Algorithm for ABR Traffic Control and Formation Feedback Information

The feedback control method proposed in this paper predicts the queue length in the switch using the slope of queue length prediction function and queue length changes in time-series. The predicted congestion information is backward to the node. NLMS and neural network are used as the predictive control functions, and they are compared from performance on the queue length prediction. Simulation results show the efficiency of the proposed method compared to the feedback control method without the prediction. Therefore, we conclude that the efficient congestion and stability of the queue length controls are possible using the prediction scheme that can resolve the problems caused from the longer delays of the feedback information.

Input-queued scheduling algorithms are designed to improve the throughput which use the input and output terminal in the router to obtain the maximum packets. And the length of the input queue is an important target, that can decides the superiority-inferiority about the capability of the input algorithm. With the existing algorithms, the main content about the scheduling algorithm are ensuring the packets’ high priority and low level. But most of these algorithms do not consider the length of queue. In this paper, we consider the change of queue length when the packets enter the input channel at the router, and put forward a shortest queue time-slot adapting algorithm (SQTA algorithm). First of all, SQTA algorithm can base the changes of transmittion time to control the packets’ sending time interval, and this action can make SQTA algorithm match the processing power about the router, and at the same time, this algorithm can reduce the length of the packets queue. Second, SQTA algorithm can control of a phenomenon that the packets may overflow when the length of queue is full, and that also make the packets transmitted more stable in network. The experimental result shows that this scheduling algorithm can control the length of queue effectively and reduce the waiting time with packets queue, It can be shorten the remaining service time, and can improve the network operation obviously.

Real time queue length estimation for signalized intersections using travel times from mobile sensors

Queue length is more likely to be affected by inefficient departure behaviour near the stop line at an intersection because of the prohibition of lane changes. The inefficient departure behaviour and its influence on the evolution of queue discharging in specific lanes have significant research value. In the past few years, the deployment of automatic vehicle identification (AVI) systems has made it possible to acquire information on vehicles’ passing through an intersection. Researchers can obtain individual driving information more accurately and effectively based on an analysis of AVI data. These high‐resolution data have enabled the reconstruction and quantitative analysis of the influence of microscopic traffic behaviour. This study aims to model the departure of vehicle platoon and predict the queue length at isolated intersections. They propose a multi‐layer gated recurrent unit (GRU) network to understand the mechanism of queue length evolution. A case study of vehicle departure pattern and queue length prediction is presented with AVI data obtained from an isolated intersection in Xuancheng, Anhui province, China for the first quarter of 2018. The results of the case study indicate that their work has good application prospects. The proposed multi‐layer GRU network has potential to guide the signal scheme optimisation.

Vehicle delay and queue length are quantitative measures of intersection performance. The technological advancement in the field of vehicle sensors and traffic controllers has reached a point where it has enabled efficient measurement of these performance measures. Two techniques are presented for real-time measurement of vehicle delay and queue length at a signalized intersection, and these automated delay and queue estimates are compared with manually ground-truthed measurement. These techniques were evaluated at an instrumented intersection in Noblesville, Indiana. The root-mean-square error by both techniques was below 0.7 veh-s for estimation of average delay and less than 0.15 vehicle for estimation of average maximum queue length, both on a cycle-by-cycle basis.

With the development of intelligent transportation system (ITS) and vehicle to X (V2X), the connected vehicle is capable of sensing a great deal of useful traffic information, such as queue length at intersections. Aiming to solve the problem of existing models’ complexity and information redundancy, this paper proposes a queue length sensing model based on V2X technology, which consists of two sub-models based on shockwave sensing and back propagation (BP) neural network sensing. First, the model obtains state information of the connected vehicles and analyzes the formation process of the queue, and then it calculates the velocity of the shockwave to predict the queue length of the subsequent unconnected vehicles. Then, the neural network is trained with historical connected vehicle data, and a sub-model based on the BP neural network is established to predict the real-time queue length. Finally, the final queue length at the intersection is determined by combining the sub-models by variable weight. Simulation results show that the sensing accuracy of the combined model is proportional to the penetration rate of connected vehicles, and sensing of queue length can be achieved even in low penetration rate environments. In mixed traffic environments of connected vehicles and unconnected vehicles, the queuing length sensing model proposed in this paper has higher performance than the probability distribution (PD) model when the penetration rate is low, and it has an almost equivalent performance with higher penetration rate while the penetration rate is not needed. The proposed sensing model is more applicable for mixed traffic scenarios with much looser conditions.

Intersections are critical elements of roadway infrastructure, often characterized by high levels of congestion and potential for severe crashes. To address increasing congestion and safety concerns, transportation agencies have begun implementing innovative geometric designs such as alternative intersections. These designs have also been called innovative intersections. This paper presents findings that evaluated the change in queue lengths of alternative intersections before and after their construction. The study involved multiple sites where alternative intersections were recently constructed, with data collected before construction and after completion. The key metric for this paper was queue lengths, with data collected using extensive video recording setups. For the two sites in Arizona with median U-turns (MUTs), the queues for the major road through movements were reduced even with an increase in volume. The signalized restricted crossing U-turn in Minnesota, the displaced left-turn (DLT) in Texas and in Virginia, and the hybrid intersection (mix of DLTs and MUTs) in Virginia all experienced improvements in queue lengths after construction of the alternative intersection. The analysis revealed that alternative intersections result in shorter vehicle queues and improved traffic flow compared to conventional designs.

In this paper, stability and Hopf bifurcation analysis of a novel hyperchaotic system are investigated. Four feedback control strategies, the linear feedback control method, enhancing feedback control method, speed feedback control method and delayed feedback control method, are used to control the hyperchaotic attractor to unstable equilibrium. Moreover numerical simulations are given to verify the theoretical results.

To improve the effectiveness of model predictive control (MPC) in dynamic traffic signal control strategies, it has been combined with graph convolutional networks (GCNs) and deep reinforcement learning (DRL) technologies. In this study, a neural-network-based traffic signal control optimization method under the MPC framework is proposed. A dynamic correlation matrix is introduced in the predictive model to adapt to the dynamic changes in correlations between nodes over time. The signal control optimization strategy is solved using DRL, where the agent explores the optimal control strategy based on pre-set constraints in the future road environment. The geometric structure and traffic flow data of a real intersection were selected as the simulation validation environment, and a joint simulation was conducted using Python and SUMO. The experimental results indicate that in low-traffic scenarios, the queue length is reduced by more than 2 vehicles compared to the selected comparison methods; in high-traffic scenarios, the queue length is reduced by an average of 17 vehicles. Under the actual traffic data of the intersection, the average speed is increased by 6.4% compared to the fixed timing method; compared to the inductive signal control method, it increases from 9.76 m/s to 11.69 m/s, an improvement of 19.7%, effectively enhancing the intersection signal control performance.

A queue length function mapping model is built through the neural network, and can be optimized based on the white noise principle. Thus, a nonlinear mapping, which the input and output nodes are considered as independent variable and the dependent variable respectively, is obtained. After training the learning samples, which can be got from counting the data of the accident duration, traffic upstream, the passed coefficient, and the vehicles queue length, we can acquired the weights and thresholds, which are used to represent the mapping between the nodes. Also, through the optimization based on the white noise principle, the impact caused by the error can be reduced and a more accurate model is obtained as a predicting measuring tool of the queue length.

Queue Length Estimation along urban arterials is vital to city traffic planners for calculating ‘Level of Service’ measures and optimizing signal plans to alleviate congestion. While several data sources such as GPS, Video, Bluetooth, DSRC etc. are now available, we focus exclusively on high-resolution loop detector data, which is widely available to traffic authorities in North America and elsewhere. Analytical methods for queue length estimation rely on counting input-output vehicle flows over advance and stop-bar detectors or identifying breakpoints in the detector actuation waveforms. However, these methods are limited, as they are sensitive to assumed traffic parameters, detector placement and do not take into account driving behaviors and the effect of left-turn buffers. More recently, Machine Learning models have been developed which learn queue lengths from traffic state data. These are usually trained on localized datasets at coarse time resolutions. In this work, we show that Deep Neural Networks are capable of directly learning an abstract representation of the queuing process, from detector actuation waveforms along an intersection approach with an exclusive left-turn buffer. Using a microscopic traffic simulator, we generate a large dataset by approximately replicating traffic arrival patterns from a realworld loop detector dataset. We then feed multi-cycle data to compute maximum and residual queue lengths across a range of traffic conditions, at the cycle-level. We explore the use of lightweight Encoder-Decoder architectures with Temporal Attention, trained using teacher-forcing strategy, and show their superior performance over regular Feed-Forward Neural Networks.

With the development of satellite positioning technology and network communication technology, the floating vehicle data has the advantages of large amount of data, high precision and wide Coverage. In order to improve the green wave control effect, combines the floating vehicle trajectory data with distributed wave theory, makes the real-time prediction of queue length of current cycle and the time of complete disappearance, and analyzes the green-start coordination, green-middle coordination and green-end coordination in the advantages and disadvantages, designs an optimal control strategy with dynamic green point floating vehicle queue length based on real-time sensing green, improves the efficiency of green band, through simulation experiments to prove the effectiveness of the proposed method.

Dynamic programming for optimal packet routing control using two neural networks

Signalized intersections are nodes in the transportation network where vehicles in different directions meet and are critical points for congestion. Vehicle queue length is one of the performance parameters of a signalized intersection. Long queues of vehicles are at high risk of accidents involving many vehicles. Feedback signal control (actuated signal control) can be used to improve intersection performance. One of the variables that can be used as a feedback input is the length of the vehicle queue. Traffic in Indonesia is mixed traffic where various types of vehicles use the same road lanes and with low lane discipline. This causes the traffic system to become complex and to be stochastic and non-linear. Queue length modeling using a static linear algorithm is unable to capture the phenomenon of this complex traffic system. Therefore, this study aims to build a queue length model based on machine learning, that is, using an artificial neural network (ANN). This model studies traffic systems with historical data so that through the training process it can model queue lengths with a good degree of accuracy. An estimation model was built and applied to a section of the Muara Rapak signalized intersections, Balikpapan. Data on queue length for 10 days, 2 hours/day, obtained using CCTV and direct field surveys. The results of the model test show that ANN has a good level of accuracy with MAE, RMSE and MAPE of 3.8 m, 4.9 m and 6%, respectively.

A survey of deep learning-driven architecture for predictive maintenance