Search

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., Comment: 13 pages, 10 figures, LaTeX; typos corrected, references added, mathematical analysis added

The connected vehicle (CV) data could potentially revolutionize the traffic monitoring landscape as a new source of CV data that are collected exclusively from original equipment manufactures (OEMs) have emerged in the commercial market in recent years. Compared to existing CV data that are used by agencies, the new-generation of CV data have certain advantages including nearly ubiquitous coverage, high temporal resolution, high spatial accuracy, and enriched vehicle telematics data (e.g., hard braking events). This paper proposed a traffic profiling framework that target vehicle-level performance indexes across mobility, safety, riding comfort, traffic flow stability, and fuel consumption. The proof-of-concept study of a major interstate highway (i.e., I-280 NJ), using the CV data, illustrates the feasibility of going beyond traditional aggregated traffic metrics. Lastly, potential applications for either historical analysis and even near real-time monitoring are discussed. The proposed framework can be easily scaled and is particularly valuable for agencies that wish to systemically monitoring regional or statewide roadways without substantial investment on infrastructure-based sensing (and the associated on-going maintenance costs), Comment: 7 pages, 8 figures

This paper applies a classical physics-based model to control platooning AVs in a commercial traffic simulation software. In Spring-Mass-Damper model, each vehicle is assumed as a mass coupled with its preceding vehicle with a spring and a damper: the spring constant and damper coefficient control spacing and speed adoption between vehicles. Limitations on platooning-oriented communication range and number of vehicles in each platoon are applied to the model to reflect real-world circumstances and avoid overlengthened platoons. The SMD model control both intra-platoon and inter-platoon interactions. Initial evaluation of the model reveals that the SMD model does not cause a negative spacing error between AVs in a harsh deceleration scenario, guaranteeing safety. Besides that, the SMD model produces a smaller positive average spacing error than VISSIM built-in platooning module, which prevents maximum throughput drop. The simulation result for a regular highway section reveals that the proposed platooning algorithm increases the maximum throughput by 10%, 29%, and 63% under 10%, 50%, and full market penetration rate of AVs with 0.5 sec response time. A merging section with different volume combinations on the main section and merging section and different market penetration rates of AVs is also modeled to test inter-platoon spacing policy effectiveness in accommodating merging vehicles. Travel time reductions of 20% and 4% are gained under low MPR of AVs on the mainlane and merging lane accordingly. Meanwhile, a more noticeable travel time reduction is observed in both mainline and merging lanes and under all volume combinations in higher AVs' MPR.

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.

The connected vehicle data (CVD) is one of the most promising emerging mobility data that greatly increases the ability to effectively monitor transportation system performance. A commercial vehicle trajectory dataset was evaluated for market penetration and coverage to establish whether it represents a sufficient sample of the vehicle volumes across the statewide roadway network of New Jersey. The dataset (officially named Wejo Vehicle Movement data) was compared to the vehicle volumes obtained from 46 weight-in-motion (WIM) traffic count stations during the corresponding two-month period. The observed market penetration rates of the Movement data for the interstate highways, non-interstate expressways, major arterials, and minor arterials are 2.55% (std. dev. 0.76%), 2.31% (std. dev. 1.07%), 3.25% (standard deviation 1.48%), and 4.39% (standard deviation 2.65%), respectively. Additionally, the temporal resolution of the dataset (i.e., the time interval between consecutive Wejo vehicle trips captured at a given roadway section, time-of-day variation, day-of-month variation) was also found to be consistent among the evaluated WIM locations. Although relatively low (less than 5%), the consistent market penetration, combined with uniform spatial distribution of equipped vehicles within the traffic flow, could enable or enhance a wide range of traffic analytics applications., Comment: 6 pages, 9 figures

Being one of the most promising applications enabled by connected and automated vehicles (CAV) technology, Cooperative Adaptive Cruise Control (CACC) is expected to be deployed in the near term on public roads.} Thus far, the majority of the CACC studies have been focusing on the overall network performance with limited insights on the potential impacts of CAVs on human-driven vehicles (HVs).This paper aims to quantify such impacts by studying the high-resolution vehicle trajectory data that are obtained from microscopic simulation. Two platoon clustering strategies for CACC- an ad hoc coordination strategy and a local coordination strategy-are implemented. Results show that the local coordination outperforms the ad hoc coordination across all tested market penetration rates (MPRs) in terms of network throughput and productivity. According to the two-sample Kolmogorov-\textcolor{re}{Smirnov} test, however, the distributions of the hard braking events (as a potential safety impact) for HVs change significantly under local coordination strategy. For both of the clustering strategy, CAVs increase the average lane change frequency for HVs. The break-even point for average lane change frequency between the two strategies is observed at 30% MPR, which decreases from 5.42 to 5.38 per vehicle. The average lane change frequency following a monotonically increasing pattern in response to MPR, and it reaches the highest 5.48 per vehicle at 40% MPR. Lastly, the interaction state of the car-following model for HVs is analyzed. It is revealed that the composition of the interaction state could be influenced by CAVs as well. One of the apparent trends is that the time spent on approaching state declines with the increasing presence of CAVs., Comment: 22 pages, 17 figures. arXiv admin note: substantial text overlap with arXiv:1909.13204

As a promising application of connected and automated vehicles (CAVs), Cooperative Adaptive Cruise Control (CACC) is expected to be deployed on the public road in the near term. Thus far the majority of the CACC studies have been focusing on the overall network performance with limited insight on the potential impact of CAVs on human-driven vehicles (HVs). This paper aims to quantify the influence of CAVs on HVs by studying the high-resolution vehicle trajectory data that is obtained from microscopic simulation. Two clustering strategies for CACC are implemented: an ad hoc coordination one and a local coordination one. Results show that the local coordination outperforms the ad hoc coordination across all tested market penetration rates (MPRs) in terms of network throughput and productivity. The greatest performance difference between the two strategies is observed at 30% and 40% MPR for throughput and productivity, respectively. However, the distributions of the hard braking observations (as a potential safety impact) for HVs change significantly under local coordination strategy. Regardless of the clustering strategy, CAVs increase the average lane change frequency for HVs. 30% MPR is the break-even point for local coordination, after which the average lane change frequency decreases from the peak 5.42 to 5.38. Such inverse relationship to MPR is not found in the ah hoc case and the average lane change frequency reaches the highest 5.48 at 40% MPR., Comment: 7 pages, 6 figures, 2019 IEEE 2nd Connected and Automated Vehicles Symposium

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., Comment: 8 pages, 8 figures

Traffic simulation is a cost-effective way to test the deployment of Cooperative Adaptive Cruise Control (CACC) vehicles in a large-scale transportation network. By using a previously developed microscopic simulation testbed, this paper examines the impacts of four managed lane strategies for the near-term deployment of CACC vehicles under mixed traffic conditions. Network-wide performance measures are investigated from the perspectives of mobility, safety, equity, and environmental impacts. In addition, the platoon formation performance of CACC vehicles is evaluated with platoon-orientated measures, such as the percentage of platooned CACC vehicles, average platoon depth, and vehicle-hour-platooned that is proposed in this paper under the imperfect DSRC communication environment. Moreover, managed lane score matrices are developed to incorporate heterogeneous categories of performance measures, aiming to provide a more comprehensive picture for stakeholders. The results show that mixing CACC traffic along with non-CACC traffic across all travel lanes is an acceptable option when the market penetration (MP) is lower than 30% for roadways where a managed lane is absent. Providing CACC with priority access to an existing managed lane, if available, is also a good strategy for improving the overall traffic performance when the MP is lower than 40%. When the MP reaches above 40%, a dedicated lane for CACC vehicles is recommended, as it provides greater opportunity for CACC vehicles to form platoons. The facilitation of homogeneous CACC traffic flow could make further improvements possible in the future., Comment: 22 pages, 15 figures

Managed lanes, such as a dedicated lane for connected and automated vehicles (CAVs), can provide not only technological accommodation but also desired market incentives for road users to adopt CAVs in the near future. In this paper, we investigate traffic flow characteristics with two configurations of the managed lane across different market penetration rates and quantify the benefits from the perspectives of lane-level headway distribution, fuel consumption, communication density, and overall network performance. The results highlight the benefits of implementing managed lane strategies for CAVs: 1) a dedicated CAV lane significantly extends the stable region of the speed-flow diagram and yields a greater road capacity. As the result shows, the highest flow rate is 3,400 vehicles per hour per lane at 90% market penetration rate with one CAV lane; 2) the concentration of CAVs in one lane results in a narrower headway distribution (with smaller standard deviation) even with partial market penetration; 3) a dedicated CAV lane is also able to eliminate duel-bell-shape distribution that is caused by the heterogeneous traffic flow; and 4) a dedicated CAV lane creates a more consistent CAV density, which facilitates communication activity and decreases the probability of packet dropping.

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., Comment: 6 pages, 6 figure, 44th Annual Conference of the IEEE Industrial Electronics Society

This article addresses the problem of controlling the speed of a number of automated vehicles before they enter a speed reduction zone on a freeway. We formulate the control problem and provide an analytical, closed-form solution that can be implemented in real time. The solution yields the optimal acceleration/deceleration of each vehicle under the hard safety constraint of rear-end collision avoidance. The effectiveness of the solution is evaluated through a microscopic simulation testbed and it is shown that the proposed approach reduces significantly both fuel consumption and travel time. In particular, for three different traffic volume levels, fuel consumption for each vehicle is reduced by 19-22% compared to the baseline scenario, where human-driven vehicles are considered, by 12-17% compared to the variable speed limit (VSL) algorithm, and by 18-34% compared to the vehicular-based speed harmonization (SPD-HARM) algorithm. Similarly, travel time is improved by 26-30% compared to the baseline scenario, by 3-19% compared to the VSL algorithm, and by 31-39% compared to the vehicular-based SPD-HARM algorithm., Comment: 13 pages, 5 figures