【摘要】：近年來,機動車保有量的迅猛增長使得停車難的問題日益凸顯。造成停車難的原因是多方面的,主要表現(xiàn)在停車場車位信息閉塞、停車場人工收費效率低下以及停車場車位流轉(zhuǎn)率不高等。為了減輕停車難的壓力,引入采用先進互聯(lián)網(wǎng)技術(shù)以及智能設(shè)備的智能停車系統(tǒng)升級改造傳統(tǒng)停車系統(tǒng)是大勢所趨。傳統(tǒng)停車系統(tǒng)無法提供實時車位信息,不支持在線繳納停車費用,也沒有車位預(yù)測功能。針對這些問題,在帶有車牌識別系統(tǒng)的線下收費軟件的基礎(chǔ)上,本課題基于安卓系統(tǒng)開發(fā)了一款具有車位預(yù)測功能的智能停車系統(tǒng)。最終實現(xiàn)的系統(tǒng)支持車位信息實時發(fā)布、用戶在線繳納停車費用以及車位預(yù)訂等功能。在開發(fā)過程中,首先,結(jié)合前期對用戶以及停車場的調(diào)查結(jié)果,對系統(tǒng)進行需求分析并討論了系統(tǒng)實現(xiàn)所涉及的相關(guān)框架與技術(shù)。系統(tǒng)通信協(xié)議為HTTP協(xié)議,數(shù)據(jù)交互格式選用JSON。然后,根據(jù)具體需求,系統(tǒng)被劃分為我要停車、停車助手以及個人中心等功能模塊。進一步地,針對各個功能模塊設(shè)計對應(yīng)的數(shù)據(jù)表,按模型-視圖-控制器模式完成系統(tǒng)的設(shè)計與實現(xiàn)。同時,為了提高停車場車位的流轉(zhuǎn)率以及車位預(yù)測的準(zhǔn)確性,分別基于灰色模型和神經(jīng)網(wǎng)絡(luò)設(shè)計了空閑車位的預(yù)測方法,并根據(jù)誤差變化動態(tài)調(diào)整學(xué)習(xí)速率以優(yōu)化神經(jīng)網(wǎng)絡(luò)。為了驗證預(yù)測方法的有效性,針對成都某停車場的空閑車位情況做了預(yù)測仿真,并就仿真結(jié)果進行了分析討論。在此基礎(chǔ)上,系統(tǒng)最終選用學(xué)習(xí)速率動態(tài)調(diào)整的神經(jīng)網(wǎng)絡(luò)作為空閑車位預(yù)測方法。最后,對該智能停車系統(tǒng)進行功能測試、非功能測試以及用戶滿意度調(diào)查。結(jié)果表明該系統(tǒng)具有較好的實時性、穩(wěn)健性、兼容性以及較高的用戶滿意度。采用該系統(tǒng)之后,停車場車位信息更加透明化、車位流轉(zhuǎn)率更高、繳費方式更加多樣化、尋找車位所花時間更加少。間接地改善了交通擁堵狀況,并在一定程度上減輕了環(huán)境污染。
[Abstract]:In recent years, with the rapid growth of vehicle ownership, the problem of parking difficulties has become increasingly prominent. There are many reasons for parking difficulties, such as the block of parking space information, the low efficiency of parking lot manual charge and the low circulation rate of parking space. In order to reduce the pressure of parking difficulties, it is the general trend to introduce the intelligent parking system which adopts advanced Internet technology and intelligent equipment to upgrade and transform the traditional parking system. Traditional parking systems can not provide real-time parking information, do not support online parking fees, and do not have parking prediction function. In order to solve these problems, an intelligent parking system with the function of parking space prediction is developed on the basis of offline charging software with license plate recognition system. The system supports real-time distribution of parking space information, online parking fees and parking reservation. In the process of development, first of all, combined with the previous survey results of users and parking lots, the requirements of the system are analyzed and the related framework and technology involved in the implementation of the system are discussed. The communication protocol of the system is HTTP protocol, and the format of data exchange is JSON.. Then, according to the specific needs, the system is divided into my parking, parking assistant and personal center and other functional modules. Furthermore, the corresponding data table is designed for each functional module, and the system is designed and implemented according to the model-view-controller mode. At the same time, in order to improve the flow rate of parking space and the accuracy of parking space prediction, the prediction method of free parking space is designed based on grey model and neural network respectively, and the learning rate is dynamically adjusted according to the error change to optimize the neural network. In order to verify the effectiveness of the prediction method, the free parking space in a parking lot in Chengdu is predicted and simulated, and the simulation results are analyzed and discussed. On this basis, the system finally chooses the learning rate dynamic adjustment neural network as the free parking prediction method. Finally, the function test, non-function test and user satisfaction survey of the intelligent parking system are carried out. The results show that the system has good real-time, robustness, compatibility and high user satisfaction. With this system, the parking space information is more transparent, the parking space flow rate is higher, the payment method is more diversified, and the time to find the parking space is less. Indirectly improve the traffic congestion, and to some extent reduce environmental pollution.
【學(xué)位授予單位】：四川師范大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：U495;U491.7

【參考文獻】

相關(guān)期刊論文 前10條

1 許匯文;;“互聯(lián)網(wǎng)+”破解停車難的實踐探索[J];前線;2016年12期

2 洪偉;任芬;;云服務(wù)器技術(shù)的發(fā)展與安全遠程訪問研究[J];網(wǎng)絡(luò)安全技術(shù)與應(yīng)用;2016年09期

3 王新普;周想凌;邢杰;楊軍;;一種基于改進灰色BP神經(jīng)網(wǎng)絡(luò)組合的光伏出力預(yù)測方法[J];電力系統(tǒng)保護與控制;2016年18期

4 張以帥;賴惠鴿;李勇;唐光耀;張晨藝;;基于MEA優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的天然氣短期負(fù)荷預(yù)測[J];自動化與儀表;2016年05期

5 江媛;;物聯(lián)網(wǎng)技術(shù)發(fā)展與應(yīng)用[J];電子測試;2016年05期

6 周盼;張蕾;;基于Android平臺的變電站智能巡檢系統(tǒng)[J];計算機測量與控制;2015年09期

7 王廣鵬;;“互聯(lián)網(wǎng)+”城市智能交通發(fā)展趨勢探討[J];中國安防;2015年15期

8 王正玉;李斌;;基于DAO模式的Hibernate框架在Java Web開發(fā)中的應(yīng)用[J];微型機與應(yīng)用;2015年11期

9 朱佳偉;喻梁文;關(guān)志;陳鐘;;Android權(quán)限機制安全研究綜述[J];計算機應(yīng)用研究;2015年10期

10 魯曉天;李永全;;Android中的BroadcastReceiver注冊方式研究[J];電腦知識與技術(shù);2015年10期

相關(guān)碩士學(xué)位論文 前6條

1 萬木林;面向Android手機應(yīng)用的功能測試軟件的設(shè)計與實現(xiàn)[D];蘇州大學(xué);2015年

2 王瑾;基于SSH框架下的校園網(wǎng)上交易平臺[D];電子科技大學(xué);2014年

3 張承寰;基于Android平臺的通用傳感器系統(tǒng)的設(shè)計與實現(xiàn)[D];北京郵電大學(xué);2014年

4 韋藍鑫;基于RFID技術(shù)與百度地圖技術(shù)結(jié)合的城市停車誘導(dǎo)信息系統(tǒng)研究[D];長安大學(xué);2013年

5 張星波;基于GMR傳感器的無線車位探測器設(shè)計[D];杭州電子科技大學(xué);2013年

6 荀亞敏;基于無線傳感器網(wǎng)絡(luò)的交通信息采集系統(tǒng)設(shè)計[D];長安大學(xué);2012年

，

本文編號：2206841