mlsanityData to 14 Oct 2026

Training Directional Locomotion for Quadrupedal Low-Cost Robotic Systems via Deep Reinforcement Learning

arXiv 2503.11059 ↗First posted 14 Mar 2025cs.RO cs.AIcited by 0

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

In this work we present Deep Reinforcement Learning (DRL) training of directional locomotion for low-cost quadrupedal robots in the real world. In particular, we exploit randomization of heading that the robot must follow to foster exploration of action-state transitions most useful for learning both forward locomotion as well as course adjustments. 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 involving frequent turns in both directions as well as long straight-line stretches. By repeatedly changing the heading, this method keeps the robot moving within the training platform and thus reduces human involvement and need for manual resets during the training. Real world experiments on a custom-built, low-cost quadruped demonstrate the efficacy of our method with the robot successfully navigating all validation tests. When trained with other approaches, the robot only succeeds in forward locomotion test and fails when turning is required.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 17 of its 27 references that are in the map. First posted on arXiv 14 Mar 2025.

Builds on

17 of 27 references.

References · 17

  1. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
  2. 2015Continuous control with deep reinforcement learningcited by 3,407
  3. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  4. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  5. 2017Sim-to-Real Transfer of Robotic Control with Dynamics Randomizationcited by 672
  6. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 604
  7. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 512
  8. 2020Sim-to-Real Transfer in Deep Reinforcement Learning for Robotics: a Surveycited by 413
  9. 2022DayDreamer: World Models for Physical Robot Learningcited by 400
  10. 2018Sim-to-Real: Learning Agile Locomotion For Quadruped Robotscited by 319
  11. 2020Learning to Walk in the Real World with Minimal Human Effortcited by 111
  12. 2018GPU-Accelerated Robotic Simulation for Distributed Reinforcement Learningcited by 90
5 more
  1. 2022A Walk in the Park: Learning to Walk in 20 Minutes With Model-Free Reinforcement Learningcited by 83
  2. 2020Dynamics Randomization Revisited:A Case Study for Quadrupedal Locomotioncited by 59
  3. 2021Vision-aided Dynamic Quadrupedal Locomotion on Discrete Terrain using Motion Librariescited by 13
  4. 2020Adaptive Force-based Control for Legged Robotscited by 6
  5. 2021Verifying Safe Transitions between Dynamic Motion Primitives on Legged Robotscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Learning to Walk With Less: A Dyna-Style Approach to Quadrupedal Locomotion
  2. 2025Quadruped Robot Simulation Using Deep Reinforcement Learning -- A step towards locomotion policy
  3. 2023A Real-World Quadrupedal Locomotion Benchmark for Offline Reinforcement Learning
  4. 2024Reinforcement Learning For Quadrupedal Locomotion: Current Advancements And Future Perspectives
  5. 2025RANDPOL: Parameter-Efficient End-to-End Quadruped Locomotion via Randomized Policy Learning
  6. 2025PALo: Learning Posture-Aware Locomotion for Quadruped Robots
  7. 2024Learning Quadrupedal Locomotion with Impaired Joints Using Random Joint Masking
  8. 2024Offline Adaptation of Quadruped Locomotion using Diffusion Models