Reinforcement learning (RL)-based traffic signal control has been proven to have great potential in alleviating traffic congestion. The state definition, which is a key element in RL-based traffic signal control, plays a vital role. However, the data used for state definition in the literature are either coarse or difficult to measure directly using the prevailing detection systems for signal control. This paper proposes a deep reinforcement learning-based traffic signal control method which uses high-resolution event-based data, aiming to achieve cost-effective and efficient adaptive traffic signal control. High-resolution event-based data, which records the time when each vehicle-detector actuation/de-actuation event occurs, is informative and can be collected directly from vehicle-actuated detectors (e.g., inductive loops) with current technologies. Given the event-based data, deep learning techniques are employed to automatically extract useful features for traffic signal control. The proposed method is benchmarked with two commonly used traffic signal control strategies, i.e., the fixed-time control strategy and the actuated control strategy, and experimental results reveal that the proposed method significantly outperforms the commonly used control strategies.
Increasing computing capability and high-resolution digital tracing of human behavior make large-scale computational models for individual-based realistic simulation available. Reconstructing a virtual computational environment is crucial for designing and implementing individual interactions in an artificial society as human beings behave in the real world. In this paper, we propose a methodology to recreate a virtual city by utilizing statistical data and geographic information. The synthetic population and physical environment are baseline components of the virtual city. Individual-based modeling is used to specify individuals' demographic characteristics, and each individual is endowed with heterogeneous social attributes. Various physical environments are generated with geographic locations and mapped with individuals to support daily mobility, migration, and interaction. A series of algorithms are proposed to bridge the gap between macroscopic data and microscopic models, and guarantee equivalence between them. Based on the methodology, we reconstructed a virtual city of Beijing, and presented the statistical analysis of population structure, spatial distribution of physical environments, human travel characteristics, and spatial topologies of social networks. Our synthetic population can represent individual actors in the form of households and household members, and the synthetic population is statistically equivalent to a real population. The proposed methodology is efficient to recreate a synthetic virtual city and can serve as a base for computational experiments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.