Search

Single occupancy vehicles are the most attractive transportation alternative for many commuters, leading to increased traffic congestion and air pollution. Advancements in information technologies create opportunities for smart solutions that incentivize ridesharing and mode shift to higher occupancy vehicles (HOVs) to achieve the car lighter vision of cities. In this study, we present HumanLight, a novel decentralized adaptive traffic signal control algorithm designed to optimize people throughput at intersections. Our proposed controller is founded on reinforcement learning with the reward function embedding the transportation-inspired concept of pressure at the person-level. By rewarding HOV commuters with travel time savings for their efforts to merge into a single ride, HumanLight achieves equitable allocation of green times. Apart from adopting FRAP, a state-of-the-art (SOTA) base model, HumanLight introduces the concept of active vehicles, loosely defined as vehicles in proximity to the intersection within the action interval window. The proposed algorithm showcases significant headroom and scalability in different network configurations considering multimodal vehicle splits at various scenarios of HOV adoption. Improvements in person delays and queues range from 15% to over 55% compared to vehicle-level SOTA controllers. We quantify the impact of incorporating active vehicles in the formulation of our RL model for different network structures. HumanLight also enables regulation of the aggressiveness of the HOV prioritization. The impact of parameter setting on the generated phase profile is investigated as a key component of acyclic signal controllers affecting pedestrian waiting times. HumanLight's scalable, decentralized design can reshape the resolution of traffic management to be more human-centric and empower policies that incentivize ridesharing and public transit systems. [Display omitted] • HumanLight is the first scalable human-centric RL-based adaptive signal controller. • Scalability is enabled from the decentralized design and algorithmic formulation. • HumanLight can democratize urban traffic by equitably allocating green times. • Active vehicles are proposed to handle the variance in occupancy of multimodal traffic. • HumanLight offers policymakers control of the aggressiveness in HOV prioritization. [ABSTRACT FROM AUTHOR]

Digital maps of the transportation network are the foundation of future mobility solutions. Autonomous and connected vehicles rely on real-time, at-scale updating of the environment in which they operate. Successful operation in a hybrid environment, where human and machine intelligence coexist, requires explicit knowledge of the traffic regulator infrastructure. Future generation traffic management strategies and path planning systems must be tightly integrated with the regulator infrastructure in order to improve traffic dynamics and reduce congestion in urban environments. In this article, we present Intersense, a regulator identification and categorization system that leverages raw, naturalistic GPS trajectory data to infer the existence and type of regulators present at network intersections. A supervised learning approach, based on the eXtreme Gradient Boosting (XGBoost) algorithm, combines infrastructure and vehicle telemetry data features to analyze movement patterns vehicles follow when approaching intersections. For the most widely used regulator types (traffic light, stop sign and right of way), the achieved accuracy surpasses 96%. Permutation feature importance is used to evaluate the relative importance of the generated features, providing a detailed view of the classification model and insights for future data collection and processing stages. We showcase the Intersense's adaptability to various GPS data sources with uneven penetration and sampling rates and we illustrate the system's transferability across very diverse cities by performing experiments in San Jose and San Francisco. Finally, the relationship between the number of available trajectories and classification accuracy is derived to determine the necessary data collection investment. [Display omitted] • Intersense enables wide-scale regulator identification via GPS data with 96% accuracy. • The system is transferable to cities with different relief and regulator densities. • Intersense is robust to GPS data sources with uneven penetration and sampling rates. • Data size requirements and feature importance are evaluated. [ABSTRACT FROM AUTHOR]

• Planning bike trips with novel data sources. • Using percolation theory to optimize cost and maximize global connectivity. • Network Science for detecting affinity of trips by income level and further inform local interventions. Cities around the world are turning to non-motorized transport alternatives to help solve congestion and pollution issues. This paradigm shift demands on new infrastructure that serves and boosts local cycling rates. This creates the need for novel data sources, tools, and methods that allow us to identify and prioritize locations where to intervene via properly planned cycling infrastructure. Here, we define potential demand as the total trips of the population that could be supported by bicycle paths. To that end, we use information from a phone-based travel demand and the trip distance distribution from bike apps. Next, we use percolation theory to prioritize paths with high potential demand that benefit overall connectivity if a bike path would be added. We use Bogotá as a case study to demonstrate our methods. The result is a data science framework that informs interventions and improvements to an urban cycling infrastructure. [ABSTRACT FROM AUTHOR]