In order to improve the neural network structure and parameters set methods, on the basis of Glowworm Swarm Algorithm and the BP neural network, a Glowworm Swarm Algorithm to optimize the BP neural network algorithm is proposed. The algorithm uses the Glowworm Swarm Algorithm to obtain the better initial weights and thresholds of the network, to make up for the random defects of the BP neural network in the selection of connection weights and threshold and display the mapping ability of the generalization of BP neural network, and also make the BP neural network has fast convergence and strong learning ability. Applying the algorithm to the measured network flow algorithm and compared with the BP neural network and the Glowworm Swarm Algorithm to optimize the BP neural network, the simulation results show that the algorithm has higher forecast accuracy, thus proves the feasibility and effectiveness of the algorithm in the field of the forecast.

Urban network traffic congestion can be caused by disturbances, such as fluctuation and disequilibrium of traffic demand. This paper designs a distributed control method for preventing disturbance-based urban network traffic congestion by integrating Multi-Agent Reinforcement Learning (MARL) and regional Mixed Strategy Nash-Equilibrium (MSNE). To enhance the disturbance-rejection performance of Urban Network Traffic Control (UNTC), a regional MSNE concept is integrated, which models the competitive relationship between each agent and its neighboring agents in order to improve the decision-making process of MARL. The learning rate is enhanced with a self-adaptive ability to avoid a local optimal dilemma; Jensen-Shannon (JS) divergence is utilized to define the learning rate of the modified MARL. A two-way rectangular grid network with nine intersections is modeled via a Cell Transmission Model (CTM). A probability distribution mechanism, which can update the turn ratio of each approach dynamically and discretely, is established to represent the segmented route-decision process of the vehicles. The effectiveness of the proposed control method is evaluated through simulations in the grid network. The results show the influence of major disturbances, such as fluctuation of vehicle arrival rate, fluctuation of traffic demand (e.g. a rapidly rising flow and extreme changes in origin-destination distribution), and disequilibrium of traffic demand (e.g. different arrival flows at each boundary of the urban network), on the performance of the suggested control method. The results can be used to improve the state of the art in order to reduce urban network traffic congestion due to these disturbances.

Using a deep learning method to detect buried pipes on ground penetrating radar (GPR) data is popular for its excellent performance potential. However, this needs a large amount of high-quality training data, which leads to time-consuming and labor-intensive data annotation work. Semi-supervised learning method provides a solution for the situation of small sample size. However, the original semi-supervised learning methods lack control over the quality of pseudo labels during model training. To address the problem, a new prior knowledge-guided semi-supervised deep learning method is proposed to improve the model performance under the small sample. In the method, a prior structure feature (PSF) is constructed to control the quality of pseudo label during semi-supervised learning. The structure response index (SRI) is designed to segment the structure subject out for eliminating disturbance from unstructured information. Then, the PSF is represented using histograms of oriented gradients (HOGs) and Fourier descriptors (FDs) to describe the structure’s edge orientation and shape. Based on the PSF, LightGBM is employed as a discriminator to screen out high-quality pseudo labels for model training. In the experiments, comparison experiments with some state-of-the-art supervised learning methods and semi-supervised learning methods were carried out to validate the method’s performance. The ablation studies on labeled dataset size, pseudo-label confidence threshold, and PSF were conducted to validate the effectiveness of the prior structural feature in controlling the pseudo label’s quality. The results show that the PSF is effective, and the proposed method outperforms the other methods.

The Dynamic Traffic Assignment (DTA) is one of the important measures to alleviate urban network traffic congestion. The congestions are usually caused by stochastic traffic demands, which are generally unassignable from time dimension in the real-world but are assumed to be assignable in existing DTA methods (i.e. real-time travel demands). In this paper, a distributed DTA method for preventing urban network traffic congestion caused by stochastic real-time travel demands by improving Multi-Agent Reinforcement Learning (MARL). A team structure, which consists of decision-makers and advisers, is designed to learn parallelly in realistic DTA tasks. To reduce the size of the solution space adaptively, the dynamic critical values advised by adviser agents are adopted as constraints for the strategy space of decision-makers (i.e. main agents). A collaborative heterogeneous-adviser mechanism is designed to avoid deviation of guidance. To enhance the adaptability of DTA to the changeable external environment, the mixed strategy concept is introduced to improve the decision-making process of main agents. The respective mapping mechanisms are designed to define adaptive learning rates to improve the sensitivity of MARL. The Sioux Falls (SF) network is established as a test platform via a Dynamic Network Loading (DNL). The effectiveness of the suggested DTA method is assessed through numerical simulations SF network. Under the influence of the scenario with stochastic real-time travel demands, the results show that the proposed method outperforms in terms of the throughput of the network and the individual average travel time among the overall network. Additionally, the ability of the proposed method in response to the external environment rapidly has also been demonstrated. Adopting the suggested method can improve the state of the art to assign stochastic real-time travel demands dynamically and to avoid potential traffic congestion fundamentally.

In complex traffic environment, it is still a great challenge for Autonomous Vehicles (AVs) to understand the surrounding drivers’ Lane-Changing Decision (LCD) intention accurately. The LCD intention is affected by Driver’s Psychology (DP) and Driving Style (DS). But few LCD studies considered DP and DS simultaneously. We previously proposed a LCD model, by combing DP and DS, termed as DP&DS-LCD model. Nevertheless, there are some factors not fully considered in this model, including the driver’s Visual Attention (VA) in DP quantification and the influence of driving state on DS. Therefore, an enhanced LCD Model is developed, by integrating the DP under VA (DPVA) and DS Layering (DSL), named as DPVA&DSL-LCD model. In the model, a psychological field model coupling driver’s VA mechanism is established to represent the surrounding vehicles’ influence on the driver. Then, a DSL framework is proposed by adding the influence of driving state on DS. The Gaussian Mixture Model (GMM) clustering and Support Vector Machine (SVM) classifier are respectively adopted in training and recognition phases to identify the current driving style. Finally, integrating the DPVA and DSL, the Light Gradient Boosting Machine (LightGBM) algorithm is used to train the LCD model. In experiments, the open I-80 database from Next Generation Simulation (NGSIM) is adopted to train the DPVA&DSL-LCD. And compared with other three LCD models, the prediction performance of DPVA&DSL-LCD model achieved the best. Therefore, the DPVA&DSL-LCD model is effective and could provide support for the decision-making of AVs by predicting surrounding vehicles’ LCD intention.