发布求助
文献互助 智能选刊 最新文献

2020 Fourth IEEE International Conference on Robotic Computing (IRC)最新文献

筛选
英文 中文
Formal Verification for Safe Deep Reinforcement Learning in Trajectory Generation 轨迹生成中安全深度强化学习的形式化验证
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00062
Davide Corsi, Enrico Marchesini, A. Farinelli, P. Fiorini
{"title":"Formal Verification for Safe Deep Reinforcement Learning in Trajectory Generation","authors":"Davide Corsi, Enrico Marchesini, A. Farinelli, P. Fiorini","doi":"10.1109/IRC.2020.00062","DOIUrl":"https://doi.org/10.1109/IRC.2020.00062","url":null,"abstract":"We consider the problem of Safe Deep Reinforcement Learning (DRL) using formal verification in a trajectory generation task. In more detail, we propose an approach to verify whether a trained model can generate trajectories that are guaranteed to meet safety properties (e.g., operate in a limited work-space). We show that our verification approach based on interval analysis, provably guarantees whether a model meets pre-specified safety properties and it returns the input values that cause a violation of such properties. Furthermore, we show that an optimized DRL approach (i.e., using scaling discount factor and a mixed exploration policy based on a directional controller) can reach the target with millimeter precision while reducing the set of inputs that cause safety violations. Crucially, in our experiments, the number of undesirable inputs is so low that they can be directly removed with a simple controller.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116983473","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}
引用次数: 12
Self-supervised learning of object slippage: An LSTM model trained on low-cost tactile sensors 物体滑动的自监督学习:基于低成本触觉传感器训练的LSTM模型
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00038
Ainur Begalinova, Ross D. King, B. Lennox, R. Batista-Navarro
{"title":"Self-supervised learning of object slippage: An LSTM model trained on low-cost tactile sensors","authors":"Ainur Begalinova, Ross D. King, B. Lennox, R. Batista-Navarro","doi":"10.1109/IRC.2020.00038","DOIUrl":"https://doi.org/10.1109/IRC.2020.00038","url":null,"abstract":"This paper presents a combination of machine learning techniques for slip detection in grasping, based on temporal features collected by low-cost tactile sensors. A slippage is an event that is subsequent to prior micro-slippages that have occurred at hand-object contact. The method is based on the application of a sequential classification technique (a variant of recurrent neural networks known as long short-term memory networks or LSTMs), whereby time-series pressure readings from tactile sensors are classified as either slip or non-slip events. We also propose a novel method for autonomous labelling, removing the need for humans in the labelling process. Lastly, this paper proposes a new design for an adaptable wearable tactile sensing device that integrates non-expensive sensors. Our proposed method achieved high accuracy in the classification of slip and non-slip events, obtaining over 95% in offline classification and 89% in online classification using a Sawyer robot.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121873825","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}
引用次数: 4
Catching a Robot Intruder with Limited Information 用有限的信息捕捉机器人入侵者
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00010
D. Nussbaum, Fedor Ilitchev
{"title":"Catching a Robot Intruder with Limited Information","authors":"D. Nussbaum, Fedor Ilitchev","doi":"10.1109/IRC.2020.00010","DOIUrl":"https://doi.org/10.1109/IRC.2020.00010","url":null,"abstract":"This paper provides solutions and insight into a new set of problems of catching a mobile robot intruder using limited information about the intruder's location. The information about the intruder's location, which is termed a snapshot, is only available upon request. We formulate the problem of tracking and catching an intruder with limited number of snapshots, which we termed the Moving Target Search with Snapshots (MTSWS). In the MTSWS problem a mobile guard Ah is chasing a mobile intruder At in R2. Here, Ah knows the location of At either from a requested snapshot or if Ah is sufficiently close to At. The objective is to reduce the number of required snapshots and/or to reduce the distance travelled by Ah• We compute the number of snapshots that are necessary and sufficient to catch an intruder in the worst case. We also provide algorithmic solutions under a number of assumptions on the intruder's behaviour (e.g., intruder's random motion where At is oblivious to the actions taken by Ah• Last we provide solution to the problem when Ah is allowed to use $k$ snapshots and determine the locations and the time that Ah should take the snapshot.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128656233","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}
引用次数: 2
Remarks on Adaptive Compensator with Quaternion Neural Network in Computed Torque Control 四元数神经网络自适应补偿器在计算转矩控制中的应用
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00084
Kazuhiko Takahashi
{"title":"Remarks on Adaptive Compensator with Quaternion Neural Network in Computed Torque Control","authors":"Kazuhiko Takahashi","doi":"10.1109/IRC.2020.00084","DOIUrl":"https://doi.org/10.1109/IRC.2020.00084","url":null,"abstract":"Model-based control such as computed torque control is frequently employed to ensure the accurate control of a robot manipulator. However, in some cases control performance is not satisfactory due to unmodeled nonlinearities or dynamics. To overcome this issue, this study investigates how using a quaternion neural network can adaptively compensate for the computed torque control. The control system consists of the quaternion neural network, feedforward model and feedback controller, resulting in a feedback error learning scheme utilised for the training of the quaternion neural network with a backpropagation algorithm extended to quaternion numbers. In computational experiments, the trajectory control of a three-link robot manipulator is performed using the proposed control system. Simulation results confirm the effectiveness of the quaternion neural network in practical control applications.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129735532","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}
引用次数: 0
Multi-label UAV sound classification using Stacked Bidirectional LSTM 基于堆叠双向LSTM的多标签无人机声音分类
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00086
D. Utebayeva, A. Almagambetov, Manal Alduraibi, Yelmurat Temirgaliyev, L. Ilipbayeva, Sungat Marxuly
{"title":"Multi-label UAV sound classification using Stacked Bidirectional LSTM","authors":"D. Utebayeva, A. Almagambetov, Manal Alduraibi, Yelmurat Temirgaliyev, L. Ilipbayeva, Sungat Marxuly","doi":"10.1109/IRC.2020.00086","DOIUrl":"https://doi.org/10.1109/IRC.2020.00086","url":null,"abstract":"Nowadays Unmanned Aerial Vehicles (UAVs) pose an increasing threat to public areas such as parks, schools, hospitals and official buildings. Different methods of dealing with UAV detection are developing more and more actively. This paper primarily focuses on two key aims: the first aim is to perform a multi-label classification system and the second aim is to develop Stacked Bidirectional Long Short-Term Memory (LSTM) with two hidden layers to categorize multiple UAVs sounds. Frame-wise spectral-domain features are applied as inputs of the proposed system. Overall, the results of the study show that the sound of UAVs can be classified into multiple labels. This study has been one of the first attempts to thoroughly examine Stacked Bidirectional LSTM for UAV sound classification task.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121976025","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}
引用次数: 6
Dynamics Modelling and Parameter Identification of a Reaction Wheel Based Pendulum 反作用轮摆的动力学建模与参数辨识
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00049
C. Su, Chao-Chung Peng, Ankit A. Ravankar, A. Ravankar
{"title":"Dynamics Modelling and Parameter Identification of a Reaction Wheel Based Pendulum","authors":"C. Su, Chao-Chung Peng, Ankit A. Ravankar, A. Ravankar","doi":"10.1109/IRC.2020.00049","DOIUrl":"https://doi.org/10.1109/IRC.2020.00049","url":null,"abstract":"In control system research, the pendulum system is one of the important nonlinear systems, and it has been widely used in many control applications. Different from the conventional pendulum system, this paper presents a reaction wheel based pendulum control system. From the application point of view, the design task is to stabilize the pendulum system through the reaction force generated from the flywheel. To this aim, dynamics modelling of the reaction wheel based pendulum must be considered first. Second, the system parameter identification scheme is required for future controller designs. Therefore, in this paper, dynamics modelling of the reaction wheel based pendulum is derived in term of Lagrange equation. For the associated nonlinear model, a discrete model approximation is proposed for the identification of system parameters. Finally, simulations are carried out to demonstrate the feasibility and effectiveness of the proposed method.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121624530","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}
引用次数: 0
An Artificial Soft Somatosensory System for a Cognitive Robot 用于认知机器人的人工软体感系统
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00058
A. Augello, Ignazio Infantino, S. Gaglio, U. Maniscalco, G. Pilato, Filippo Vella
{"title":"An Artificial Soft Somatosensory System for a Cognitive Robot","authors":"A. Augello, Ignazio Infantino, S. Gaglio, U. Maniscalco, G. Pilato, Filippo Vella","doi":"10.1109/IRC.2020.00058","DOIUrl":"https://doi.org/10.1109/IRC.2020.00058","url":null,"abstract":"The paper proposes an artificial somatosensory system loosely inspired by human beings' biology and embedded in a cognitive architecture (CA). It enables a robot to receive the stimulation from its embodiment, and use these sensations, we called roboceptions, to behave according to both the external environment and the internal robot status. In such a way, the robot is aware of its body and able to interpret physical sensations can be more effective in the task while maintaining its well being. The robot's physiological urges are tightly bound to the specific physical state of the robot. Positive and negative physical information can, therefore, be processed and let the robot behave in a more realistic way adopting the right trade-off between the achievement of the task and the well-being of the robot. This goal has been achieved through a reinforcement learning approach. To test these statements we considered, as a test-bench, the execution of working performances with an SoftBank NAO robot that are modulated according its body well-being.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130719956","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}
引用次数: 3
SAFESTOP: Disturbance Handling in Prioritized Multi-robot Trajectory Planning 安全停止:优先多机器人轨迹规划中的干扰处理
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00043
Felix Keppler, Sebastian Wagner, K. Janschek
{"title":"SAFESTOP: Disturbance Handling in Prioritized Multi-robot Trajectory Planning","authors":"Felix Keppler, Sebastian Wagner, K. Janschek","doi":"10.1109/IRC.2020.00043","DOIUrl":"https://doi.org/10.1109/IRC.2020.00043","url":null,"abstract":"Prioritized planning allows the generation of coordinated motion plans for multiple robots in a shared workspace by tracking their spatiotemporal progression along pre-planned paths. With the knowledge where and when collisions would occur, collision-free velocity profiles are calculated according to a prioritization scheme. However, in real-world applications, disturbances in the exact execution of the planned motions are likely. This can affect other planned trajectories and leave the system prone to deadlocks. We present a novel methodology, SAFESTOP, for handling interruptions with minimal impact on other planned tasks and show that it can drastically reduce the number of affected robots as well as the overall delay. The key idea is to maintain the paths and vary the motion of the affected robots, making them yield for others. In contrast to other approaches it supports complex vehicle structures and takes kinematic and dynamic constraints into account.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128877081","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}
引用次数: 0
Design of a Reusable Controller for Pushing Unliftable Objects Using a Multicopter 多旋翼机推不可举物体的可重复使用控制器设计
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00026
N. Shirakura, G. A. G. Ricardez, J. Takamatsu, T. Ogasawara
{"title":"Design of a Reusable Controller for Pushing Unliftable Objects Using a Multicopter","authors":"N. Shirakura, G. A. G. Ricardez, J. Takamatsu, T. Ogasawara","doi":"10.1109/IRC.2020.00026","DOIUrl":"https://doi.org/10.1109/IRC.2020.00026","url":null,"abstract":"Research on aerial manipulation, or the manipulation of objects using multicopters, has been attracting attention in recent years. Commonly, aerial manipulation is performed with a manipulator mounted on a multicopter. However, these research projects focus on manipulating objects which can be grasped and lifted. This research focuses on manipulating unliftable objects using a multicopter. With the proposed method, the multicopter pushes a target object. To push an object automatically, a physical model for the pushing motion is constructed. An attitude control part of an existing flight controller is reused for reducing the effort of implementing the controller. The pushing motion is achieved by controlling the attitude using the physical model. In the evaluation experiments, the effectiveness of the proposed method is verified in simulation.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115538996","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}
引用次数: 0
Robust Distance Estimation of Capacitive Proximity Sensors in HRI using Neural Networks 基于神经网络的电容式接近传感器鲁棒距离估计
2020 Fourth IEEE International Conference on Robotic Computing (IRC) Pub Date : 2020-11-01 DOI: 10.1109/IRC.2020.00061
Alexander Poeppel, A. Hoffmann, Martin Siehler, W. Reif
{"title":"Robust Distance Estimation of Capacitive Proximity Sensors in HRI using Neural Networks","authors":"Alexander Poeppel, A. Hoffmann, Martin Siehler, W. Reif","doi":"10.1109/IRC.2020.00061","DOIUrl":"https://doi.org/10.1109/IRC.2020.00061","url":null,"abstract":"With Industry 4.0 and the idea of flexible and hybrid manufacturing systems, the need for safe human-robot-interaction has become increasingly important. For this purpose, it is necessary to reliably detect persons in the workspace of a robot. Capacitive sensors mounted to the robot structure can be used to measure the presence of conductive objects and, hence, allow the detection of persons. However, for a reliable detection in an adequate range, it is necessary to compensate for various additional influences on capacitive sensors. In this paper, we propose a segmentation of the world model and the use of machine learning to compensate for the self-influence of the robot. Moreover, a correlation between the measured capacitance and the distance of a human hand can be calculated using machine learning, as well. Finally, it is possible to estimate the distance between capacitive sensors at the robot structure and a human hand while the robot is in motion. A motion capturing system is used as ground truth for the distance.","PeriodicalId":232817,"journal":{"name":"2020 Fourth IEEE International Conference on Robotic Computing (IRC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127891178","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}
引用次数: 5
下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信