Joyoung Lee

Mahmoud Nazzal, Abdallah Khreishah, Joyoung Lee et al.

Prediction of taxi service demand and supply is essential for improving customer's experience and provider's profit. Recently, graph neural networks (GNNs) have been shown promising for this application. This approach models city regions as nodes in a transportation graph and their relations as edges. GNNs utilize local node features and the graph structure in the prediction. However, more efficient forecasting can still be achieved by following two main routes; enlarging the scale of the transportation graph, and simultaneously exploiting different types of nodes and edges in the graphs. However, both approaches are challenged by the scalability of GNNs. An immediate remedy to the scalability challenge is to decentralize the GNN operation. However, this creates excessive node-to-node communication. In this paper, we first characterize the excessive communication needs for the decentralized GNN approach. Then, we propose a semi-decentralized approach utilizing multiple cloudlets, moderately sized storage and computation devices, that can be integrated with the cellular base stations. This approach minimizes inter-cloudlet communication thereby alleviating the communication overhead of the decentralized approach while promoting scalability due to cloudlet-level decentralization. Also, we propose a heterogeneous GNN-LSTM algorithm for improved taxi-level demand and supply forecasting for handling dynamic taxi graphs where nodes are taxis. Extensive experiments over real data show the advantage of the semi-decentralized approach as tested over our heterogeneous GNN-LSTM algorithm. Also, the proposed semi-decentralized GNN approach is shown to reduce the overall inference time by about an order of magnitude compared to centralized and decentralized inference schemes.

Ardeshir Mirbakhsh, Joyoung Lee, Dejan Besenski

This paper presents a signal-free intersection control system for CAVs by combination of a pixel reservation algorithm and a Deep Reinforcement Learning (DRL) decision-making logic, followed by a corridor-level impact assessment of the proposed model. The pixel reservation algorithm detects potential colliding maneuvers and the DRL logic optimizes vehicles' movements to avoid collision and minimize the overall delay at the intersection. The proposed control system is called Decentralized Sparse Coordination System (DSCLS) since each vehicle has its own control logic and interacts with other vehicles in coordinated states only. Due to the chain impact of taking random actions in the DRL's training course, the trained model can deal with unprecedented volume conditions, which poses the main challenge in intersection management. The performance of the developed model is compared with conventional and CAV-based control systems, including fixed traffic lights, actuated traffic lights, and the Longest Queue First (LQF) control system under three volume regimes in a corridor of four intersections in VISSIM software. The simulation result revealed that the proposed model reduces delay by 50%, 29%, and 23% in moderate, high, and extreme volume regimes compared to the other CAV-based control system. Improvements in travel time, fuel consumption, emission, and Surrogate Safety Measures (SSM) are also noticeable.

Zijia Zhong, Joyoung Lee

Accessible pedestrian signal (APS) was proposed as a mean to achieve the same level of service that is set forth by the American with Disability Act (ADA) for the visually impaired. One of the major issues of existing APSs is the failure to deliver adequate crossing information for the visually impaired. This paper presents a mobile-based APS application, namely Virtual Guide Dog (VGD). Integrating intersection information and onboard sensors (e.g., GPS, compass, accelerometer, and gyroscope sensor) of modern smartphones, the VGD application can notify the visually impaired: 1) the close proximity of an intersection and 2) the street information for crossing. By employing a screen tapping interface, VGD can remotely place a pedestrian crossing call to the controller, without the need of using a push button. In addition, VGD informs VIs the start of a crossing phase by using text-to-speech technology. The proof-of-concept test shows that VGD keeps the users informed about the remaining distance as their approaching the intersection. It was also found that the GPS-only mode is accompanied by greater distance deviation compared to the mode jointly operating with both GPS and cellular positioning.

Zijia Zhong, Joyoung Lee

Wireless communication plays a vital role in the promising performance of connected and automated vehicle (CAV) technology. This paper proposes a Vissim-based microscopic traffic simulation framework with an analytical dedicated short-range communication (DSRC) module for packet reception. Being derived from ns-2, a packet-level network simulator, the DSRC probability module takes into account the imperfect wireless communication that occurs in real-world deployment. Four managed lane deployment strategies are evaluated using the proposed framework. While the average packet reception rate is above 93\% among all tested scenarios, the results reveal that the reliability of the vehicle-to-vehicle (V2V) communication can be influenced by the deployment strategies. Additionally, the proposed framework exhibits desirable scalability for traffic simulation and it is able to evaluate transportation-network-level deployment strategies in the near future for CAV technologies.