A reinforcement learning model for personalized driving policies identification
| Autor: | Eleni I. Vlahogianni, Dimitris M. Vlachogiannis, John Golias |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Traffic data
Computer science media_common.quotation_subject Personal mobility Transportation Intelligent transportation systems 010501 environmental sciences Management Monitoring Policy and Law 01 natural sciences 0502 economics and business Reinforcement learning Machine learning Quality (business) Intelligent transportation system 0105 earth and related environmental sciences Civil and Structural Engineering media_common 050210 logistics & transportation Data stream mining 05 social sciences lcsh:TA1001-1280 Identification (information) Work (electrical) Risk analysis (engineering) Automotive Engineering Q-learning lcsh:Transportation engineering |
| Zdroj: | International Journal of Transportation Science and Technology, Vol 9, Iss 4, Pp 299-308 (2020) |
| ISSN: | 2046-0430 |
| Popis: | Optimizing driving performance by addressing personalized aspects of driving behavior and without posing unrealistic restrictions on personal mobility may have far reaching implications to traffic safety, flow operations and the environment, as well as significant benefits for users. The present work addresses the problem of delivering personalized driving policies based on Reinforcement Learning for enhancing existing Intelligent Transportation Systems (ITS) to the benefit of traffic management and road safety. The proposed framework is implemented on appropriate driving behavior metrics derived from smartphone sensors’ data streams. Aggressiveness, speeding and mobile usage are considered to describe the driving profile per trip and are presented as inputs to the Q-learning algorithm. The implementation of the proposed methodological approach produces personalized quantified driving policies to be exploited for self-improvement. Finally, this paper establishes validation measures of the quality and effectiveness of the produced policies and methodological tools for comparing and classifying the examined drivers. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|