☆ 4.6 Article

Learning Bipedal Walking for Humanoids With Current Feedback

IEEE ACCESS (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Proceedings Paper Computer Science, Interdisciplinary Applications

mc-mujoco: Simulating Articulated Robots with FSM Controllers in MuJoCo

Rohan P. Singh et al.

Summary: To ensure the safe and reliable deployment of robot controllers on real hardware platforms, it is necessary to evaluate their performance in a realistic simulation environment beforehand. However, interfacing simulation environments with robot controllers can be complex and error-prone. This study introduces mc-mujoco, an open-source software framework that acts as an interface between the MuJoCo physics simulator and the mc-rtc robot control framework, allowing for the addition of support for new robots.

2023 IEEE/SICE INTERNATIONAL SYMPOSIUM ON SYSTEM INTEGRATION, SII (2023)

添加到收藏夹

Article Computer Science, Information Systems

Reinforcement Learning-Based Cascade Motion Policy Design for Robust 3D Bipedal Locomotion

Guillermo A. Castillo et al.

Summary: This paper presents a novel reinforcement learning framework for designing cascade feedback control policies for 3D bipedal locomotion. The proposed solution decouples the locomotion problem into two modules, incorporating physical insights from walking dynamics and the Hybrid Zero Dynamics approach. The framework offers advantages such as lightweight network structure, sample efficiency, and reduced dependence on prior knowledge.

IEEE ACCESS (2022)

添加到收藏夹

Article Robotics

Centroidal Trajectory Generation and Stabilization Based on Preview Control for Humanoid Multi-Contact Motion

Masaki Murooka et al.

Summary: This study proposes a trajectory generation and stabilization control method for humanoid dynamic multi-contact motion, which combines preview control, centroidal state feedback, and wrench distribution strategy to achieve stable multi-contact motion.

IEEE ROBOTICS AND AUTOMATION LETTERS (2022)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Adapting Rapid Motor Adaptation for Bipedal Robots

Ashish Kumar et al.

Summary: Recent advances in legged locomotion have focused on quadrupeds, but designing walking controllers for bipedal robots is more challenging due to their inherent instability. In this work, the researchers leverage rapid adaptation techniques and propose A-RMA, which adapts the base policy of a bipedal robot for imperfect extrinsics estimation using model-free reinforcement learning. The results show that A-RMA outperforms other controllers in simulation and enables a bipedal robot to walk in real-world scenarios it hasn't encountered before.

2022 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) (2022)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Sim-to-Real Learning of All Common Bipedal Gaits via Periodic Reward Composition

Jonah Siekmann et al.

Summary: By defining a parameterized reward function using simple probabilistic periodic costs, we successfully achieved the transfer of various gaits of bipedal robots from simulation to reality. This method can assist in learning different gaits and is applicable to common bipedal locomotion.

2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021) (2021)

添加到收藏夹

Proceedings Paper Automation & Control Systems

DeepWalk: Omnidirectional Bipedal Gait by Deep Reinforcement Learning

Diego Rodriguez et al.

Summary: This paper presents a novel DRL approach for enabling robots to learn omnidirectional locomotion for humanoid robots by gradually increasing task difficulty. The method does not require reference motions, reducing overall complexity, and provides strategies for sim-to-real transfer.

2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021) (2021)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Learning Linear Policies for Robust Bipedal Locomotion on Terrains with Varying Slopes

Lokesh Krishna et al.

Summary: This paper presents a robust and fast feedback control law for bipedal walking on terrains with varying slopes, achieved through a single linear feedback policy learned via a model-free and gradient-free learning algorithm. Additional behaviors such as walking backwards, stepping-in-place, and recovery from external pushes are also demonstrated. The preliminary results of hardware transfer to Digit platform are provided at the end.

2021 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) (2021)

添加到收藏夹

Article Robotics

Learning quadrupedal locomotion over challenging terrain

Joonho Lee et al.

SCIENCE ROBOTICS (2020)

添加到收藏夹

Article Robotics

Multi-expert learning of adaptive legged locomotion

Chuanyu Yang et al.

SCIENCE ROBOTICS (2020)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Reliable chattering-free simulation of friction torque in joints presenting high stiction

Rafael Cisneros et al.

2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) (2020)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Sim-to-Real Transfer for Biped Locomotion

Wenhao Yu et al.

2019 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) (2019)

添加到收藏夹

Article Robotics

Learning agile and dynamic motor skills for legged robots

Jemin Hwangbo et al.

SCIENCE ROBOTICS (2019)

添加到收藏夹

Proceedings Paper Automation & Control Systems

Stair Climbing Stabilization of the HRP-4 Humanoid Robot using Whole-body Admittance Control

Stephane Caron et al.

2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA) (2019)

添加到收藏夹

Article Robotics

Humanoid Robot HRP-5P: An Electrically Actuated Humanoid Robot With High-Power and Wide-Range Joints

Kenji Kaneko et al.

IEEE ROBOTICS AND AUTOMATION LETTERS (2019)

添加到收藏夹

Article Robotics

Three-Dimensional Bipedal Walking Control Based on Divergent Component of Motion

Johannes Englsberger et al.

IEEE TRANSACTIONS ON ROBOTICS (2015)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.