Kevin Spieser

Toward a Systematic Approach to the Design and Evaluation of Automated Mobility-on-Demand Systems: A Case Study in Singapore

The objective of this work is to provide analytical guidelines and financial justification for the design of shared-vehicle mobility-on-demand systems. Specifically, we consider the fundamental issue of determining the appropriate number of vehicles to field in the fleet, and estimate the financial benefits of several models of car sharing. As a case study, we consider replacing all modes of personal transportation in a city such as Singapore with a fleet of shared automated vehicles, able to drive themselves, e.g., to move to a customer’s location. Using actual transportation data, our analysis suggests a shared-vehicle mobility solution can meet the personal mobility needs of the entire population with a fleet whose size is approximately 1/3 of the total number of passenger vehicles currently in operation.

Models, algorithms, and evaluation for autonomous mobility-on-demand systems

This tutorial paper examines the operational and economic aspects of autonomous mobility-on-demand (AMoD) systems, a rapidly emerging mode of personal transportation wherein robotic, self-driving vehicles transport customers in a given environment. We address AMoD systems along three dimensions: (1) modeling - analytical models capable of capturing the salient dynamic and stochastic features of customer demand, (2) control - coordination algorithms for the vehicles aimed at stability and subsequently throughput maximization, and (3) economic - fleet sizing and financial analyses for case studies of New York City and Singapore. Collectively, the models and algorithms presented in this paper enable a rigorous assessment of the value of AMoD systems. In particular, the case study of New York City shows that the current taxi demand in Manhattan can be met with about 8,000 robotic vehicles (roughly 70% of the size of the current taxi fleet), while the case study of Singapore suggests that an AMoD system can meet the personal mobility need of the entire population of Singapore with a number of robotic vehicles that is less than 40% of the current number of passenger vehicles. Directions for future research on AMoD systems are presented and discussed.

The Cow-Path Game: A competitive vehicle routing problem

This work considers multi-vehicle systems in which self-interested, mobile agents compete to capture a target that has been distributed on a ring. In the scenarios studied, agents face the added difficulty of having minimal sensing capabilities and limited knowledge of where the target is located. We consider strategic algorithms that allow agents to effectively make decisions and plan trajectories in these settings. Specifically, we characterize equilibria strategies for a search game in which two cows compete to find a patch of clover located somewhere on the unit ring. Throughout, we motivate the work using the example of taxi drivers that compete with one another to garner fares in a busy urban landscape.

Vehicle routing for shared-mobility systems with time-varying demand

This work considers mobility systems in which a shared fleet of self-driving vehicles is used to transport passengers. More specifically, we focus on policies to route both passenger-filled and empty vehicles when the travel demand is time-varying. In this setting, we argue that metrics, such as the cost to relocate empty vehicles, which are well-defined in a stand-alone capacity under steady-state conditions, now make sense only within a framework that reflects inherent tradeoffs with other metrics, e.g., the fleet size and the quality of service provided. As a first step toward developing a general theory of time-varying, shared-mobility systems, we provide an optimization framework that models passengers and vehicles as continuous fluids, and their movement as fluid flows. The model is used to develop some initial performance results related to the minimum number of vehicles required to avoid passenger queueing. Finally, simulation results of a hypothetical shared mobility system based in Singapore demonstrate how a fleet manager could use our optimization approach to select a vehicle routing policy.

On the transfer time complexity of cooperative vehicle routing

Motivated by next-generation air transportation systems, this paper investigates the relationship between traffic volume and congestion in a multi-agent system, assuming that the agents can communicate their intentions with one another. In particular, we consider n independent mobile agents, each assigned an origin and a destination point, and study how the minimum time necessary to safely transfer all agents from their origin to their destination scales with the number of agents n. We provide an algorithm for which the transfer time scales logarithmically in n. This is an improvement over previous results that rely on more conservative conflict models because the agents do not leverage inter-agent cooperation to the same degree, resulting in transfer times that scale as √n.

Cow-Path Games in dynamic environments: Strategic search algorithms for a changing world

This work investigates multi-vehicle systems in which vehicles compete, among themselves, to capture targets. In line with previous formulations, vehicles have minimal sensing capabilities and limited prior knowledge of target locations. That is, vehicles must actively scour the workspace to find targets. However, the games featured herein have the distinguishing feature of unfolding in dynamic environments whereby targets repopulate the workspace over time. For a variety of such games, we characterize persistent search strategies that enable vehicles to plan trajectories that, in the long-run, prove efficient in the face of inter-vehicle competition and an everchanging world.

On the Nash equilibria of a timed asymmetric skirmish

In this work, we present a Nash equilibrium solution for a timed, asymmetric skirmish between two agents: an attacker, and a defender. We derive a solution by focusing on strategy profiles in which both the attacker and defender randomize their actions, which correspond to times, over a common atomic support. We show this class of strategies admits a unique mixed-strategy Nash equilibrium and give an algorithm for its computation. A numerical example highlights interesting features of a typical equilibrium strategy profile.