Shared Autonomous Mobility on Demand: A Learning-Based Approach and Its Performance in the Presence of Traffic Congestion

Abstract

Mobility-on-demand systems consisting of shared autonomous vehicles (SAVs) are expected to improve the efficiency of urban mobility through reduced vehicle ownership and parking demand. However, several issues in their implementation remain open, such as unifying the vehicle and ride-sharing assignment with rebalancing non-occupied vehicles. Furthermore, proposed SAV systems are evaluated in isolation from other traffic; no congestion is taken into account when assigning requests or calculating routes. To address this gap, we present Shared Autonomous Mobility-on-Demand system (SAMoD), a reinforcement learning-based approach to vehicle relocation and ride-sharing request assignment. Each vehicle learns its pickup and rebalancing behaviour based on local current and observed historical demand. We evaluate SAMoD on Manhattan network using NYC taxi data in microsimulator SUMO. We investigate SAMoD performance in the presence of congestion generated by private vehicles, as well as investigate impact of different percentages of SAMoD vehicles in the system on overall traffic network performance.

Can autonomous vehicles enable sustainable mobility in future cities? Insights and policy challenges from user preferences over different urban transport options

“…A key technological innovation that is expected to radically shape urban passenger and freight transportation is the autonomous vehicle (Paddeu and Parkhurst, 2020;Cugurullo and Acheampong, 2020;Milakis et al, 2017). On the one hand, aside the promise of making motorized transport safe, driverless vehicles are expected to integrate clean technologies and support flexible, free-floating carsharing, thereby helping to reduce car-ownership, congestion, travel-related energy consumptions and C02 emissions (Guériau et al, 2020;Milakis et al, 2017;Chan, 2017). On the other hand, however, initial exploratory studies suggest that driverless vehicles could trigger urban form and travel behaviour changes with potentially negative social and environmental consequences by increasing car-use, vehicle miles of travel, energy use and pollution (Thomopoulos and Givoni, 2015;Wadud et al, 2016;Harper et al, 2016;Stead and Vaddadi, 2019;Legacy et al, 2019).…”

The Sustainability of Artificial Intelligence: An Urbanistic Viewpoint from the Lens of Smart and Sustainable Cities

“…On the one hand, the prospects of AI in the environment domain include: assisting environmental monitoring, optimizing energy consumption and production, optimizing transport systems, and assisting the development of more environmentally efficient transport and logistic systems [157][158][159]. On the other hand, the constraints of AI involve: making biased decisions, increasing urban sprawl, leading to more motor vehicle kilometers traveled, destabilizing property values, establishing heavy energy dependency due to intensive use of technology, and increasing carbon footprints [160][161][162].…”

Intelligent Traffic Management system based on IoT Challenges and Applications

“…Intelligent transportation systems with integrated components such as highway management adaptive traffic signal controls, roadside units, and emergency management services are all examples of smart traffic management [8]. These systems collect real-time traffic data and take the required measures to prevent or reduce any social issues that arise as a result of traffic congestion [9]. For example, people will be able to use real-time traffic maps to help them choose the best route to save effort and time.…”