Learning Transferability: A Two-Stage Reinforcement Learning Approach for Enhancing Quadruped Robots' Performance in U-Shaped Stair Climbing

• URL: http://arxiv.org/abs/2602.14473v1
• Date: Mon, 16 Feb 2026 05:19:06 GMT
• Title: Learning Transferability: A Two-Stage Reinforcement Learning Approach for Enhancing Quadruped Robots' Performance in U-Shaped Stair Climbing
• Authors: Baixiao Huang, Baiyu Huang, Yu Hou,
• Abstract summary: We employ a two-stage end-to-end deep reinforcement learning approach to optimize a robot's performance on U-shaped stairs.<n>The training robot-dog modality, Unitree Go2, was first trained to climb stairs on Isaac Lab's pyramid-stair terrain.<n>The results showed (1) the successful goal reached for robot dogs climbing U-shaped stairs with a stall penalty, and (2) the transferability from the policy trained on U-shaped stairs to deployment on straight, L-shaped, and spiral stair terrains.
• Score: 4.901516178319544
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Quadruped robots are employed in various scenarios in building construction. However, autonomous stair climbing across different indoor staircases remains a major challenge for robot dogs to complete building construction tasks. In this project, we employed a two-stage end-to-end deep reinforcement learning (RL) approach to optimize a robot's performance on U-shaped stairs. The training robot-dog modality, Unitree Go2, was first trained to climb stairs on Isaac Lab's pyramid-stair terrain, and then to climb a U-shaped indoor staircase using the learned policies. This project explores end-to-end RL methods that enable robot dogs to autonomously climb stairs. The results showed (1) the successful goal reached for robot dogs climbing U-shaped stairs with a stall penalty, and (2) the transferability from the policy trained on U-shaped stairs to deployment on straight, L-shaped, and spiral stair terrains, and transferability from other stair models to deployment on U-shaped terrain.

Related papers

• Training and Simulation of Quadrupedal Robot in Adaptive Stair Climbing for Indoor Firefighting: An End-to-End Reinforcement Learning Approach [4.901516178319544]
  Quadruped robots are used for primary searches during the early stages of indoor fires.<n> situational awareness in complex indoor environments and rapid stair climbing across different staircases remain the main challenges.<n>This project explores how to balance navigation and locomotion and how end-to-end RL methods can enable quadrupeds to adapt to different stair shapes.
  arXiv  Detail & Related papers  (2026-02-03T04:23:50Z)
• Training Directional Locomotion for Quadrupedal Low-Cost Robotic Systems via Deep Reinforcement Learning [4.669957449088593]
  We present Deep Reinforcement Learning training of directional locomotion for low-cost quadpedalru robots in the real world.<n>We exploit randomization of heading that the robot must follow to foster exploration of action-state transitions.<n>Changing the heading in episode resets to current yaw plus a random value drawn from a normal distribution yields policies able to follow complex trajectories.
  arXiv  Detail & Related papers  (2025-03-14T03:53:01Z)
• BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholds [35.62230804783507]
  Existing learning-based approaches often struggle on complex terrains due to sparse foothold rewards and inefficient learning processes.<n>We introduce BeamDojo, a reinforcement learning framework designed for enabling agile humanoid locomotion on sparse footholds.<n>We show that BeamDojo achieves efficient learning in simulation and enables agile locomotion with precise foot placement on sparse footholds in the real world.
  arXiv  Detail & Related papers  (2025-02-14T18:42:42Z)
• Learning Humanoid Locomotion over Challenging Terrain [84.35038297708485]
  We present a learning-based approach for blind humanoid locomotion capable of traversing challenging natural and man-made terrains. Our model is first pre-trained on a dataset of flat-ground trajectories with sequence modeling, and then fine-tuned on uneven terrain using reinforcement learning. We evaluate our model on a real humanoid robot across a variety of terrains, including rough, deformable, and sloped surfaces.
  arXiv  Detail & Related papers  (2024-10-04T17:57:09Z)
• Reinforcement Learning for Versatile, Dynamic, and Robust Bipedal Locomotion Control [106.32794844077534]
  This paper presents a study on using deep reinforcement learning to create dynamic locomotion controllers for bipedal robots. We develop a general control solution that can be used for a range of dynamic bipedal skills, from periodic walking and running to aperiodic jumping and standing. This work pushes the limits of agility for bipedal robots through extensive real-world experiments.
  arXiv  Detail & Related papers  (2024-01-30T10:48:43Z)
• Barkour: Benchmarking Animal-level Agility with Quadruped Robots [70.97471756305463]
  We introduce the Barkour benchmark, an obstacle course to quantify agility for legged robots. Inspired by dog agility competitions, it consists of diverse obstacles and a time based scoring mechanism. We present two methods for tackling the benchmark.
  arXiv  Detail & Related papers  (2023-05-24T02:49:43Z)
• Robust and Versatile Bipedal Jumping Control through Reinforcement Learning [141.56016556936865]
  This work aims to push the limits of agility for bipedal robots by enabling a torque-controlled bipedal robot to perform robust and versatile dynamic jumps in the real world. We present a reinforcement learning framework for training a robot to accomplish a large variety of jumping tasks, such as jumping to different locations and directions. We develop a new policy structure that encodes the robot's long-term input/output (I/O) history while also providing direct access to a short-term I/O history.
  arXiv  Detail & Related papers  (2023-02-19T01:06:09Z)
• Legged Locomotion in Challenging Terrains using Egocentric Vision [70.37554680771322]
  We present the first end-to-end locomotion system capable of traversing stairs, curbs, stepping stones, and gaps. We show this result on a medium-sized quadruped robot using a single front-facing depth camera.
  arXiv  Detail & Related papers  (2022-11-14T18:59:58Z)
• Learning Bipedal Walking On Planned Footsteps For Humanoid Robots [5.127310126394387]
  Deep reinforcement learning (RL) based controllers for legged robots have demonstrated impressive robustness for walking in different environments for several robot platforms. To enable the application of RL policies for humanoid robots in real-world settings, it is crucial to build a system that can achieve robust walking in any direction. In this paper, we tackle this problem by learning a policy to follow a given step sequence. We show that simply feeding the upcoming 2 steps to the policy is sufficient to achieve omnidirectional walking, turning in place, standing, and climbing stairs.
  arXiv  Detail & Related papers  (2022-07-26T04:16:00Z)
• Adaptation of Quadruped Robot Locomotion with Meta-Learning [64.71260357476602]
  We demonstrate that meta-reinforcement learning can be used to successfully train a robot capable to solve a wide range of locomotion tasks. The performance of the meta-trained robot is similar to that of a robot that is trained on a single task.
  arXiv  Detail & Related papers  (2021-07-08T10:37:18Z)
• LBGP: Learning Based Goal Planning for Autonomous Following in Front [16.13120109400351]
  This paper investigates a hybrid solution which combines deep reinforcement learning (RL) and classical trajectory planning. An autonomous robot aims to stay ahead of a person as the person freely walks around. Our system outperforms the state-of-the-art in following ahead and is more reliable compared to end-to-end alternatives in both the simulation and real world experiments.
  arXiv  Detail & Related papers  (2020-11-05T22:29:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.