Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8594223
Yuanzhouhan Cao, O. Canévet, J. Odobez
We propose a head pose estimation framework that leverages on a recent keypoint detection model. More specifically, we apply the convolutional pose machines (CPMs) to input images, extract different types of facial keypoint features capturing appearance information and keypoint relationships, and train multilayer perceptrons (MLPs) and convolutional neural networks (CNNs) for head pose estimation. The benefit of leveraging on the CPMs (which we apply anyway for other purposes like tracking) is that we can design highly efficient models for practical usage. We evaluate our approach on the Annotated Facial Landmarks in the Wild (AFLW) dataset and achieve competitive results with the state-of-the-art.
{"title":"Leveraging Convolutional Pose Machines for Fast and Accurate Head Pose Estimation","authors":"Yuanzhouhan Cao, O. Canévet, J. Odobez","doi":"10.1109/IROS.2018.8594223","DOIUrl":"https://doi.org/10.1109/IROS.2018.8594223","url":null,"abstract":"We propose a head pose estimation framework that leverages on a recent keypoint detection model. More specifically, we apply the convolutional pose machines (CPMs) to input images, extract different types of facial keypoint features capturing appearance information and keypoint relationships, and train multilayer perceptrons (MLPs) and convolutional neural networks (CNNs) for head pose estimation. The benefit of leveraging on the CPMs (which we apply anyway for other purposes like tracking) is that we can design highly efficient models for practical usage. We evaluate our approach on the Annotated Facial Landmarks in the Wild (AFLW) dataset and achieve competitive results with the state-of-the-art.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"18 1","pages":"1089-1094"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83113269","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}
Pub Date : 2018-10-01DOI: 10.1109/iros.2018.8593782
{"title":"IROS 2018 Technical Program","authors":"","doi":"10.1109/iros.2018.8593782","DOIUrl":"https://doi.org/10.1109/iros.2018.8593782","url":null,"abstract":"","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83133932","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8594126
Jongseo Lee, T. Muskardin, Cristina Ruiz Pacz, P. Oettershagen, Thomas Stastny, Inkyu Sa, R. Siegwart, K. Kondak
System identification of High Altitude Long Endurance fixed-wing aerial vehicles is challenging as its operating flight envelope covers wide ranges of altitudes and Mach numbers. We present a new global system identification framework geared towards such fixed-wing aerial platforms where the aim is to build a global aerodynamic model without many repetitions of local system identification procedures or the use of any aerodynamic database. Instead we apply parameter identification techniques to virtually created system identification data and update the identified parameters with available flight test data. The proposed framework was evaluated using data set outside the flight envelope of the available flight test data, i.e. at different airspeeds considering both interpolation and extrapolation scenarios. The error analysis has shown that the obtained longitudinal aerodynamic model can accurately predict the pitch rate and pitch angle, mostly within a tolerance of $pm pmb{1.5}$ degrees/s and $pm pmb{2}$ degrees respectively. Such a cost and time efficient model development framework enables high fidelity simulation and precise control which ultimately leads to higher success rates in autonomous missions.
{"title":"Towards Autonomous Stratospheric Flight: A Generic Global System Identification Framework for Fixed-Wing Platforms","authors":"Jongseo Lee, T. Muskardin, Cristina Ruiz Pacz, P. Oettershagen, Thomas Stastny, Inkyu Sa, R. Siegwart, K. Kondak","doi":"10.1109/IROS.2018.8594126","DOIUrl":"https://doi.org/10.1109/IROS.2018.8594126","url":null,"abstract":"System identification of High Altitude Long Endurance fixed-wing aerial vehicles is challenging as its operating flight envelope covers wide ranges of altitudes and Mach numbers. We present a new global system identification framework geared towards such fixed-wing aerial platforms where the aim is to build a global aerodynamic model without many repetitions of local system identification procedures or the use of any aerodynamic database. Instead we apply parameter identification techniques to virtually created system identification data and update the identified parameters with available flight test data. The proposed framework was evaluated using data set outside the flight envelope of the available flight test data, i.e. at different airspeeds considering both interpolation and extrapolation scenarios. The error analysis has shown that the obtained longitudinal aerodynamic model can accurately predict the pitch rate and pitch angle, mostly within a tolerance of $pm pmb{1.5}$ degrees/s and $pm pmb{2}$ degrees respectively. Such a cost and time efficient model development framework enables high fidelity simulation and precise control which ultimately leads to higher success rates in autonomous missions.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"339 1","pages":"6233-6240"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80730099","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8593984
F. Morbidi, Estelle Bretagne
In this paper, we present a new characterization of mobility for formations of unicycle robots defined by distance-bearing constraints. In fact, by introducing a simple reduction procedure which associates a prescribed formation with a “macro-robot”, we extend the classification by type proposed by Campion et al., to multi-agent systems. To simplify the classification task, which only leverages the nonslip condition for a conventional centered wheel, we assume that the robots are disposed at the vertices of a regular convex polygon. We demonstrate the practical utility of the notion of macro-robot in a trajectory-tracking control problem for a formation of unicycles.
{"title":"A New Characterization of Mobility for Distance-Bearing Formations of Unicycle Robots","authors":"F. Morbidi, Estelle Bretagne","doi":"10.1109/IROS.2018.8593984","DOIUrl":"https://doi.org/10.1109/IROS.2018.8593984","url":null,"abstract":"In this paper, we present a new characterization of mobility for formations of unicycle robots defined by distance-bearing constraints. In fact, by introducing a simple reduction procedure which associates a prescribed formation with a “macro-robot”, we extend the classification by type proposed by Campion et al., to multi-agent systems. To simplify the classification task, which only leverages the nonslip condition for a conventional centered wheel, we assume that the robots are disposed at the vertices of a regular convex polygon. We demonstrate the practical utility of the notion of macro-robot in a trajectory-tracking control problem for a formation of unicycles.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"4 1","pages":"4833-4839"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83796817","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8594180
Connor Brooks, Madhur Atreya, D. Szafir
Successful human-robot collaboration depends on a shared understanding of task state and current goals. In nonlinear or freeform tasks without an explicit task model, robot partners are unable to provide assistance without the ability to translate perception into meaningful task knowledge. In this paper, we explore the utility of multimodal recurrent neural networks (RNNs) with long short-term memory (LSTM) units for real-time subtask recognition in order to provide context-aware assistance during generic assembly tasks. We train RNNs to recognize specific subtasks in individual modalities, then combine the high-level representations of these networks through a nonlinear connection layer to create a multimodal subtask recognition system. We report results from implementing the system on a robot that uses the subtask recognition system to provide predictive assistance to a human partner during a laboratory experiment involving a human-robot team completing an assembly task. Generalizability of the system is evaluated through training and testing on separate tasks with some similar subtasks. Our results demonstrate the value of such a system in providing assistance to human partners during a freeform assembly scenario and increasing humans' perception of the robot's agency and usefulness.
{"title":"Proactive Robot Assistants for Freeform Collaborative Tasks Through Multimodal Recognition of Generic Subtasks","authors":"Connor Brooks, Madhur Atreya, D. Szafir","doi":"10.1109/IROS.2018.8594180","DOIUrl":"https://doi.org/10.1109/IROS.2018.8594180","url":null,"abstract":"Successful human-robot collaboration depends on a shared understanding of task state and current goals. In nonlinear or freeform tasks without an explicit task model, robot partners are unable to provide assistance without the ability to translate perception into meaningful task knowledge. In this paper, we explore the utility of multimodal recurrent neural networks (RNNs) with long short-term memory (LSTM) units for real-time subtask recognition in order to provide context-aware assistance during generic assembly tasks. We train RNNs to recognize specific subtasks in individual modalities, then combine the high-level representations of these networks through a nonlinear connection layer to create a multimodal subtask recognition system. We report results from implementing the system on a robot that uses the subtask recognition system to provide predictive assistance to a human partner during a laboratory experiment involving a human-robot team completing an assembly task. Generalizability of the system is evaluated through training and testing on separate tasks with some similar subtasks. Our results demonstrate the value of such a system in providing assistance to human partners during a freeform assembly scenario and increasing humans' perception of the robot's agency and usefulness.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"1 1","pages":"8567-8573"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79113286","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8593372
Imeen Ben Salah, Sébastien Kramm, C. Demonceaux, P. Vasseur
Recent progress in 3D sensor devices and in semantic mapping allows to build very rich HD 3D maps very useful for autonomous navigation and localization. However, these maps are particularly huge and require important memory capabilities as well computational resources. In this paper, we propose a new method for summarizing a 3D map (Mesh)as a set of compact spheres in order to facilitate its use by systems with limited resources (smartphones, robots, UAVs,…). This vision-based summarizing process is applied in a fully automatic way using jointly photometric, geometric and semantic information of the studied environment. The main contribution of this research is to provide a very compact map that maximizes the significance of its content while maintaining the full visibility of the environment. Experimental results in summarizing large-scale 3D map demonstrate the feasibility of our approach and evaluate the performance of the algorithm.
{"title":"Summarizing Large Scale 3D Mesh","authors":"Imeen Ben Salah, Sébastien Kramm, C. Demonceaux, P. Vasseur","doi":"10.1109/IROS.2018.8593372","DOIUrl":"https://doi.org/10.1109/IROS.2018.8593372","url":null,"abstract":"Recent progress in 3D sensor devices and in semantic mapping allows to build very rich HD 3D maps very useful for autonomous navigation and localization. However, these maps are particularly huge and require important memory capabilities as well computational resources. In this paper, we propose a new method for summarizing a 3D map (Mesh)as a set of compact spheres in order to facilitate its use by systems with limited resources (smartphones, robots, UAVs,…). This vision-based summarizing process is applied in a fully automatic way using jointly photometric, geometric and semantic information of the studied environment. The main contribution of this research is to provide a very compact map that maximizes the significance of its content while maintaining the full visibility of the environment. Experimental results in summarizing large-scale 3D map demonstrate the feasibility of our approach and evaluate the performance of the algorithm.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"27 1","pages":"1-9"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79455115","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8593890
Jungwook Suh
Many surgical robots are remotely actuated by means of wire cables. In the past, the cables wound around circular pulleys at the robot joints did not constitute a problem of the cable driver structure. However, the pulleys inside the joints are removed recently in order to miniaturize the joints, so a specially designed cable driver suitable for the miniature joint structure is required for stable driving. In this paper, we propose a novel cable driver design for driving a pulleyless rolling joint and extend it to 2-DOF structure. Then, the proposed cable driver is manufactured using 3D printing with the 2-DOF bending joint, and an experiment is performed to evaluate them using the prototype. The cable driver proposed in this paper can drive pulleyless rolling joints stably with low cable tension. In addition, it can decouple yaw and pitch motion of the joints completely, therefore it can be applied to a variety of thin robots and instruments including steerable endoscopes and surgical robots.
{"title":"A Novel Cable Actuation Mechanism for 2-DOF Hyper-redundant Bending Robot Composed of Pulleyless Rolling Joints","authors":"Jungwook Suh","doi":"10.1109/IROS.2018.8593890","DOIUrl":"https://doi.org/10.1109/IROS.2018.8593890","url":null,"abstract":"Many surgical robots are remotely actuated by means of wire cables. In the past, the cables wound around circular pulleys at the robot joints did not constitute a problem of the cable driver structure. However, the pulleys inside the joints are removed recently in order to miniaturize the joints, so a specially designed cable driver suitable for the miniature joint structure is required for stable driving. In this paper, we propose a novel cable driver design for driving a pulleyless rolling joint and extend it to 2-DOF structure. Then, the proposed cable driver is manufactured using 3D printing with the 2-DOF bending joint, and an experiment is performed to evaluate them using the prototype. The cable driver proposed in this paper can drive pulleyless rolling joints stably with low cable tension. In addition, it can decouple yaw and pitch motion of the joints completely, therefore it can be applied to a variety of thin robots and instruments including steerable endoscopes and surgical robots.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"56 1 1","pages":"961-966"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83200989","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8593916
A. Mandow, J. Morales, J. A. Gómez-Ruiz, A. García-Cerezo
Multi-beam lidar (MBL) scanners are compact, light, and accessible 3D sensors with high data rates, but they offer limited vertical resolution and field of view (FOV). Some recent robotics research has profited from the addition of a degree-of-freedom (DOF) to an MBL to build rotating multibeam lidars (RMBL) that can achieve high-resolution scans with full spherical FOV. In a previous work, we offered a methodology to analyze the complex 3D scan measurement distributions produced by RMBLs with a rolling DOF and no pitching. In this paper, we investigate the effect of introducing constant pitch angles in the construction of the RMBLs with the purpose of finding a kinematic configuration that optimizes scan homogeneity with a spherical FOV. To this end, we propose a scalar index of 3D sensor homogeneity that is based on the spherical formulation of Ripley's K function. The optimization is performed for the widely used Puck (VLP-16) and HDL-32 sensors by Velodyne.
{"title":"Optimizing Scan Homogeneity for Building Full-3D Lidars Based on Rotating a Multi-Beam Velodyne Range-Finder","authors":"A. Mandow, J. Morales, J. A. Gómez-Ruiz, A. García-Cerezo","doi":"10.1109/IROS.2018.8593916","DOIUrl":"https://doi.org/10.1109/IROS.2018.8593916","url":null,"abstract":"Multi-beam lidar (MBL) scanners are compact, light, and accessible 3D sensors with high data rates, but they offer limited vertical resolution and field of view (FOV). Some recent robotics research has profited from the addition of a degree-of-freedom (DOF) to an MBL to build rotating multibeam lidars (RMBL) that can achieve high-resolution scans with full spherical FOV. In a previous work, we offered a methodology to analyze the complex 3D scan measurement distributions produced by RMBLs with a rolling DOF and no pitching. In this paper, we investigate the effect of introducing constant pitch angles in the construction of the RMBLs with the purpose of finding a kinematic configuration that optimizes scan homogeneity with a spherical FOV. To this end, we propose a scalar index of 3D sensor homogeneity that is based on the spherical formulation of Ripley's K function. The optimization is performed for the widely used Puck (VLP-16) and HDL-32 sensors by Velodyne.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"330 1","pages":"4788-4793"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80813264","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8594106
J. M. Prendergast, Gregory A. Formosa, C. Heckman, M. Rentschler
Capsule endoscopes have gained popularity over the last decade as minimally invasive devices for diagnosing gastrointestinal abnormalities such as colorectal cancer. While this technology offers a less invasive and more convenient alternative to traditional scopes, these capsules are only able to provide observational capabilities due to their passive nature. With the addition of a reliable mobility system and a real-time navigation system, capsule endoscopes could transform from observational devices into active surgical tools, offering biopsy and therapeutic capabilities and even autonomous navigation in a single minimally invasive device. In this work, a vision system is developed to allow for autonomous lumen center tracking and haustral fold identification and tracking during colonoscopy. This system is tested for its ability to accurately identify and track multiple haustral folds across many frames in both simulated and in vivo video, and the lumen center tracking is tested onboard a robotic endoscope platform (REP) within an active simulator to demonstrate autonomous navigation. In addition, real-time localization is demonstrated using open source ORB-SLAM2. The vision system successfully identified 95.6% of Haustral folds in simulator frames and 70.6% in in vivo frames and false positives occurred in less than 1% of frames. The center tracking algorithm showed in vivo center estimates within a mean error of 6.6% of physician estimates and allowed for the REP to traverse 2 m of the active simulator in 6 minutes without intervention.
{"title":"Autonomous Localization, Navigation and Haustral Fold Detection for Robotic Endoscopy","authors":"J. M. Prendergast, Gregory A. Formosa, C. Heckman, M. Rentschler","doi":"10.1109/IROS.2018.8594106","DOIUrl":"https://doi.org/10.1109/IROS.2018.8594106","url":null,"abstract":"Capsule endoscopes have gained popularity over the last decade as minimally invasive devices for diagnosing gastrointestinal abnormalities such as colorectal cancer. While this technology offers a less invasive and more convenient alternative to traditional scopes, these capsules are only able to provide observational capabilities due to their passive nature. With the addition of a reliable mobility system and a real-time navigation system, capsule endoscopes could transform from observational devices into active surgical tools, offering biopsy and therapeutic capabilities and even autonomous navigation in a single minimally invasive device. In this work, a vision system is developed to allow for autonomous lumen center tracking and haustral fold identification and tracking during colonoscopy. This system is tested for its ability to accurately identify and track multiple haustral folds across many frames in both simulated and in vivo video, and the lumen center tracking is tested onboard a robotic endoscope platform (REP) within an active simulator to demonstrate autonomous navigation. In addition, real-time localization is demonstrated using open source ORB-SLAM2. The vision system successfully identified 95.6% of Haustral folds in simulator frames and 70.6% in in vivo frames and false positives occurred in less than 1% of frames. The center tracking algorithm showed in vivo center estimates within a mean error of 6.6% of physician estimates and allowed for the REP to traverse 2 m of the active simulator in 6 minutes without intervention.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"99 1","pages":"783-790"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81178008","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}
Pub Date : 2018-10-01DOI: 10.1109/IROS.2018.8593633
Fethiye Irmak Dogan, Ilker Bozcan, Sinan Kalkan
There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.
{"title":"CINet: A Learning Based Approach to Incremental Context Modeling in Robots","authors":"Fethiye Irmak Dogan, Ilker Bozcan, Sinan Kalkan","doi":"10.1109/IROS.2018.8593633","DOIUrl":"https://doi.org/10.1109/IROS.2018.8593633","url":null,"abstract":"There have been several attempts at modeling context in robots. However, either these attempts assume a fixed number of contexts or use a rule-based approach to determine when to increment the number of contexts. In this paper, we pose the task of when to increment as a learning problem, which we solve using a Recurrent Neural Network. We show that the network successfully (with 98% testing accuracy) learns to predict when to increment, and demonstrate, in a scene modeling problem (where the correct number of contexts is not known), that the robot increments the number of contexts in an expected manner (i.e., the entropy of the system is reduced). We also present how the incremental model can be used for various scene reasoning tasks.","PeriodicalId":6640,"journal":{"name":"2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","volume":"40 1","pages":"4641-4646"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88766213","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：