Path Integral Control for Stochastic Dynamic Traffic Routing Problems

Hibbard, Michael; Wasa, Yasuaki; Tanaka, Takashi

doi:10.1109/cdc42340.2020.9304312

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While path integral control has enjoyed a broad range of research [6]- [8] and applications [9]- [13] in the single-agent case, there is very limited literature on its potential extension to multi-agent systems. Practical applications of PI 2 to multi-agent systems, such as UAV planning, essentially still work in a centralized manner [14].…”

Section: Introductionmentioning

confidence: 99%

Multi-Agent Stochastic Control using Path Integral Policy Improvement

Várnai

Dimarogonas

2022

2022 American Control Conference (ACC)

View full text Add to dashboard Cite

Path integral policy improvement (PI 2 ) is a datadriven method for solving stochastic optimal control problems. Both feedforward and feedback controls are calculated based on a sample of noisy open-loop trajectories of the system and their costs, which can be obtained in a highly parallelizable manner. The control strategy offers theoretical performance guarantees related to the expected cost achieved by the resulting closed-loop system. This paper extends the single-agent case to a multi-agent setting, where such theoretical guarantees have not been attained previously. We provide both a decentralized and a leader-follower scheme for distributing the feedback calculations under different communication constraints. The theoretical results are verified numerically through simulations.

show abstract

Section: Introductionmentioning

confidence: 99%