Abstract
This paper proposes a method for adaptively assigning service areas to self-driving taxi agents in ride-share services by using a centralized deep Q-network (DQN) and demand prediction data. A number of (taxi) companies have participated in ride-share services with the increase of passengers due to the mutual benefits for taxi companies and customers. However, an excessive number of participants has often resulted in many empty taxis in a city, leading to traffic jams and energy waste problems. Therefore, an effective strategy to appropriately decide the service areas where agents, which are self-driving programs, have to wait for passengers is crucial for easing such problems and achieving the quality service. Thus, we propose a service area adaptation method for ride share (SAAMS) to allocate service areas to agents for this purpose. We experimentally show that the SAAMS manager can effectively control the agents by allocating their service areas to cover passengers using demand prediction data with some errors. We also evaluated the SAAMS by comparing its performance with those of the conventional methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Yaraghi, N., Ravi, S.: The current and future state of the sharing economy. SSRN Electron. J. (2017)
Erhardt, G.D., et al.: Do transportation network companies decrease or increase congestion? Sci. Adv. 5(5), Kindly provide volume and page number for Ref. [9], if applicable. (2019)
Ke, J., Zheng, H., Yang, H., Chen, X.(Michael): Short-term forecasting of passenger demand under on-demand ride services: a spatio-temporal deep learning approach. Transp. Res. Part C: Emerg. Technol. 85, 591–608 (2017)
Miao, F., et al.: Taxi dispatch with real-time sensing data in metropolitan areas: a receding horizon control approach. CoRR, abs/1603.04418 (2016)
Cordeau, J.-F., Laporte, G.: A tabu search heuristic for the static multi-vehicle dial-a-ride problem. Transp. Res. Part B: Methodol. 37(6), 579–594 (2003)
Berbeglia, G., Cordeau, J.-F., Laporte, G.: A hybrid tabu search and constraint programming algorithm for the dynamic dial-a-ride problem. INFORMS J. Comput. 24, 343–355 (2012)
Nakashima, H., et al.: Design of the smart access vehicle system with large scale ma simulation. In: Proceedings of the 1st International Workshop on Multiagent-Based Societal Systems (2013)
Alonso-Mora, J., Samaranayake, S., Wallar, A., Frazzoli, E., Rus, D.: On-demand high-capacity ride-sharing via dynamic trip-vehicle assignment. Proc. Natl. Acad. Sci. 114(3), 462–467 (2017)
Oda, T., Joe-Wong, C.: MOVI: a model-free approach to dynamic fleet management. In: IEEE INFOCOM 2018 - IEEE Conference on Computer Communications, pp. 2708–2716, April 2018
Lin, K., Zhao, R., Xu, Z., Zhou, J.: Efficient large-scale fleet management via multi-agent deep reinforcement learning. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, USA, pp. 1774–1783. ACM (2018)
van Hasselt, H., Guez, A., Silver, D.: Deep reinforcement learning with double Q-learning. In: Proceedings of the 30th Conference on Artificial Intelligence. AAAI 2016, pp. 2094–2100. AAAI Press (2016)
Tokyo Hire-Taxi Association. Result of taxi research in 2018 (2018). http://taxi-tokyo.or.jp/enquete/pdf/research2018.pdf
Acknowledgements
This work was supported by JSPS KAKENHI (17KT0044) and JST-Mirai Program Grant Number JPMJMI19B5, Japan.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Yoshida, N., Noda, I., Sugawara, T. (2020). Multi-agent Service Area Adaptation for Ride-Sharing Using Deep Reinforcement Learning. In: Demazeau, Y., Holvoet, T., Corchado, J., Costantini, S. (eds) Advances in Practical Applications of Agents, Multi-Agent Systems, and Trustworthiness. The PAAMS Collection. PAAMS 2020. Lecture Notes in Computer Science(), vol 12092. Springer, Cham. https://doi.org/10.1007/978-3-030-49778-1_29
Download citation
DOI: https://doi.org/10.1007/978-3-030-49778-1_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-49777-4
Online ISBN: 978-3-030-49778-1
eBook Packages: Computer ScienceComputer Science (R0)