This work provides a complete framework for the simulation, co-optimization, and sim-to-real
transfer of the design and control of soft legged robots. The compliance of soft robots …

The automatic design of soft robots characterizes as jointly optimizing structure and control.
As reinforcement learning is gradually used to optimize control, the time-consuming …

Modular robots are made up of a set of components which can be configured and
reconfigured to form customized robots for a wide range of tasks. Fully utilizing the flexibility …

Learning a locomotion policy for quadruped robots has traditionally been constrained to
specific robot morphology, mass, and size. The learning process must usually be repeated …

A popular paradigm in robotic learning is to train a policy from scratch for every new robot.
This is not only inefficient but also often impractical for complex robots. In this work, we …

Biological and artificial agents need to deal with constant changes in the real world. We
study this problem in four classical continuous control environments, augmented with …

Reinforcement learning (RL) has achieved remarkable success in complex robotic systems
(eg. quadruped locomotion). In previous works, the RL-based controller was typically …

具身智能强调智能受脑, 身体与环境协同影响, 更侧重关注智能体与环境的“交互”. 因此,
在具身智能的研究中, 智能体的物理形态与感知, 学习, 控制的关系起到至关重要的作用. 当前 …

Metarobotics aims to combine next generation wireless communication, multisense
immersion, and collective intelligence to provide a pervasive, itinerant, and noninvasive …

This work provides a complete framework for the simulation, co-optimization, and sim-to-real
transfer of the design and control of soft legged robots. Soft robots have “mechanical …