4 . the hardware of electronic control system is developed . it includes display module whose tasks are man - machine interaction , displaying humidity and temperature , setting humidity , time , date and week , communicating with computer ; power device module which provides power of all modules and drives control equipments to accomplish intention of main control module and can provide zero crossing protection circuit , excess voltage and under voltage protection circuit and excess current protection circuit ; water sensor group , the responsibilities of which are testing the states of water and send the states to scm , then the scm control the water level after judging the states 開發(fā)并實(shí)現(xiàn)了電控系統(tǒng)的硬件部分,包括顯示模塊:主要負(fù)責(zé)人機(jī)交互,顯示濕度、溫度,修改濕度、時(shí)間、日期、星期,和主控模塊通訊;電源驅(qū)動(dòng)模塊:主要提供各個(gè)模塊的電源,驅(qū)動(dòng)被控設(shè)備,實(shí)現(xiàn)主控模塊的意圖,并能實(shí)現(xiàn)繼電器的過零保護(hù),過流、過壓、欠壓保護(hù);水位傳感器組:負(fù)責(zé)測(cè)知水位的狀態(tài),并把此狀態(tài)送入單片機(jī),單片機(jī)經(jīng)過判斷、處理反過來控制水位,實(shí)現(xiàn)水位的閉環(huán)控制。
Due to disadvantage of fusion method generally , in this paper , the advantage of neural network based is used , these advantage include the ability of real - time processing data , the ability of knowledge overflowing and tolerance mistake of structure . taking the image of obstacle from ccd vidicon , these image must pretreat . and distill valid feature , at the same time , provide third dimension distance information from ultrasonic sensor group 在本文中,利用基于神經(jīng)網(wǎng)絡(luò)的信息融合方法的實(shí)時(shí)處理大量數(shù)據(jù)的能力、知識(shí)泛化能力以及結(jié)構(gòu)的容錯(cuò)性等優(yōu)點(diǎn),把ccd攝像機(jī)攝取的障礙物圖象,進(jìn)行預(yù)處理并且提取有效特征,同時(shí)有超聲波傳感器組來提供其第三維的距離信息。
In the end , according to the idea of inertial sensor group that comprises several same kind micromachined sensors , the fusion estimated algorithm of multisensor data is presented in this paper . it is theoretically proved that this algorithm has least mean square errors . as a result , more precise and reliable data can be obtained from the integrated output 最后,基于將硅微慣性敏感元件構(gòu)成慣性敏感器群的思想,本文應(yīng)用了簡(jiǎn)單而有效的多傳感器數(shù)據(jù)融合算法,提高了綜合輸出的精度,得出了比單一慣性傳感器更可靠、更準(zhǔn)確的測(cè)量結(jié)果;并在理論上證明算法的最小方差性;計(jì)算機(jī)上的仿真結(jié)果也從實(shí)際應(yīng)用上進(jìn)一步說明了融合算法的有效性。
First , the author builds the improved architecture based on behaviors for the mobile robot moro - i , proposes a full strategy of obstacle avoidance for region filling , and defines random obstacle and general obstacle which will be dealt with differently , - moreover , the multi - sensor system is set up which includes five groups of ultrasonic sensors for detecting the environmental information and the artificial neural network is used to fuse the information from each sensor group to achieve a precision of obstacle identification ; finally , the hardware and software are designed for the realization of obstacle avoidance 本文作者建立了移動(dòng)機(jī)器人改進(jìn)型基于行為的體系結(jié)構(gòu),提出了較完善的全區(qū)域覆蓋移動(dòng)機(jī)器人的避障理論和策略,定義了突發(fā)障礙和一般障礙,并給予不同的處理方式;同時(shí)重點(diǎn)構(gòu)建了多傳感器系統(tǒng),采用五組超聲波傳感器用于檢測(cè)環(huán)境信息;基于人工神經(jīng)網(wǎng)絡(luò)對(duì)每組獲取的信息進(jìn)行分析和融合,從而較準(zhǔn)確地完成了對(duì)障礙的識(shí)別;最后,通過軟硬件的設(shè)計(jì),實(shí)現(xiàn)了特定的避障行為。