Realizing Learned Quadruped Locomotion Behaviors through Kinematic Motion Primitives

@article{Singla2019RealizingLQ,
  title={Realizing Learned Quadruped Locomotion Behaviors through Kinematic Motion Primitives},
  author={Abhik Singla and Shounak Bhattacharya and Dhaivat Dholakiya and Shalabh Bhatnagar and Ashitava Ghosal and Bharadwaj S. Amrutur and Shishir Kolathaya},
  journal={2019 International Conference on Robotics and Automation (ICRA)},
  year={2019},
  pages={7434-7440}
}
  • Abhik Singla, Shounak Bhattacharya, +4 authors Shishir Kolathaya
  • Published 2019
  • Engineering, Computer Science
  • 2019 International Conference on Robotics and Automation (ICRA)
  • Humans and animals are believed to use a very minimal set of trajectories to perform a wide variety of tasks including walking. Our main objective in this paper is two fold 1) Obtain an effective tool to realize these basic motion patterns for quadrupedal walking, called the kinematic motion primitives (kMPs), via trajectories learned from deep reinforcement learning (D-RL) and 2) Realize a set of behaviors, namely trot, walk, gallop and bound from these kinematic motion primitives in our… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-8 OF 8 CITATIONS

    Gait Library Synthesis for Quadruped Robots via Augmented Random Search

    VIEW 3 EXCERPTS
    CITES RESULTS & BACKGROUND

    Trajectory based Deep Policy Search for Quadrupedal Walking

    VIEW 8 EXCERPTS

    Learning Active Spine Behaviors for Dynamic and Efficient Locomotion in Quadruped Robots

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS

    Learning a Faster Locomotion Gait for a Quadruped Robot with Model-Free Deep Reinforcement Learning

    • Biao Hu, Shibo Shao, +3 authors Chao Ma
    • Engineering, Computer Science
    • 2019 IEEE International Conference on Robotics and Biomimetics (ROBIO)
    • 2019
    VIEW 1 EXCERPT
    CITES BACKGROUND

    Design, Development and Experimental Realization of A Quadrupedal Research Platform: Stoch

    VIEW 6 EXCERPTS
    CITES RESULTS, METHODS & BACKGROUND

    Rapidly Adaptable Legged Robots via Evolutionary Meta-Learning

    VIEW 1 EXCERPT
    CITES METHODS

    Reinforcement Learning Algorithms for Autonomous Adaptive Agents

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 39 REFERENCES

    On the Kinematic Motion Primitives (kMPs) – Theory and Application

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Sim-to-Real: Learning Agile Locomotion For Quadruped Robots

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Policy gradient reinforcement learning for fast quadrupedal locomotion

    • Nate Kohl, Peter Stone
    • Engineering, Computer Science
    • IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004
    • 2004
    VIEW 3 EXCERPTS

    Realizing dynamic and efficient bipedal locomotion on the humanoid robot DURUS

    VIEW 1 EXCERPT

    Reinforcement Learning Control for Biped Robot Walking on Uneven Surfaces

    VIEW 2 EXCERPTS