adaptive robot learning in a non-stationary environmentnew

adaptive robot learning in a non-stationary environmentnew

ID:34486912

大小:717.77 KB

页数:6页

时间:2019-03-06

adaptive robot learning in a non-stationary environmentnew_第1页
adaptive robot learning in a non-stationary environmentnew_第2页
adaptive robot learning in a non-stationary environmentnew_第3页
adaptive robot learning in a non-stationary environmentnew_第4页
adaptive robot learning in a non-stationary environmentnew_第5页
资源描述:

《adaptive robot learning in a non-stationary environmentnew》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、Adaptiverobotlearninginanon-stationaryenvironmentKaryFrämlingHelsinkiUniversityofTechnology,DepartmentofComputerScience,FI-02015HUT,FinlandKary.Framling@hut.fiAbstract.Adaptivecontrolischallenginginreal-worldapplicationssuchasrobotics.Learninghastoberapidenoughtobeperformedinrealtimeandtoavo

2、iddamagetotherobot.Modelsusinglinearfunctionapproximationareinterestinginsuchtasksbecausetheyofferrapidlearningandhavesmallmemoryandprocessingrequirements.Thismakesthemsuitableasadaptivecontrollersinnon-stationaryenvironments,especiallywhenthecontrollerneedstobeanembeddedsystem.Experimentswi

3、thalight-seekingrobotillustratehowtherobotadaptstotheenvironmentbyReinforcementLearningwheretherobotcollectstrainingsamplesbyexploringtheenvironment.1IntroductionTheuseofmachinelearninginreal-worldcontrolapplicationsischallenging.Real-worldtasks,suchasthoseusingrealrobots,involvenoisecomingf

4、romsensors,non-deterministicactionsanduncontrollablechangesintheenvironment.Inrobotics,learningmustberelativelyrapidandpossibletoperformwithoutcausingdamagetotherobot.Onlyinformationthatisavailablefromrobotsensorscanbeusedforlearning.Thismeansthatthelearningmethodshavetobeabletohandlepartial

5、lymissinginformationandsensornoise,whichmaybedifficulttotakeintoaccountinsimulatedenvironments.Artificialneuralnetworks(ANN)areawell-knowntechniqueformachinelearninginnoisyenvironments.Inrealroboticsapplications,however,ANNlearningmaybecometooslowtobepractical.One-layerlinearfunctionapproxim

6、ationANNs(oftencalledAdalines[7])offerfastertrainingthannon-linearANNsandtheirconvergencetoanoptimalsolutioncanusuallybeguaranteed.Thesearepropertiesthatareparticularlyusefulinnon-stationaryenvironmentsthatrequirerapidadaptation,especiallyiftherobothastoexploretheenvironmentandcollecttrainin

7、gsamplesbyitself.Learningbyautonomousexplorationoftheenvironmentisoftenperformedusingreinforcementlearning(RL)methods.Finally,thelimitedmemory-andcomputingpowerneedsofAdalinesmakethemeasytouseinembeddedsystems.Thestructureofthispaperisasfollows.Sec

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。