WARL: Wrench-Augmented Reinforcement Learning (IROS 2026)

Loading video

WARL (Wrench-Augmented Reinforcement Learning) accepted to IROS 2026 augments joint actions with virtual wrenches (forces and torques) to improve exploration during learning for legged robots.

强化学习 RL 腿式机器人 IROS2026 Twitter

Category: reinforcement-learning

Author: @kyoneda777

Date: 2026-07-28T00:00:00

Duration: 30.789s

Source: https://x.com/kyoneda777/status/2082024528155418983

REGRIND: Dexterous RL from Human Retargeting Initialization

REGRIND: Dexterous RL from Human Retargeting Initialization

@robotsdigest · 6 views · 2026-08-24

灵巧操作Dexterous ManipulationRL

Physics-Based Character Controllers with RL

Physics-Based Character Controllers with RL

@neural_avb · 1 views · 2026-08-09

强化学习物理角色控制RL

PAC-MAN: Humanoid Dodges Balls for Whole-Body Safety

PAC-MAN: Humanoid Dodges Balls for Whole-Body Safety

@yokoshaan · 4 views · 2026-08-02

人形机器人安全强化学习

RL Framework Lets Robots Autonomously Choose Gait on Uneven Terrain

RL Framework Lets Robots Autonomously Choose Gait on Uneven Terrain

@ScienceMagazine · 6 views · 2026-07-20

强化学习RL步态

FlexionAI: RGB-Based RL Policy Training in Simulation

FlexionAI: RGB-Based RL Policy Training in Simulation

@FlexionAI · 2 views · 2026-07-20

强化学习RLRGB

CMU open-sources the HTD whole-body controller for humanoids — playable in your browser

CMU open-sources the HTD whole-body controller for humanoids — playable in your browser

@yaru_niu · 14 views · 2026-08-25

人形机器人全身控制Isaac Lab