2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)最新文献_第6页

Improving Humanoid Grasp Success Rate based on Uncertainty-aware Metrics and Sensitivity Optimization 基于不确定性感知度量和灵敏度优化的仿人抓取成功率提高

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000206

W.-J. Baek, C. Pohl, Philipp Pelcz, T. Kröger, T. Asfour

{"title":"Improving Humanoid Grasp Success Rate based on Uncertainty-aware Metrics and Sensitivity Optimization","authors":"W.-J. Baek, C. Pohl, Philipp Pelcz, T. Kröger, T. Asfour","doi":"10.1109/Humanoids53995.2022.10000206","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000206","url":null,"abstract":"We present an approach for the selection of robot grasp candidates by treating specified metrics in a probabilistic manner and maximizing the success rate through statistical optimization. Recently, progress has been made in grasping unknown objects in cluttered scenes by using deep neural networks or incorporating classifiers. Although existing methods deliver promising results, they either lack explainability or fail to account for uncertainties that accumulate over the entire system. To address this shortcoming, we optimize a ranking score based on the sensitivities of the grasp success with respect to a set of metrics. These sensitivities reflect each metric's contribution to the success. To perform this optimization, we refer to a dataset of 932 randomly selected grasps recorded under real-world conditions with the humanoid robot ARMAR-6. By validating our approach on a separate data collection of 187 physical real- world grasps, we demonstrate that our approach yields a success rate of 73.8 %, amounting to an improvement of more than 40 % compared to a random grasp selection. The results exemplify that sensitivity optimization, scarcely applied in the context of robotic applications so far, can significantly enhance the grasp success by considering respective metrics in the face of uncertainties.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126916016","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

Optimizing Facial Expressions of an Android Robot Effectively: a Bayesian Optimization Approach 有效优化Android机器人面部表情:贝叶斯优化方法

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000154

Dongsheng Yang, Wataru Sato, Qianying Liu, T. Minato, Shushi Namba, Shin’ya Nishida

{"title":"Optimizing Facial Expressions of an Android Robot Effectively: a Bayesian Optimization Approach","authors":"Dongsheng Yang, Wataru Sato, Qianying Liu, T. Minato, Shushi Namba, Shin’ya Nishida","doi":"10.1109/Humanoids53995.2022.10000154","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000154","url":null,"abstract":"Expressing various facial emotions is an important social ability for efficient communication between humans. A key challenge in human-robot interaction research is providing androids with the ability to make various human-like facial expressions for efficient communication with humans. The android Nikola, we have developed, is equipped with many actuators for facial muscle control. While this enables Nikola to simulate various human expressions, it also complicates identification of the optimal parameters for producing desired expressions. Here, we propose a novel method that automati-cally optimizes the facial expressions of our android. We use a machine vision algorithm to evaluate the magnitudes of seven basic emotions, and employ the Bayesian Optimization algorithm to identify the parameters that produce the most convincing facial expressions. Evaluations by naïve human participants demonstrate that our method improves the rated strength of the android's facial expressions of anger, disgust, sadness, and surprise compared with the previous method that relied on Ekman's theory and parameter adjustments by a human expert.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125670910","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

Biomechanical Risk Assessment of Human Lifting Tasks via Supervised Classification of Multiple Sensor Data 基于多传感器数据监督分类的人类起重作业生物力学风险评估

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000147

A. Santopaolo, Marta Lorenzini, Luigi Privitera, T. Varrecchia, G. Chini, A. Ranavolo, P. Ariano, A. Ajoudani

{"title":"Biomechanical Risk Assessment of Human Lifting Tasks via Supervised Classification of Multiple Sensor Data","authors":"A. Santopaolo, Marta Lorenzini, Luigi Privitera, T. Varrecchia, G. Chini, A. Ranavolo, P. Ariano, A. Ajoudani","doi":"10.1109/Humanoids53995.2022.10000147","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000147","url":null,"abstract":"Manual lifting tasks are among the primary causes of work-related lower back disorders (WLBD), which are the most common and costly musculoskeletal conditions reported. Aiming to improve risk prevention, the National Institute for Occupational Safety and Health (NIOSH) established a method to evaluate lifting activities based on the kinematic parameters of the lift. The resulting Lifting Index (LI) proved to be a good indicator of the associated biomechanical risk, but it only considers job-related factors, is constrained by equations and parameters, and cannot be calculated when lifting is performed with the assistance of a human-robot collaboration technology such as an exoskeleton. In this paper, we exploit a k-nearest neighbors algorithm to combine and compare different types of sensor information in their ability to classify the risk level associated with lifting tasks. Data are collected on eight healthy participants while performing six load lifting under different task conditions. An instantaneous lifting index (i-LI) is estimated to refine the risk computation. Based on it, an actual lifting index (a-LI) is computed to train the learning algorithm. Then, three different data sets are designed, which include only kinematic data, only muscle electrical activity data, and their combination, respectively, and compared based on the algorithm's performance. Results prove that our framework can classify the ergonomic risk level with high accuracy and show its potential in the automatic and comprehensive assessment of lifting tasks. A very similar performance was found among different sensor data, highlighting its generalization capability.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115986516","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: 1

Accelerating Interactive Human-like Manipulation Learning with GPU-based Simulation and High-quality Demonstrations 利用基于gpu的仿真和高质量演示加速交互式类人操作学习

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000161

Malte Mosbach, Kara Moraw, Sven Behnke

{"title":"Accelerating Interactive Human-like Manipulation Learning with GPU-based Simulation and High-quality Demonstrations","authors":"Malte Mosbach, Kara Moraw, Sven Behnke","doi":"10.1109/Humanoids53995.2022.10000161","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000161","url":null,"abstract":"Dexterous manipulation with anthropomorphic robot hands remains a challenging problem in robotics because of the high-dimensional state and action spaces and complex contacts. Nevertheless, skillful closed-loop manipulation is required to enable humanoid robots to operate in unstructured real-world environments. Reinforcement learning (RL) has traditionally imposed enormous interaction data requirements for optimizing such complex control problems. We introduce a new framework that leverages recent advances in GPU-based simulation along with the strength of imitation learning in guiding policy search towards promising behaviors to make RL training feasible in these domains. To this end, we present an immersive virtual reality teleoperation interface designed for interactive human-like manipulation on contact rich tasks and a suite of manipulation environments inspired by tasks of daily living. Finally, we demonstrate the complementary strengths of massively parallel RL and imitation learning, yielding robust and natural behaviors. Videos of trained policies, our source code, and the collected demonstration datasets are available at https://maltemosbach.github.io/interactive_human_like_manipulation/.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116024218","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

Grasp Pose Sampling for Precision Grasp Types with Multi-fingered Robotic Hands 多指机械手精确抓取类型的抓取姿态采样

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000203

D. Dimou, J. Santos-Victor, Plinio Moreno

{"title":"Grasp Pose Sampling for Precision Grasp Types with Multi-fingered Robotic Hands","authors":"D. Dimou, J. Santos-Victor, Plinio Moreno","doi":"10.1109/Humanoids53995.2022.10000203","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000203","url":null,"abstract":"Generation of promising hand and finger poses for multi-fingered robotic hands cannot be simplified as the 2-dimensional model for grippers. Current approaches rely on heuristics that reduce the search space while ignoring a large number of candidates. We present a generative model that samples 6DoF poses for several types of precision grasps. Similarly to previous works, we start with a geometric heuristic to gather data. However, with a large enough samples we are able to sample grasp poses that are by a large margin more successful than using the heuristics. The model consists of 3 cascaded generative models that are based on the conditional Variational Auto-Encoder framework, and takes as input the desired grasp type, the object label, and the object's size. It generates a grasp posture, meaning the configuration of the fingers of the robotic hand, and a 6DoF pose. Our cascaded model samples first the finger joint configuration, followed by the Cartesian position of the object and finally the rotation of the object, our sampler divides the 6DoF in simpler problems, which lead to more successful grasps. In our experiments we show that our model improves the percentage of successful grasps sampled compared to the heuristic and compare several variants of the model to support our design choices, showing the benefits of the cascaded sampling.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123300565","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

Whole-Body Model Predictive Control for Biped Locomotion on a Torque-Controlled Humanoid Robot 力矩控制仿人机器人两足运动的全身模型预测控制

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000129

Ewen Dantec, M. Naveau, Pierre Fernbach, N. Villa, Guilhem Saurel, O. Stasse, M. Taïx, N. Mansard

{"title":"Whole-Body Model Predictive Control for Biped Locomotion on a Torque-Controlled Humanoid Robot","authors":"Ewen Dantec, M. Naveau, Pierre Fernbach, N. Villa, Guilhem Saurel, O. Stasse, M. Taïx, N. Mansard","doi":"10.1109/Humanoids53995.2022.10000129","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000129","url":null,"abstract":"Locomotion of biped robots requires predictive controllers due to its unstable dynamics and physical limitations of contact forces. A real-time controller designed to perform complex motions while maintaining balance over feet must generate whole-body trajectories, predicting a few seconds in the future with a high enough updating rate to reduce model errors. Due to the huge computational power demanded by such solvers, future trajectories are usually generated using a reduced order model that contains the unstable dynamics. However, this simplification introduces feasibility problems on many edge cases. Considering the permanent improvement of computers and algorithms, whole-body locomotion in real-time is becoming a viable option for humanoids, and this article aims at illustrating this point. We propose a whole-body model predictive control scheme based on differential dynamic programming that takes into account the full dynamics of the system and decides the optimal actuation for the robot's lower body (20 degrees of freedom) along a preview horizon of 1.5 s. Our experimental validation on the torque-controlled robot Talos shows good and promising results for dynamic locomotion at different gaits as well as 10 cm height stairstep crossing.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127527517","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: 9

Estimation of Human Base Kinematics using Dynamical Inverse Kinematics and Contact-Aided Lie Group Kalman Filter 基于动态逆运动学和接触辅助李群卡尔曼滤波的人体基础运动学估计

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000199

Prashanth Ramadoss, Lorenzo Rapetti, Yeshasvi Tirupachuri, Riccardo Grieco, Gianluca Milani, Enrico Valli, Stefano Dafarra, Silvio Traversaro, D. Pucci

引用次数: 1

Combining Reinforcement Learning and Lazy Learning for Faster Few-Shot Transfer Learning 结合强化学习和惰性学习实现更快的几次迁移学习

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000095

Zvezdan Loncarevic, Mihael Simonič, A. Ude, A. Gams

{"title":"Combining Reinforcement Learning and Lazy Learning for Faster Few-Shot Transfer Learning","authors":"Zvezdan Loncarevic, Mihael Simonič, A. Ude, A. Gams","doi":"10.1109/Humanoids53995.2022.10000095","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000095","url":null,"abstract":"Since repeating a task with a humanoid robot many times is typically far too time consuming and strenuous for the robotic mechanism, learning is often shifted to simulation. Bridging the sim-to-real gap, however, still requires considerable real-world effort. In this paper we explore how to reduce the number of required repetitions with a novel few-shot transfer learning methodology. The skill is initially encoded with a deep neural network in one domain, and later adapted for a different target domain by re-training only a partllayer of this deep neural network with real data. For retraining we propose to combine lazy learning and reinforcement learning. Our experiments show that such combination is considerably faster than only using either one of these and an order of magnitude faster than learning from scratch. We demonstrated the approach on the example of robotic throwing, a complex dynamic skill where the outcome of the task is not explicitly dependent on the final position of the robot motion. The experiments were performed for sim-to-sim transfer learning on the full-sized humanoid robot TALOS, with initial throwing implementation on the real robot.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128132154","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: 1

SoftTouch: A Sensor-Placement Framework for Soft Robotic Hands 软触:一种用于软机械手的传感器放置框架

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000138

C. Li, N. Pollard

{"title":"SoftTouch: A Sensor-Placement Framework for Soft Robotic Hands","authors":"C. Li, N. Pollard","doi":"10.1109/Humanoids53995.2022.10000138","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000138","url":null,"abstract":"Sensor placement for grasping tasks in conventional robotic hands has been extensively studied, with goals including sensorizing essential contact areas or determining the effect of number of sensors on performance. However, with the new generation of dexterous soft robotic hands that deform to the shape of the object, the former frameworks may not be sufficient. In particular, we find that real-world experiments are essential to determine the value of different sensors and the effect of different sensor placements due to the complex interactions between the deformable robot body, sensor material properties, and sensor and task performance. In this paper, we propose a sensor-placement framework for dexterous soft robotic hands that is easily reconfigurable to different hand designs using off-the-shelf sensors. Our three-step framework selects and evaluates candidate sensor configurations to de-termine the effectiveness of sensors in each configuration for estimating qualitative and quantitative manipulation metrics. We tested our framework on a soft robotic hand to select the optimum sensor placement for a given set of manipulation patterns using force and inertial sensors. Our studies show that sensors placed at contact points are best for predicting the qualitative success of the manipulation. However, when it comes to estimating quantitative manipulation metrics, off-the-shelf sensors placed at contact points decrease performance for some manipulation types. This performance decrease may be due to the disturbance they create to surface texture, deformation patterns, and weight of soft robotic systems.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"299 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133236599","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: 1

Quadratically constrained quadratic programs using approximations of the step-to-step dynamics: application on a 2D model of Digit 采用步进动力学近似的二次约束二次规划:在二维数字模型上的应用

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000251

Ernesto Hernandez-Hinojosa, Daniel Torres, Pranav A. Bhounsule

{"title":"Quadratically constrained quadratic programs using approximations of the step-to-step dynamics: application on a 2D model of Digit","authors":"Ernesto Hernandez-Hinojosa, Daniel Torres, Pranav A. Bhounsule","doi":"10.1109/Humanoids53995.2022.10000251","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000251","url":null,"abstract":"Bipedal robots are yet to achieve mainstream application because they lack robustness in real-world settings. One of the major control challenges arises due to the ankle motors' limited control authority, which prevents these robots from being fully controllable at a particular instant (e.g., like an inverted pendulum). We show that to stabilize such robots, they must achieve stability over the time scale of a step, also known as step-to-step (S2S) stability. Past approaches have used the linearization of the S2S dynamics to develop controllers, but these have limited regions of validity. Here, we use a data-driven approach to approximate S2S dynamics, including its region of validity. Our results show that linear and quadratic models can approximate the region of validity and S2S dynamics, respectively. We show that the quadratic S2S approximation generated using a data-driven full-body dynamics simulator outperforms those generated using the analytical linear S2S generated from the popularly used linear inverted pendulum model (LIPM). The S2S approximation enables us to formulate and solve a quadratically constrained quadratic program to develop walking controllers. We demonstrate the efficacy of the approach in simulation using a 2D model of Digit walking on patterned terrain. A video is linked here: https://youtu.be/MniABg2jGEA","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131325317","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