Pub Date : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246927
X. Long, P. Long, T. Padır
This paper lays the foundations of risk-aware decision-making within the context of compositional robot autonomy for humanoid robots. In a nutshell, the idea is to compose task-level autonomous robot behaviors into a holistic motion plan by selecting a sequence of actions from a feasible action set. In doing so, we establish a total risk function to evaluate and assign a risk value to individual robot actions which then can be used to find the total risk of executing a plan. As a result, various actions can be composed into a complete autonomous motion plan while the robot is being cognizant to risks associated with executing one composition over another. In order to illustrate the concept, we introduce two specific risk measures, namely, the collision risk and the fall risk. We demonstrate the results from this foundational study of risk-aware compositional robot autonomy in simulation using NASA's Valkyrie humanoid robot.
{"title":"Compositional autonomy for humanoid robots with risk-aware decision-making","authors":"X. Long, P. Long, T. Padır","doi":"10.1109/HUMANOIDS.2017.8246927","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246927","url":null,"abstract":"This paper lays the foundations of risk-aware decision-making within the context of compositional robot autonomy for humanoid robots. In a nutshell, the idea is to compose task-level autonomous robot behaviors into a holistic motion plan by selecting a sequence of actions from a feasible action set. In doing so, we establish a total risk function to evaluate and assign a risk value to individual robot actions which then can be used to find the total risk of executing a plan. As a result, various actions can be composed into a complete autonomous motion plan while the robot is being cognizant to risks associated with executing one composition over another. In order to illustrate the concept, we introduce two specific risk measures, namely, the collision risk and the fall risk. We demonstrate the results from this foundational study of risk-aware compositional robot autonomy in simulation using NASA's Valkyrie humanoid robot.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124983379","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246916
L. Peternel, A. Ajoudani
In this paper we study the concept of robots learning from collaboration with skilled robots. The advantage of this concept is that the human involvement is reduced, while the skill can be propagated faster among the robots performing similar collaborative tasks or the ones being executed in hostile environments. The expert robot initially obtains the skill through the observation of, and physical collaboration with the human. We present a novel approach to how a novice robot can learn the specifics of the co-manipulation task from the physical interaction with an expert robot. The method consists of a multi-stage learning process that can gradually learn the appropriate motion and impedance behaviour under given task conditions. The trajectories are encoded with Dynamical Movement Primitives and learnt by Locally Weighted Regression, while their phase is estimated by adaptive oscillators. The learnt trajectories are replicated by a hybrid force/impedance controller. To validate the proposed approach we performed experiments on two robots learning and executing a challenging co-manipulation task.
{"title":"Robots learning from robots: A proof of concept study for co-manipulation tasks","authors":"L. Peternel, A. Ajoudani","doi":"10.1109/HUMANOIDS.2017.8246916","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246916","url":null,"abstract":"In this paper we study the concept of robots learning from collaboration with skilled robots. The advantage of this concept is that the human involvement is reduced, while the skill can be propagated faster among the robots performing similar collaborative tasks or the ones being executed in hostile environments. The expert robot initially obtains the skill through the observation of, and physical collaboration with the human. We present a novel approach to how a novice robot can learn the specifics of the co-manipulation task from the physical interaction with an expert robot. The method consists of a multi-stage learning process that can gradually learn the appropriate motion and impedance behaviour under given task conditions. The trajectories are encoded with Dynamical Movement Primitives and learnt by Locally Weighted Regression, while their phase is estimated by adaptive oscillators. The learnt trajectories are replicated by a hybrid force/impedance controller. To validate the proposed approach we performed experiments on two robots learning and executing a challenging co-manipulation task.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123528063","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246889
Miriam Bilac, Marine Chamoux, Angelica Lim
Let the human speak! Interactive robots and voice interfaces such as Pepper, Amazon Alexa, and OK Google are becoming more and more popular, allowing for more natural interaction compared to screens or keyboards. One issue with voice interfaces is that they tend to require a “robotic” flow of human speech. Humans must be careful to not produce disfluencies, such as hesitations or extended pauses between words. If they do, the agent may assume that the human has finished their speech turn, and interrupts them mid-thought. Interactive robots often rely on the same limited dialogue technology built for speech interfaces. Yet humanoid robots have the potential to also use their vision systems to determine when the human has finished their speaking turn. In this paper, we introduce HOMAGE (Human-rObot Multimodal Audio and Gaze End-of-turn), a multimodal turntaking system for conversational humanoid robots. We created a dataset of humans spontaneously hesitating when responding to a robot's open-ended questions such as, “What was your favorite moment this year?”. Our analyses found that users produced both auditory filled pauses such as “uhhh”, as well as gaze away from the robot to keep their speaking turn. We then trained a machine learning system to detect the auditory filled pauses and integrated it along with gaze into the Pepper humanoid robot's real-time dialog system. Experiments with 28 naive users revealed that adding auditory filled pause detection and gaze tracking significantly reduced robot interruptions. Furthermore, user turns were 2.1 times longer (without repetitions), suggesting that this strategy allows humans to express themselves more, toward less time pressure and better robot listeners.
让人类说话吧!交互式机器人和语音界面(如Pepper、Amazon Alexa和OK Google)正变得越来越受欢迎，与屏幕或键盘相比，它们允许更自然的交互。语音界面的一个问题是，它们往往需要一种“机器人式”的人类语言流。人们必须注意不要产生不流畅，比如单词之间的犹豫或长时间停顿。如果他们这样做，代理可能会认为人类已经完成了他们的演讲，并打断他们的思考。交互式机器人通常依赖于为语音界面构建的同样有限的对话技术。然而，人形机器人也有可能利用它们的视觉系统来确定人类何时完成了他们的讲话。在本文中，我们介绍了HOMAGE (Human-rObot Multimodal Audio and Gaze end -turn)，这是一个用于会话类人机器人的多模态轮转系统。我们创建了一个数据集，记录了人类在回答机器人提出的开放式问题时的自发犹豫，比如“你今年最喜欢的时刻是什么?”我们的分析发现，用户既会发出“啊”这样充满听觉的停顿，也会把目光从机器人身上移开，以保持说话的顺序。然后，我们训练了一个机器学习系统来检测充满听觉的停顿，并将其与凝视整合到Pepper人形机器人的实时对话系统中。对28名天真用户的实验表明，添加听觉填充暂停检测和凝视跟踪显著减少了机器人的干扰。此外，用户的回合数增加了2.1倍(没有重复)，这表明这种策略可以让人类更多地表达自己，减少时间压力，让机器人更好地倾听。
{"title":"Gaze and filled pause detection for smooth human-robot conversations","authors":"Miriam Bilac, Marine Chamoux, Angelica Lim","doi":"10.1109/HUMANOIDS.2017.8246889","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246889","url":null,"abstract":"Let the human speak! Interactive robots and voice interfaces such as Pepper, Amazon Alexa, and OK Google are becoming more and more popular, allowing for more natural interaction compared to screens or keyboards. One issue with voice interfaces is that they tend to require a “robotic” flow of human speech. Humans must be careful to not produce disfluencies, such as hesitations or extended pauses between words. If they do, the agent may assume that the human has finished their speech turn, and interrupts them mid-thought. Interactive robots often rely on the same limited dialogue technology built for speech interfaces. Yet humanoid robots have the potential to also use their vision systems to determine when the human has finished their speaking turn. In this paper, we introduce HOMAGE (Human-rObot Multimodal Audio and Gaze End-of-turn), a multimodal turntaking system for conversational humanoid robots. We created a dataset of humans spontaneously hesitating when responding to a robot's open-ended questions such as, “What was your favorite moment this year?”. Our analyses found that users produced both auditory filled pauses such as “uhhh”, as well as gaze away from the robot to keep their speaking turn. We then trained a machine learning system to detect the auditory filled pauses and integrated it along with gaze into the Pepper humanoid robot's real-time dialog system. Experiments with 28 naive users revealed that adding auditory filled pause detection and gaze tracking significantly reduced robot interruptions. Furthermore, user turns were 2.1 times longer (without repetitions), suggesting that this strategy allows humans to express themselves more, toward less time pressure and better robot listeners.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121882545","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246929
Phillip Hyatt, Marc D. Killpack
With humanoid robots becoming more complex and operating in un-modeled or human environments, there is a growing need for control methods that are scalable and robust, while still maintaining compliance for safety reasons. Model Predictive Control (MPC) is an optimal control method which has proven robust to modeling error and disturbances. However, it can be difficult to implement for high degree of freedom (DoF) systems due to the optimization problem that must be solved. While evolutionary algorithms have proven effective for complex large-scale optimization problems, they have not been formulated to find solutions quickly enough for use with MPC. This work details the implementation of a parallelized evolutionary MPC (EMPC) algorithm which is able to run in real-time through the use of a Graphics Processing Unit (GPU). This parallelization is accomplished by simulating candidate control input trajectories in parallel on the GPU. We show that this framework is more flexible in terms of cost function definition than traditional MPC and that it shows promise for finding solutions for high DoF systems.
{"title":"Real-time evolutionary model predictive control using a graphics processing unit","authors":"Phillip Hyatt, Marc D. Killpack","doi":"10.1109/HUMANOIDS.2017.8246929","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246929","url":null,"abstract":"With humanoid robots becoming more complex and operating in un-modeled or human environments, there is a growing need for control methods that are scalable and robust, while still maintaining compliance for safety reasons. Model Predictive Control (MPC) is an optimal control method which has proven robust to modeling error and disturbances. However, it can be difficult to implement for high degree of freedom (DoF) systems due to the optimization problem that must be solved. While evolutionary algorithms have proven effective for complex large-scale optimization problems, they have not been formulated to find solutions quickly enough for use with MPC. This work details the implementation of a parallelized evolutionary MPC (EMPC) algorithm which is able to run in real-time through the use of a Graphics Processing Unit (GPU). This parallelization is accomplished by simulating candidate control input trajectories in parallel on the GPU. We show that this framework is more flexible in terms of cost function definition than traditional MPC and that it shows promise for finding solutions for high DoF systems.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129795740","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8239531
Yu-Chi Lin, D. Berenson
In this paper we explore discrete search-based contact space planning for humanoids using both palm and foot contact in complex unstructured environments. With a high branching factor and sparse contactable regions, it is challenging for the planner to find a contact sequence in such environments quickly. Therefore, we propose to learn a function which predicts traversability — a measure of how quickly the contact space planner can generate contact sequences to traverse a certain region. By including a learned traversability estimate into the heuristic function of the contact space planner, we can bias the planner to search the areas with more contactable regions, and thus find contact sequences more efficiently. In this paper we propose and evaluate two kinds of feature vectors for estimating traversability: Exact Contact Checking (ECC) and Approximate Contact Checking (ACC), which make different trade-offs between speed and accuracy. The experimental results show that the proposed approach using ACC outperforms both ECC and the baseline heuristic for contact space planning; ACC increases the planning success rate by 19% and reduces average planning time by 24% compared to the baseline in difficult environments with uneven terrain.
{"title":"Humanoid navigation in uneven terrain using learned estimates of traversability","authors":"Yu-Chi Lin, D. Berenson","doi":"10.1109/HUMANOIDS.2017.8239531","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239531","url":null,"abstract":"In this paper we explore discrete search-based contact space planning for humanoids using both palm and foot contact in complex unstructured environments. With a high branching factor and sparse contactable regions, it is challenging for the planner to find a contact sequence in such environments quickly. Therefore, we propose to learn a function which predicts traversability — a measure of how quickly the contact space planner can generate contact sequences to traverse a certain region. By including a learned traversability estimate into the heuristic function of the contact space planner, we can bias the planner to search the areas with more contactable regions, and thus find contact sequences more efficiently. In this paper we propose and evaluate two kinds of feature vectors for estimating traversability: Exact Contact Checking (ECC) and Approximate Contact Checking (ACC), which make different trade-offs between speed and accuracy. The experimental results show that the proposed approach using ACC outperforms both ECC and the baseline heuristic for contact space planning; ACC increases the planning success rate by 19% and reduces average planning time by 24% compared to the baseline in difficult environments with uneven terrain.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130381443","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246904
Francesco Cursi, J. Malzahn, N. Tsagarakis, D. Caldwell
In this paper, an alternative novel method for calibrating a non-redundant six-axis force/torque sensor is presented. Calibration is conducted online, with the aid of a visual interactive interface, and is based on robust regression, allowing to discard outliers from the data and continuously monitor the sensor's calibration quality online while calibration data are being collected. The method is experimentally applied and validated in the calibration of the custom six-dimensional force/torque load-cells used in the feet of WALK-MAN humanoid.
{"title":"An online interactive method for guided calibration of multi-dimensional force/torque transducers","authors":"Francesco Cursi, J. Malzahn, N. Tsagarakis, D. Caldwell","doi":"10.1109/HUMANOIDS.2017.8246904","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246904","url":null,"abstract":"In this paper, an alternative novel method for calibrating a non-redundant six-axis force/torque sensor is presented. Calibration is conducted online, with the aid of a visual interactive interface, and is based on robust regression, allowing to discard outliers from the data and continuously monitor the sensor's calibration quality online while calibration data are being collected. The method is experimentally applied and validated in the calibration of the custom six-dimensional force/torque load-cells used in the feet of WALK-MAN humanoid.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122357458","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246970
Weiqiao Han, Russ Tedrake
To recover from large perturbations, a legged robot must make and break contact with its environment at various locations. These contact switches make it natural to model the robot as a hybrid system. If we apply Model Predictive Control to the feedback design of this hybrid system, the on/off behavior of contacts can be directly encoded using binary variables in a Mixed Integer Programming problem, which scales badly with the number of time steps and is too slow for online computation. We propose novel techniques for the design of stabilizing controllers for such hybrid systems. We approximate the dynamics of the system as a discrete-time Piecewise Affine (PWA) system, and compute the state feedback controllers across the hybrid modes offline via Lyapunov theory. The Lyapunov stability conditions are translated into Linear Matrix Inequalities. A Piecewise Quadratic Lyapunov function together with a Piecewise Linear (PL) feedback controller can be obtained by Semidefinite Programming (SDP). We show that we can embed a quadratic objective in the SDP, designing a controller approximating the Piecewise-Affine Quadratic Regulator. Moreover, we observe that our formulation restricted to the linear system case appears to always produce exactly the unique stabilizing solution to the Discrete Algebraic Riccati Equation. In addition, we extend the search from the PL controller to the PWA controller via Bilinear Matrix Inequalities. Finally, we demonstrate and evaluate our methods on a few PWA systems, including a simplified humanoid robot model.
{"title":"Feedback design for multi-contact push recovery via LMI approximation of the Piecewise-Affine Quadratic Regulator","authors":"Weiqiao Han, Russ Tedrake","doi":"10.1109/HUMANOIDS.2017.8246970","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246970","url":null,"abstract":"To recover from large perturbations, a legged robot must make and break contact with its environment at various locations. These contact switches make it natural to model the robot as a hybrid system. If we apply Model Predictive Control to the feedback design of this hybrid system, the on/off behavior of contacts can be directly encoded using binary variables in a Mixed Integer Programming problem, which scales badly with the number of time steps and is too slow for online computation. We propose novel techniques for the design of stabilizing controllers for such hybrid systems. We approximate the dynamics of the system as a discrete-time Piecewise Affine (PWA) system, and compute the state feedback controllers across the hybrid modes offline via Lyapunov theory. The Lyapunov stability conditions are translated into Linear Matrix Inequalities. A Piecewise Quadratic Lyapunov function together with a Piecewise Linear (PL) feedback controller can be obtained by Semidefinite Programming (SDP). We show that we can embed a quadratic objective in the SDP, designing a controller approximating the Piecewise-Affine Quadratic Regulator. Moreover, we observe that our formulation restricted to the linear system case appears to always produce exactly the unique stabilizing solution to the Discrete Algebraic Riccati Equation. In addition, we extend the search from the PL controller to the PWA controller via Bilinear Matrix Inequalities. Finally, we demonstrate and evaluate our methods on a few PWA systems, including a simplified humanoid robot model.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123387021","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246921
Jaesug Jung, Soonwook Hwang, Yisoo Lee, Jaehoon Sim, Jaeheung Park
This study investigates the position tracking performance of torque controlled humanoid robots in the presence of joint elasticity and time delay in torque command. One of the main purposes using torque control for humanoid robots is to achieve compliant behaviors on uncertain external disturbance such as uneven terrain and interaction with human. On the other hand, high performance of position tracking is also required to implement motion control of robots. In this study, the effects of joint elasticity and time delay in torque command area investigated in terms of position tracking. First, a joint model is derived and validated, which reflects the elasticity and time delay. Frequency response analysis is exploited to theoretically evaluate the performance of the control system for position tracking. This joint model with the elasticity and time delay is used to estimate the limitations in the controller design for our torque controlled humanoid robot. Theoretical analysis and its comparison with experimental results demonstrate that the joint elasticity and time delay significantly affect the system performance.
{"title":"Analysis of position tracking in torque control of humanoid robots considering joint elasticity and time delay","authors":"Jaesug Jung, Soonwook Hwang, Yisoo Lee, Jaehoon Sim, Jaeheung Park","doi":"10.1109/HUMANOIDS.2017.8246921","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246921","url":null,"abstract":"This study investigates the position tracking performance of torque controlled humanoid robots in the presence of joint elasticity and time delay in torque command. One of the main purposes using torque control for humanoid robots is to achieve compliant behaviors on uncertain external disturbance such as uneven terrain and interaction with human. On the other hand, high performance of position tracking is also required to implement motion control of robots. In this study, the effects of joint elasticity and time delay in torque command area investigated in terms of position tracking. First, a joint model is derived and validated, which reflects the elasticity and time delay. Frequency response analysis is exploited to theoretically evaluate the performance of the control system for position tracking. This joint model with the elasticity and time delay is used to estimate the limitations in the controller design for our torque controlled humanoid robot. Theoretical analysis and its comparison with experimental results demonstrate that the joint elasticity and time delay significantly affect the system performance.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128679348","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246886
A. Drimus, Vedran Mikov
This paper shows that it is possible to differentiate among various type of footwear solely by using highly dimensional pressure information provided by a sensorised insole. In order to achieve this, a person equipped with two sensorised insoles streaming real-time tactile data to a computer performs normal walking patterns. The sampled data is further transformed and reduced to sets of time series which are used for the classification of footwear. The pressure sensor is formed as a footwear inlay and is based on piezoresistive rubber having 1024 tactile cells providing normal pressure information in the form of a tactile image. The data is transmitted in realtime wirelessly at 30 fps from two such sensors. The online classification is using the dynamic time warping distances for different extracted features to assess the most similar type of footwear based on time series similarities. The paper shows that various footwear types yield distinct tactile patterns which can be assessed by the proposed algorithm.
{"title":"Footwear discrimination using dynamic tactile information","authors":"A. Drimus, Vedran Mikov","doi":"10.1109/HUMANOIDS.2017.8246886","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246886","url":null,"abstract":"This paper shows that it is possible to differentiate among various type of footwear solely by using highly dimensional pressure information provided by a sensorised insole. In order to achieve this, a person equipped with two sensorised insoles streaming real-time tactile data to a computer performs normal walking patterns. The sampled data is further transformed and reduced to sets of time series which are used for the classification of footwear. The pressure sensor is formed as a footwear inlay and is based on piezoresistive rubber having 1024 tactile cells providing normal pressure information in the form of a tactile image. The data is transmitted in realtime wirelessly at 30 fps from two such sensors. The online classification is using the dynamic time warping distances for different extracted features to assess the most similar type of footwear based on time series similarities. The paper shows that various footwear types yield distinct tactile patterns which can be assessed by the proposed algorithm.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115491986","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246915
Atabak Dehban, L. Jamone, A. R. Kampff, J. Santos-Victor
The perception of affordances provides an action-centered parametric representation of the environment. By perceiving an object's visual features in terms of what actions they afford, novel behavior opportunities can be inferred about previously unseen objects. In this paper, a flexible deep probabilistic framework is proposed which allows an explorative agent to learn tool-object affordances in continuous space. To this end, we use a deep variational auto-encoder with heterogeneous probabilistic distributions to infer the most probable action that achieves a desired effect or to predict a parametric probability distribution over action consequences i.e. effects. Our experiments show the generalization of the method to unseen objects and tools and we have analyzed the influence of different design choices. Our framework goes beyond other proposals by incorporating various probability distributions tailored for each individual modality and by eliminating the need for any pre-processing of the data.
{"title":"A deep probabilistic framework for heterogeneous self-supervised learning of affordances","authors":"Atabak Dehban, L. Jamone, A. R. Kampff, J. Santos-Victor","doi":"10.1109/HUMANOIDS.2017.8246915","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246915","url":null,"abstract":"The perception of affordances provides an action-centered parametric representation of the environment. By perceiving an object's visual features in terms of what actions they afford, novel behavior opportunities can be inferred about previously unseen objects. In this paper, a flexible deep probabilistic framework is proposed which allows an explorative agent to learn tool-object affordances in continuous space. To this end, we use a deep variational auto-encoder with heterogeneous probabilistic distributions to infer the most probable action that achieves a desired effect or to predict a parametric probability distribution over action consequences i.e. effects. Our experiments show the generalization of the method to unseen objects and tools and we have analyzed the influence of different design choices. Our framework goes beyond other proposals by incorporating various probability distributions tailored for each individual modality and by eliminating the need for any pre-processing of the data.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115248921","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：