Optimization of Ride Sharing Systems Using Event-driven Receding Horizon Control

TL;DR

This study tackles real-time optimization of ride sharing in a connected transport network by modeling the RSS as a discrete-event system and applying an event-driven Receding Horizon Control (RHC) strategy. A bespoke RSS-specific RHC scheme is developed, featuring a travel-value based vehicle assignment, graph-aware active-target sets, improved future-reward estimation, and oscillation prevention to address trajectory instabilities and computational complexity. The approach demonstrates substantial performance gains over greedy heuristics on realistic city maps and with real taxi data, while maintaining real-time feasibility. The results suggest meaningful practical impact for reducing passenger waiting/travel times and improving fleet efficiency in urban RSS deployments.

Abstract

We develop an event-driven Receding Horizon Control (RHC) scheme for a Ride Sharing System (RSS) in a transportation network where vehicles are shared to pick up and drop off passengers so as to minimize a weighted sum of passenger waiting and traveling times. The RSS is modeled as a discrete event system and the event-driven nature of the controller significantly reduces the complexity of the vehicle assignment problem, thus enabling its real-time implementation. Simulation results using actual city maps and real taxi traffic data illustrate the effectiveness of the RH controller in terms of real-time implementation and performance relative to known greedy heuristics.

Figures (12)

• Figure 1: A typical sample path of passenger $i$'s clock state $z_{i}(t)$.
• Figure 2: Event-Driven receding horizon control.
• Figure 3: Travel value of passenger $i$ evaluated by vehicle $j$ when $s_{i}(t)=0$.
• Figure 4: Example of the reachability set of vehicle $j$.
• Figure 5: A RSS in the Ann Arbor map.