Melanie Zimmer, Ali Al-Yacoub, P. Ferreira, N. Lohse
In this paper, a technique that reduces the changeover time in industrial workstations is presented. A Learning from Demonstration-based algorithm is used to acquire a new skill through a series of real-world human demonstrations in which the human shows the desired task. Initially, the collected data are filtered and aligned applying Fast Dynamic Time Warping (FastDTW). Then the aligned trajectories are modelled with a Gaussian Mixture Model (GMM), which is used as an input to generate a generalisation of the motion through a Gaussian Mixture Regression (GMR). The proposed approach is set into the context of the openMOS framework to efficiently add new skills that can be performed on different workstations. The main benefit of this work in progress is providing an intuitive, simple technique to add new robotics skills to an industrial platform which accelerates the changeover phase in manufacturing scenarios.
{"title":"Trajectory Creation Towards Fast Skill Deployment in Plug-and- Produce Assembly Systems: A Gaussian-Mixture Model Approach","authors":"Melanie Zimmer, Ali Al-Yacoub, P. Ferreira, N. Lohse","doi":"10.31256/UKRAS19.23","DOIUrl":"https://doi.org/10.31256/UKRAS19.23","url":null,"abstract":"In this paper, a technique that reduces the changeover time in industrial workstations is presented. A Learning from Demonstration-based algorithm is used to acquire a new skill through a series of real-world human demonstrations in which the human shows the desired task. Initially, the collected data are filtered and aligned applying Fast Dynamic Time Warping (FastDTW). Then the aligned trajectories are modelled with a Gaussian Mixture Model (GMM), which is used as an input to generate a generalisation of the motion through a Gaussian Mixture Regression (GMR). The proposed approach is set into the context of the openMOS framework to efficiently add new skills that can be performed on different workstations. The main benefit of this work in progress is providing an intuitive, simple technique to add new robotics skills to an industrial platform which accelerates the changeover phase in manufacturing scenarios.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125757737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This work explored the requirements of accurately and reliably predicting user intention using a deep learning methodology when performing fine-grained movements of the human hand. The focus was on combining a feature engineering process with the effective capability of deep learning to further identify salient characteristics from a biological input signal. 3 time domain features (root mean square, waveform length, and slope sign changes) were extracted from the surface electromyography (sEMG) signal of 17 hand and wrist movements performed by 40 subjects. The feature data was mapped to 6 sensor bend resistance readings from a CyberGlove II system, representing the associated hand kinematic data. These sensors were located at specific joints of interest on the human hand (the thumb’s metacarpophalangeal joint, the proximal interphalangeal joint of each finger, and the radiocarpal joint of the wrist). All datasets were taken from database 2 of the NinaPro online database repository. A 3-layer long short-term memory model with dropout was developed to predict the 6 glove sensor readings using a corresponding sEMG feature vector as input. Initial results from trials using test data from the 40 subjects produce an average mean squared error of 0.176. This indicates a viable pathway to follow for this prediction method of hand movement data, although further work is needed to optimize the model and to analyze the data with a more detailed set of metrics.
{"title":"A Deep Adaptive Framework for Robust Myoelectric Hand Movement Prediction","authors":"Carl Peter Robinson, Baihua Li, Q. Meng, M. Pain","doi":"10.31256/UKRAS19.1","DOIUrl":"https://doi.org/10.31256/UKRAS19.1","url":null,"abstract":"This work explored the requirements of accurately\u0000and reliably predicting user intention using a deep learning\u0000methodology when performing fine-grained movements of the\u0000human hand. The focus was on combining a feature engineering\u0000process with the effective capability of deep learning to further\u0000identify salient characteristics from a biological input signal. 3\u0000time domain features (root mean square, waveform length, and\u0000slope sign changes) were extracted from the surface\u0000electromyography (sEMG) signal of 17 hand and wrist\u0000movements performed by 40 subjects. The feature data was\u0000mapped to 6 sensor bend resistance readings from a CyberGlove\u0000II system, representing the associated hand kinematic data.\u0000These sensors were located at specific joints of interest on the\u0000human hand (the thumb’s metacarpophalangeal joint, the\u0000proximal interphalangeal joint of each finger, and the\u0000radiocarpal joint of the wrist). All datasets were taken from\u0000database 2 of the NinaPro online database repository. A 3-layer\u0000long short-term memory model with dropout was developed to\u0000predict the 6 glove sensor readings using a corresponding sEMG\u0000feature vector as input. Initial results from trials using test data\u0000from the 40 subjects produce an average mean squared error of\u00000.176. This indicates a viable pathway to follow for this\u0000prediction method of hand movement data, although further\u0000work is needed to optimize the model and to analyze the data with\u0000a more detailed set of metrics.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114350045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of a Simulated Production Environment for Plug-And-Produce Architecture Testing","authors":"William Eaton","doi":"10.31256/UKRAS19.27","DOIUrl":"https://doi.org/10.31256/UKRAS19.27","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123699927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Visual Features as Frames of Reference in Task-Parametrised Learning from Demonstration","authors":"Shirine El Zaatari, Weidong Li","doi":"10.31256/ukras19.25","DOIUrl":"https://doi.org/10.31256/ukras19.25","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122112322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Controlling a Bipedal Robot with Pattern Generators Trained with Reinforcement Learning*","authors":"Christos Kouppas, Q. Meng, M. King, D. Majoe","doi":"10.31256/ukras19.5","DOIUrl":"https://doi.org/10.31256/ukras19.5","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130161909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of a Multi-robotic System for Exploration of Biomass Power Plants","authors":"Sihai An, F. Arvin, S. Watson, B. Lennox","doi":"10.31256/UKRAS19.15","DOIUrl":"https://doi.org/10.31256/UKRAS19.15","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115476349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Haibin Cai, Lei Jiang, Junyi Wang, Mohamad Saada, Q. Meng
{"title":"An Embedded System for Real-Time 3D Human Detection","authors":"Haibin Cai, Lei Jiang, Junyi Wang, Mohamad Saada, Q. Meng","doi":"10.31256/UKRAS19.31","DOIUrl":"https://doi.org/10.31256/UKRAS19.31","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123527265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：