2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)最新文献_第4页

A complete framework for personalised modeling and control of children's cerebral palsy 儿童脑瘫个性化建模与控制的完整框架

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000230

Sabrina Otmani, G. Michon, B. Watier

引用次数: 0

From Offline to Online: A Perception-Based Local Planner for Dynamic Obstacle Avoidance 从离线到在线:基于感知的动态避障局部规划

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000245

L. Rossini, N. Tsagarakis

{"title":"From Offline to Online: A Perception-Based Local Planner for Dynamic Obstacle Avoidance","authors":"L. Rossini, N. Tsagarakis","doi":"10.1109/Humanoids53995.2022.10000245","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000245","url":null,"abstract":"The deployment of robots within realistic environments necessitates robots to be capable of replanning their loco-manipulation trajectories on the fly to avoid unexpected interactions that may occur due to the uncertainty that is present in such dynamic and varying workspaces. This work introduces a novel method for the online local replanning of precomputed global trajectories for redoundant robots. The local nature of the problem leads to a sparse system that a hyper-graph can encode more intuitively. Using a graph, we can also store the entire global trajectory, preventing the local planner from getting stuck in local minima, and vertices and edges can be dynamically added or removed to ignore those constraints that do not interfere with the local problem, further increasing the computational efficiency. This process is accompanied by a control layer that iteratively takes the online refined solution and safely moves the robot. The method is validated both in simulation and experimentally on the wheeled-legged quadrupedal robot CENTAURO, demonstrating its effectiveness in replanning online the loco-manipulation trajectories of the robot under the occurrence of obstacles that intervene with the initially planned trajectories.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"285 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120874568","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

Adapting Object-Centric Probabilistic Movement Primitives with Residual Reinforcement Learning 残差强化学习适应以对象为中心的概率运动原语

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000148

João Carvalho, Dorothea Koert, Marek Daniv, Jan Peters

{"title":"Adapting Object-Centric Probabilistic Movement Primitives with Residual Reinforcement Learning","authors":"João Carvalho, Dorothea Koert, Marek Daniv, Jan Peters","doi":"10.1109/Humanoids53995.2022.10000148","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000148","url":null,"abstract":"It is desirable for future robots to quickly learn new tasks and adapt learned skills to constantly changing environments. To this end, Probabilistic Movement Primitives (ProMPs) have shown to be a promising framework to learn generalizable trajectory generators from distributions over demonstrated trajectories. However, in practical applications that require high precision in the manipulation of objects, the accuracy of ProMPs is often insufficient, in particular when they are learned in cartesian space from external observations and executed with limited controller gains. Therefore, we propose to combine ProMPs with the Residual Reinforcement Learning (RRL) framework, to account for both, corrections in position and orientation during task execution. In particular, we learn a residual on top of a nominal ProMP trajectory with Soft Actor-Critic and incorporate the variability in the demonstrations as a decision variable to reduce the search space for RRL. As a proof of concept, we evaluate our proposed method on a 3D block insertion task with a 7-DoF Franka Emika Panda robot. Experimental results show that the robot successfully learns to complete the insertion, which was not possible before with using basic ProMPs.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"1 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":"131134596","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

Passivity-based multi-contact balancing of humanoid robots based on power-conserved transformations 基于功率守恒变换的仿人机器人无源多接触平衡

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000174

R. Ozawa, Takumi Kamioka

引用次数: 0

End-To-End Deep Reinforcement Learning for First-Person Pedestrian Visual Navigation in Urban Environments 城市环境中第一人称行人视觉导航的端到端深度强化学习

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000201

Honghu Xue, Rui Song, Julian Petzold, Benedikt Hein, Heiko Hamann, Elmar Rueckert

引用次数: 1

Inverse Kinematics of Hybrid Multi-link System and its Application to Motion Capture for Athlete Wearing Sports Prosthesis 混合多连杆系统运动学逆解及其在运动员运动捕捉中的应用

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000195

Sung-Ho Kim, Taiki Ishigaki, Yuta Shimane, Y. Ikegami, Ko Yamamoto

引用次数: 1

Analysis of Effects of Leg Inertia Distribution on Contact Stability and Dynamic Performance During Bipedal Walking 腿惯性分布对两足行走接触稳定性和动态性能的影响分析

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000082

Junhyung Kim, Beomyeong Park, Jae-Wook Shin, Jaeheung Park

{"title":"Analysis of Effects of Leg Inertia Distribution on Contact Stability and Dynamic Performance During Bipedal Walking","authors":"Junhyung Kim, Beomyeong Park, Jae-Wook Shin, Jaeheung Park","doi":"10.1109/Humanoids53995.2022.10000082","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000082","url":null,"abstract":"This paper analyzes the effects of leg inertia distribution on contact stability and dynamic performance during bipedal walking. Recently, to ensure that the leg inertia distribution is mainly concentrated on the hip, biped robots have been developed with legs designed based on the design principle of placing the leg actuators near the pelvis. It is often mentioned that the dynamic performance of biped robots is improved by concentrating leg inertia distribution on the hip. However, it is not clear which aspect of the dynamic performance is improved and how it affects contact stability during walking. In this study, to analyze the improved dynamic performance and effects on the contact stability by concentrating the leg inertia distribution on the hip, the contact condition margins and maximum walking velocity for each model with various leg inertia distributions are calculated using the estimated contact wrench in walking simulations. The Linear Inverted Pendulum Model and the centroidal dynamics with full-body kinematics model, which have different degrees of model accuracy, are used as walking pattern generation models for walking simulations. As a result, it is confirmed that the contact stability and dynamic performance are improved as the leg inertia distribution is concentrated on the hip from the simulation results.","PeriodicalId":180816,"journal":{"name":"2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)","volume":"35 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":"131363058","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

A Dexterous, Adaptive, Affordable, Humanlike Robot Hand: Towards Prostheses with Dexterous Manipulation Capabilities 一个灵巧的，自适应的，负担得起的，类人的机器人手:迈向具有灵巧操作能力的假肢

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000237

Jayden Chapman, Anany Dwivedi, Minas Liarokapis

{"title":"A Dexterous, Adaptive, Affordable, Humanlike Robot Hand: Towards Prostheses with Dexterous Manipulation Capabilities","authors":"Jayden Chapman, Anany Dwivedi, Minas Liarokapis","doi":"10.1109/Humanoids53995.2022.10000237","DOIUrl":"https://doi.org/10.1109/Humanoids53995.2022.10000237","url":null,"abstract":"In this work, we present a 16 degree of freedom (DoF) humanlike, adaptive robotic hand specifically designed to achieve increased dexterity and the execution of dexterous manipulation tasks with minimal cost. Three Dynamixel XM430-W350-R actuate the hand via a combination of direct driven and tendon driven systems. The unique, coupled, two degree-of-freedom thumb mechanism is capable of achieving a complex trajectory for the thumb, providing increased thumb dexterity over traditional underactuated designs. This can facilitate the execution of improved in-hand manipulation tasks while minimising the control effort required for operation. The effectiveness of the proposed device was validated through kinematic, grasping, and manipulation experiments, including the Kapandji test, as well as force exertion experiments. It was found that the hand achieved a Kapandji score of 9 out of 11, and peak contact forces of 19.43 N for pinch grasps and 35.24 N for power grasps. The device was manufactured using the Selective Laser Sintering (SLS) process with nylon and molding using silicone rubber. The device has a weight of 603 g and costs approximately $1,000 USD to manufacture.","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":"114696300","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

Design of a Five-Fingered Hand with Full-Fingered Tactile Sensors Using Conductive Filaments and Its Application to Bending after Insertion Motion 基于导电丝的五指全指触觉传感器设计及其在插入后弯曲运动中的应用

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000181

K. Miyama, Shun Hasegawa, Kento Kawaharazuka, N. Yamaguchi, K. Okada, M. Inaba

引用次数: 0

Design of Robot Foot with Outer Edge Measurement Structure and Chair Rotation Motion by Friction Control 机器人脚外缘测量结构及椅子摩擦控制旋转运动的设计

2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids) Pub Date : 2022-11-28 DOI: 10.1109/Humanoids53995.2022.10000127

Yoshimoto Ribayashi, Kento Kawaharazuka, Yasunori Toshimitsu, Daiki Kusuyama, Akihiro Miki, Koki Shinjo, Masahiro Bando, Temma Suzuki, Yuta Kojio, K. Okada, M. Inaba

引用次数: 0