A reinforcement learning model for personalized driving policies identification

دسته: , تاریخ انتشار: 4 اردیبهشت 1400تعداد بازدید: 379
قیمت محصول

رایگان

جزئیات بیشتر

انتشار

۲۰۲۱

پایگاه داده

نشریه الزویر

نوع نگارش مقاله

مقاله پژوهشی

نمایه

scopus – master journals – JCR

ایمپکت فاکتور

۴٫۲۷۶ در سال ۲۰۲۰

شاخص H_index

۲۶ در سال ۲۰۲۱

شاخص SJR

۰٫۹۰۱ در سال ۲۰۲۰

شاخص Quartile (چارک)

Q1 در سال ۲۰۲۰

مدل مفهومی

ندارد

پرسشنامه

ندارد

متغیر

ندارد

رفرنس

دارد

قوانین استفاده

خرید محصول توسط کلیه کارت های شتاب امکان پذیر است و بلافاصله پس از خرید، لینک دانلود محصول در اختیار شما قرار خواهد گرفت و هر گونه فروش در سایت های دیگر قابل پیگیری خواهد بود.

توضیحات مختصر محصول
A reinforcement learning model for personalized driving policies identification


دانلود رایگان مقاله

فهرست مطالب مقاله:

A reinforcement learning model for personalized driving policies identification

Abstract

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 driv- ing    policies  based   on    Reinforcement   Learning  for    enhancing   existing  Intelligent Transportation Systems (ITS) to the benefit of traffic management and road safety. The pro- posed framework is  implemented on  appropriate driving behavior metrics derived from smartphone sensors’ data streams. Aggressiveness, speeding and mobile usage are  consid- ered 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 per- sonalized quantified driving policies to be  exploited for  self-improvement.  Finally, this paper establishes validation measures of the quality and effectiveness of the produced poli- cies  and methodological tools for  comparing and classifying the examined drivers.

بخشی از متن مقاله:
  1. Introduction

Optimizing driving performance by  addressing personalized aspects of driving behavior is a focal  research area, which may have far  reaching implications to traffic safety and operations, environment, as  well  as  significant benefits for  users. The  above renders necessary the processing of large-scale data explicitly describing the driving task. To this end,  smart- phones can  constitute an alternative data collection tool  for driving behavior analysis (Weiser et al., 2016). Apart from wide diffusion, smartphones  demonstrate  several additional advantages, making instrumenting  a  vehicle for  data collection accessible to the general public, academia and the industry, compensating for the current lack  of dedicated equipment, like on-board diagnostics (OBD) devices (Johnson and Trivedi, 2011; Briante et al., 2014). They can form a non-intrusive environ- ment for continuously collecting rich and more granular data on the actual driving task,  due to a variety of plugged in sensors (Zhao,  ۲۰۰۰; Wahlström et al., 2015; Vlahogianni and Barmpounakis, 2017). Moreover, their communication features enable the connection with cloud based services and platforms (Lane et al., 2010), while the information collected, when exchanged between vehicles or with the road infrastructure, can  contribute to a wide range of Intelligent Transportation Systems (ITS) services (Briante et al., 2014). For example, smartphones have recently penetrated to  usage-based  motor insurance (UBI)

schemes. The concept of UBI systems entails the idea of adjusting insurance costs according to  parameters influencing the probability of accident involvement, like  total exposure (pay as you  drive – PAYD) or individual driving behavior (pay how you drive – PHYD) (Litman, 2004; Tselentis et al., 2016). UBI schemes come to revolutionize traditional insurance approaches with the implementation of advanced custom-made feedback mechanisms to persuade users to  understand their driving behavior and raise awareness on  the effects of  safe  driving (Toledo et al.,  ۲۰۰۸; Lane  et al.,  ۲۰۱۰; Birrell  et al.,  ۲۰۱۴; Baecke and Bocca,  ۲۰۱۷; Tselentis et al., 2017).

Although conceptually the use  of smartphones for the above-mentioned applications is utterly justified, the complexities that come with the smartphone-based driving applications are  numerous and methodologically difficult to be tackled. Apart from the difficulties in  delivering meaningful data on  the travel mode, driving patterns, harsh maneuverings and phone interaction while driving (Handel et al.,  ۲۰۱۴; Vlahogianni and Barmpounakis, 2017; Kanarachos et al.,  ۲۰۱۸), robust approaches should be developed to identify driving profiles. These profiles can lead to meaningful, personalized driving poli- cies useful for raising the driver’s awareness, or monitoring and ranking drivers with respect to their performance. Delivering driving profiles is a research field  that is constantly gaining a lot  of attention (Júnior et al., 2017; Yu et al., 2017; Mantouka et al., 2018). However, the manner these profiles may be combined to produce meaningful – for both the user and the system

– policies is still  at a native stage.

The present work aims to deliver a methodology that automatically formulates personalized quantified driving policies, to  be  disseminated to  drivers as  recommendations for  future behavior without posing unrealistic restrictions on  personal mobility. More precisely, the recommendations produced in no way  restrict drivers by redefining their destination, redirect- ing  their pre-chosen route or even affect their travel times. They  only  lead the users to  achieve more prudent driving with less  redundant, risk prone and,  in the cases of speeding and mobile usage, illegal maneuvers. The methodological framework is founded on  Reinforcement Learning (RL). Driving behavior can  be  extremely diverse among different drivers and during different trips of the same driver. This volatility of driving metrics is attributed to various parameters. This research defines the concept of optimum driving  and seeks for the driving patterns, or else  the recurring driving maneuvers, that should be alleviated for  behavioral improvement to  be  achieved. The  degradation of the quality of trips is expressed through their divergence from optimum driving. The  proposed methodology is implemented in  a set  of 189  drivers monitored for  over a year in a naturalistic driving experiment.

The remainder of this paper is structured as follows. The next sections provide a description of data and analyze the prin- ciples of RL and the Q-learning algorithm. Following, the methodology is outlined and the findings of this research are  cited, as well  as the validation measures and tools for classifying the examined drivers. Finally, findings are  discussed and conclu- sions are  drawn, while suggestions for future research are  provided.

دانلود رایگان مقاله   

نمایش بیشتر
دیدگاه های کاربران
دیدگاهتان را با ما درمیان بگذارید
تعداد دیدگاه : 0 امتیاز کلی : 0.0 توصیه خرید : 0 نفر
بر اساس 0 خرید
0
0
0
0
0

هیچ دیدگاهی برای این محصول نوشته نشده است.

لطفا پیش از ارسال نظر، خلاصه قوانین زیر را مطالعه کنید: فارسی بنویسید و از کیبورد فارسی استفاده کنید. بهتر است از فضای خالی (Space) بیش‌از‌حدِ معمول، شکلک یا ایموجی استفاده نکنید و از کشیدن حروف یا کلمات با صفحه‌کلید بپرهیزید. نظرات خود را براساس تجربه و استفاده‌ی عملی و با دقت به نکات فنی ارسال کنید؛ بدون تعصب به محصول خاص، مزایا و معایب را بازگو کنید و بهتر است از ارسال نظرات چندکلمه‌‌ای خودداری کنید.  

اولین کسی باشید که دیدگاهی می نویسد “A reinforcement learning model for personalized driving policies identification”

نشانی ایمیل شما منتشر نخواهد شد. بخش‌های موردنیاز علامت‌گذاری شده‌اند *

قیمت محصول

رایگان