Wireless Network & Future Internet Research Unit, Department of Electrical Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, THAILAND
+1
Dr. Ei Ei Mon
Wireless Network & Future Internet Research Unit, Department of Electrical Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, THAILAND
This paper reports the release of dataset that describes the critical city inbound-outbound intersection in the central business district’s Sathorn area of Bangkok, Thailand. The available traffic volume and occupancy are captured by sensors located on the Sathorn area’s links. Induction loop coil sensors and thermal and CCTV cameras have been installed at the approaching links of the critical Sathorn-Surasak intersection. Traffic volume data have been collected from the CCTV cameras every 5 s from 2016 to 2019, and occupancy and volume data have been collected from the loop coil sensors every 5 s from May to September 2016 during a social experiment that was part of Sathorn Model project. Occupancy and volume data have also been collected from the thermal cameras every 5 s from May to June 2016. The dataset provides temporal and spatial coverage of Sathorn Road’s primary urban areas, including weekdays, weekends, and public holidays. This dataset can be resources for research on traffic state estimation, traffic light control optimization, and the analysis of critical intersections.
Adaptive traffic signal control techniques have been developed in numerous studies to increase traffic flow efficiency. Using traffic signals to design an adaptive traffic management system is ideal for reducing traffic congestion. Reinforcement learning is a branch of current approaches that try to learn a policy function through a trial-and-error process and maximize the reward through properly adjusted interaction with the learning agent’s environment. We propose a traffic signal control architecture for an oversaturated urban network using Deep Q-Network. We have enhanced the learning process by incorporating diverse state information through upstream and downstream detailed traffic states. We conduct experiments on the Simulation of Urban MObility, an open-source traffic simulator that supports large-scale traffic signal control.
Scientists in each experiment team share their data and use distributed resources for conducting their experiments. These experiments are being accompanied in collaboration with teams that are globally dispersed. Scientific data need to be replicated or cached at distributed locations around the world. Data locality problem and transferred data overhead are important challenges for scheduling such data-intensive scientific workflow application in cloud computing. These applications are leading to the era of big data and task execution involves consuming and producing huge amount of input/output data with data dependencies among tasks. Scheduling and execution overhead are high when low performance of fine-grained tasks is a common problem in widely distributed platforms. Clustering Method based Task Dependency (CMTD) to reduce execution overhead and to improve the computational granularity of scientific workflow tasks is presented in this paper. And this paper proposes the data-intensive workflow scheduling system to minimize makespan of the data-intensive workflow applications, which can be modeled as a directed acyclic graph. Clustering method is validated by using simulation based analysis though WorkflowSim.
Traffic congestion on not only highways but also complex urban road networks has attracted the attention of many researchers. Traffic congestion growing in urban road net- works is an inevitably important problem especially for populated cities during rush hours. A traffic blockage can be realized as the source of traffic congestion, which can propagate to form queues and sometimes a gridlock. Traffic blockages are triggered by complicated factors ranging from temporal and spatial situations. Recurrent congestion is a traffic congestion that occurs during morning and evening rush hours e.g. from school buses and parent vehicles to drive their children to-and-from schools. In addition, unforeseen, unexpected events that can cause as non-recurrent traffic congestion e.g. car breakdowns, accidents, road maintenance, and severe weather conditions, which can disorder normal traffic flows and reduce road capacity. Traffic blockage may spread its negative impacts to neighbouring upstream and downstream links. And that can lead to the formation of congestion gridlock, which further reduce traffic flow efficiency in a complex urban road network. These problems are vital but often tough to resolve in urban road networks. In this paper, the Chula-Sathorn SUMO Simulator (Chula-SSS) dataset has been used with Simulation of Urban Mobility (SUMO) to simulate recurrent and non-recurrent congestion cases. The detection is based on the information from simulated lane area detectors. For non-recurrent case, lanes are closed to simulate the gridlock occurrences. With the morning case of calibrated Chula-SSS dataset, both recurrent and nonrecurrent congestion based gridlock have been studied with upstream and downstream nearby detectors and preliminary results are herein reported upon the gridlock status as detected by using different combinations of traffic jam length and mean speed conditions at both the upstream and downstream detectors of every intersection within the critical looped road segments.
The traffic bottlenecks in urban road networks are more challenging to investigate and discover than in freeways or simple arterial networks. A bottleneck indicates the congestion evolution and queue formation, which consequently disturb travel delay and degrade the urban traffic environment and safety. For urban road networks, sensors are needed to cover a wide range of areas, especially for bottleneck and gridlock analysis, requiring high installation and maintenance costs. The emerging widespread availability of GPS vehicles significantly helps to overcome the geographic coverage and spacing limitations of traditional fixed-location detector data. Therefore, this study investigated GPS vehicles that have passed through the links in the simulated gridlock-looped intersection area. The sample size estimation is fundamental to any traffic engineering analysis. Therefore, this study tried a different number of sample sizes to analyze the severe congestion state of gridlock. Traffic condition prediction is one of the primary components of intelligent transportation systems. In this study, the Long Short-Term Memory (LSTM) neural network was applied to predict gridlock based on bottleneck states of intersections in the simulated urban road network. This study chose to work on the Chula-Sathorn SUMO Simulator (Chula-SSS) dataset. It was calibrated with the past actual traffic data collection by using the Simulation of Urban MObility (SUMO) software. The experiments show that LSTM provides satisfactory results for gridlock prediction with temporal dependencies. The reported prediction error is based on long-range time dependencies on the respective sample sizes using the calibrated Chula-SSS dataset. On the other hand, the low sampling rate of GPS trajectories gives high RMSE and MAE error, but with reduced computation time. Analyzing the percentage of simulated GPS data with different random seed numbers suggests the possibility of gridlock identification and reports satisfying prediction errors.