11 papers
Scaling Rough Terrain Locomotion with Automatic Curriculum Reinforcement Learning
Ziming Li, Chenhao Li, Marco Hutter
Curriculum learning has demonstrated substantial effectiveness in robot learning. However, it still faces limitations when scaling to complex, wide-ranging task spaces. Such task s…
Multi-Domain Motion Embedding: Expressive Real-Time Mimicry for Legged Robots
Matthias Heyrman, Chenhao Li, Victor Klemm +3
Effective motion representation is crucial for enabling robots to imitate expressive behaviors in real time, yet existing motion controllers often ignore inherent patterns in motio…
Collaborative Loco-Manipulation for Pick-and-Place Tasks with Dynamic Reward Curriculum
Tianxu An, Flavio De Vincenti, Yuntao Ma +2
We present a hierarchical RL pipeline for training one-armed legged robots to perform pick-and-place (P&P) tasks end-to-end -- from approaching the payload to releasing it at a tar…
Feature-Based vs. GAN-Based Learning from Demonstrations: When and Why
Chenhao Li, Marco Hutter, Andreas Krause
This survey provides a comparative analysis of feature-based and GAN-based approaches to learning from demonstrations, with a focus on the structure of reward functions and their i…
Constrained Style Learning from Imperfect Demonstrations under Task Optimality
Kehan Wen, Chenhao Li, Junzhe He +1
Learning from demonstration has proven effective in robotics for acquiring natural behaviors, such as stylistic motions and lifelike agility, particularly when explicitly defining…
Parkour in the Wild: Learning a General and Extensible Agile Locomotion Policy Using Multi-expert Distillation and RL Fine-tuning
Nikita Rudin, Junzhe He, Joshua Aurand +1
Legged robots are well-suited for navigating terrains inaccessible to wheeled robots, making them ideal for applications in search and rescue or space exploration. However, current…