digital twin experiments in the laboratory. Third, a typical experimental example of autonomous driving was selected to demonstrate the operating steps, including data collection and preprocessing, data conversion and traffic scenario construction, and autonomous driving algorithm development. Finally, six indexes were selected to evaluate the transportation digital twin teaching reform method, including learning attitude, cognition of majors, practical participation, course satisfaction, academic performance, and science and technology competition.
The final evaluation results showed the following: 1) The proposed method can meet the learning needs of students, allow students to have a deep understanding of the practical applications of transportation professional courses, and improve their learning attitudes and cognition of majors. 2) Cyber space can provide students with repeated practice opportunities, effectively stimulating their enthusiasm for innovation and creativity, thereby improving their practical participation and course satisfaction. 3) There is a positive effect in promoting students’ class performance, experimental results, and exam results, among others. This plays an important role in science and technology competitions.
The proposed teaching reform method based on the digital twin solved the problems existing in traditional transportation talent training systems and teaching platforms, which is of great significance for the cultivation of innovative talents of transportation majors.
京公网安备11010802044758号