{"title":"A Bioinspired Approach for Mental Emotional State Perception towards Social Awareness in Robotics","authors":"Jordan J. Bird, D. Faria, Luis J. Manso, A. Ekárt","doi":"10.31256/UKRAS19.3","DOIUrl":"https://doi.org/10.31256/UKRAS19.3","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132566115","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}
The 3D point cloud data was captured in outdoor fields under different weather conditions in 4 locations: 2 in the UK, 1 in Spain, and 1 more in USA using the Kinect 2 sensor. Histograms of the reference models used in our algorithm. A FPFH descriptor [1], based on a set of angular features, is computed for each data point. The descriptor is then matched to both reference models and the difference provides the final classification score. Reference models
{"title":"Model based 3D point cloud segmentation for automated selective broccoli harvesting*","authors":"Hector A. Montes, Grzegorz Cielniak, T. Duckett","doi":"10.31256/UKRAS19.7","DOIUrl":"https://doi.org/10.31256/UKRAS19.7","url":null,"abstract":"The 3D point cloud data was captured in outdoor fields under different weather conditions in 4 locations: 2 in the UK, 1 in Spain, and 1 more in USA using the Kinect 2 sensor. Histograms of the reference models used in our algorithm. A FPFH descriptor [1], based on a set of angular features, is computed for each data point. The descriptor is then matched to both reference models and the difference provides the final classification score. Reference models","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":"116692611","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}
Transfer Learning (TL) aims to learn a problem from a source reference to improve on the performance achieved in a target reference. Recently, this concept has been applied in different domains, especially, when the data in the target is insufficient. TL can be applied across domains or across tasks. However, the challenges related to what to transfer, how to transfer and when to transfer create limitations in the realisation of this concept in day to day applications. To address the challenges, this paper presents an overview of the concept of TL and how it can be applied in human-robot interaction for assistive robots requiring to learn human tasks in Ambient Assisted Living environments. The differences in feature spaces between a human (source domain) and robot (target domain) makes it difficult for tasks to be directly learned by robots. To address the challenges of this task, we propose a model for learning across feature spaces by mapping the features in the source domain to the target domain features.
{"title":"Transfer Learning in Assistive Robotics: From Human to Robot Domain","authors":"D. Adama, Ahmad Lotfi, R. Ranson, Pedro Trindade","doi":"10.31256/UKRAS19.16","DOIUrl":"https://doi.org/10.31256/UKRAS19.16","url":null,"abstract":"Transfer Learning (TL) aims to learn a problem from a source reference to improve on the performance achieved in a target reference. Recently, this concept has been applied in different domains, especially, when the data in the target is insufficient. TL can be applied across domains or across tasks. However, the challenges related to what to transfer, how to transfer and when to transfer create limitations in the realisation of this concept in day to day applications. To address the challenges, this paper presents an overview of the concept of TL and how it can be applied in human-robot interaction for assistive robots requiring to learn human tasks in Ambient Assisted Living environments. The differences in feature spaces between a human (source domain) and robot (target domain) makes it difficult for tasks to be directly learned by robots. To address the challenges of this task, we propose a model for learning across feature spaces by mapping the features in the source domain to the target domain features.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"1 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":"124768664","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}
Abstract — Maintaining continuous communication is an important factor that contributes to the success of multi-robot systems. Most research involving multi-robot teams is conducted in controlled laboratory settings, where continuous communication is assumed, typically because there is a wireless network (wifi) that keeps all the robots connected. But for multi-robot tea ms to operate successfully “in the wild”, it is crucial to consider how communication can be maintained when signals fail or robots move out of range. This paper presents a novel “leader - follower behaviour” with dynamic role switching and messaging that supports uninterrupted communication, regardless of network perturbations. A series of experiments were conducted in which it is shown how network perturbations effect performance, comparing a baseline with the new leader-follower behaviour. The experiments record metrics on team success, given the two conditions. These results are significant for real-world multi-robot systems applications that require continuous communication amongst team members.
{"title":"Establishing Continuous Communication through Dynamic Team Behaviour Switching","authors":"Tsvetan Zhivkov, Eric Schneider, E. Sklar","doi":"10.31256/ukras19.22","DOIUrl":"https://doi.org/10.31256/ukras19.22","url":null,"abstract":" Abstract — Maintaining continuous communication is an important factor that contributes to the success of multi-robot systems. Most research involving multi-robot teams is conducted in controlled laboratory settings, where continuous communication is assumed, typically because there is a wireless network (wifi) that keeps all the robots connected. But for multi-robot tea ms to operate successfully “in the wild”, it is crucial to consider how communication can be maintained when signals fail or robots move out of range. This paper presents a novel “leader - follower behaviour” with dynamic role switching and messaging that supports uninterrupted communication, regardless of network perturbations. A series of experiments were conducted in which it is shown how network perturbations effect performance, comparing a baseline with the new leader-follower behaviour. The experiments record metrics on team success, given the two conditions. These results are significant for real-world multi-robot systems applications that require continuous communication amongst team members.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"19 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":"115354024","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}
D. Becker, J. Dobrzanski, J. Hodgson, M. Goh, P. Kinnell, L. Justham
{"title":"Evaluation of U- shaped weld prep identification and tracking","authors":"D. Becker, J. Dobrzanski, J. Hodgson, M. Goh, P. Kinnell, L. Justham","doi":"10.31256/UKRAS19.2","DOIUrl":"https://doi.org/10.31256/UKRAS19.2","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"61 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":"126882800","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}
Tianxiang Zhang, Jinya Su, Cunjia Liu, Wen‐Hua Chen
This paper presents a framework for predicting canopy states in real time by adopting a recent MATLAB based crop model: AquaCrop-OS. The historical observations are firstly used to estimate the crop sensitive parameters in Bayesian approach. Secondly, the model states will be replaced by updating remotely sensed observations in a sequential way. The final predicted states will be in comparison with the groundtruth and the RMSE of these two are 39.4155 g/ 𝒎𝟐 (calibration method) and 19.3679 g/𝒎𝟐(calibration with forcing method) concluding that the system is capable of predicting the crop status timely and improve the performance of calibration strategy.
{"title":"Integration of Calibration and Forcing Methods for Predicting Timely Crop States by Using AquaCrop-OS Model","authors":"Tianxiang Zhang, Jinya Su, Cunjia Liu, Wen‐Hua Chen","doi":"10.31256/ukras19.29","DOIUrl":"https://doi.org/10.31256/ukras19.29","url":null,"abstract":"This paper presents a framework for predicting canopy states in real time by adopting a recent MATLAB based crop model: AquaCrop-OS. The historical observations are firstly used to estimate the crop sensitive parameters in Bayesian approach. Secondly, the model states will be replaced by updating remotely sensed observations in a sequential way. The final predicted states will be in comparison with the groundtruth and the RMSE of these two are 39.4155 g/ 𝒎𝟐 (calibration method) and 19.3679 g/𝒎𝟐(calibration with forcing method) concluding that the system is capable of predicting the crop status timely and improve the performance of calibration strategy.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"32 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":"121756637","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}
Zhan Yubin, W. Zhengying, Zhang Lei, Jiang Weibing
Aim at the precision control problems of water and fertilizer concentration during agricultural fertilization and irrigation periods, designing a control technology that based on PID control technology, the application of conductivity value and pH value to develop precise water and fertilizer irrigation control system. The thesis also performed theoretical analysis and experiment of digital PID control of EC value, the results indicated that the PID control system has the advantages of high control precision, but the control performance is degradation when the fertilizer density changes greatly. The intelligent PID controller with EC value was designed by using open-loop step response, PID control technology and inaccurate-control technology. The test results show that it has PID parameter self-controlling ability, good control performance, the stable time within 3 minutes, precision degree is within ±0.15mS/cm, overshoot less than 15%. The improved intelligent PID controller control the pH value, through using open-loop step control and PID control, the test shows that the stability time within 3 minutes, the accuracy is within ± 0.15pH, overshoot less than 15%. Combine fuzzy PID with grey prediction, the grey prediction Fuzzy PID control of water and fertilizer concentration was developed, which has a fastest corresponding speed and better control stability. It owns a good control effective and have a better control quality Keywords—water-fertilizer irrigation, grey-fuzzy PID control, intelligent control strategy, precise irrigation, concentration detection
{"title":"The Intelligent Control Strategy and Verification for Precise Water-fertilizer Irrigation System","authors":"Zhan Yubin, W. Zhengying, Zhang Lei, Jiang Weibing","doi":"10.31256/ukras19.34","DOIUrl":"https://doi.org/10.31256/ukras19.34","url":null,"abstract":"Aim at the precision control problems of water and fertilizer concentration during agricultural fertilization and irrigation periods, designing a control technology that based on PID control technology, the application of conductivity value and pH value to develop precise water and fertilizer irrigation control system. The thesis also performed theoretical analysis and experiment of digital PID control of EC value, the results indicated that the PID control system has the advantages of high control precision, but the control performance is degradation when the fertilizer density changes greatly. The intelligent PID controller with EC value was designed by using open-loop step response, PID control technology and inaccurate-control technology. The test results show that it has PID parameter self-controlling ability, good control performance, the stable time within 3 minutes, precision degree is within ±0.15mS/cm, overshoot less than 15%. The improved intelligent PID controller control the pH value, through using open-loop step control and PID control, the test shows that the stability time within 3 minutes, the accuracy is within ± 0.15pH, overshoot less than 15%. Combine fuzzy PID with grey prediction, the grey prediction Fuzzy PID control of water and fertilizer concentration was developed, which has a fastest corresponding speed and better control stability. It owns a good control effective and have a better control quality Keywords—water-fertilizer irrigation, grey-fuzzy PID control, intelligent control strategy, precise irrigation, concentration detection","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"167 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":"121089710","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":"Exploiting System Capacity with a Distributed Routing Strategy for UAVs*","authors":"W. Bonnell","doi":"10.31256/ukras19.13","DOIUrl":"https://doi.org/10.31256/ukras19.13","url":null,"abstract":"","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"17 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":"115599394","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}
Locust plagues are very harmful for food security, quality and quantity of agricultural products. With this consideration, precise locust detection is significant for preventing locust plagues. To achieve this task, aggregate channel feature (ACF) object detector with parameters optimization is applied to detect locusts. Experiment results show that ACF object detector with optimized parameters can achieve 0.39 for average precision and 0.86 for log-average miss rate. Moreover, ACF is a non-deep method using a simple model to detect objects. That is, the proposed method is promising to be embedded in a real-time locust detection system.
{"title":"Locust Recognition and Detection via Aggregate Channel Features","authors":"Dewei Yi, Jinya Su, Wen‐Hua Chen","doi":"10.31256/ukras19.30","DOIUrl":"https://doi.org/10.31256/ukras19.30","url":null,"abstract":"Locust plagues are very harmful for food security, quality and quantity of agricultural products. With this consideration, precise locust detection is significant for preventing locust plagues. To achieve this task, aggregate channel feature (ACF) object detector with parameters optimization is applied to detect locusts. Experiment results show that ACF object detector with optimized parameters can achieve 0.39 for average precision and 0.86 for log-average miss rate. Moreover, ACF is a non-deep method using a simple model to detect objects. That is, the proposed method is promising to be embedded in a real-time locust detection system.","PeriodicalId":424229,"journal":{"name":"UK-RAS19 Conference: \"Embedded Intelligence: Enabling and Supporting RAS Technologies\" Proceedings","volume":"33 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":"125198175","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：