Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:CDLA3MQOrecord.jsonopen to challenge →
read the original abstract
In this work, we present and study a training set-up that achieves fast policy generation for real-world robotic tasks by using massive parallelism on a single workstation GPU. We analyze and discuss the impact of different training algorithm components in the massively parallel regime on the final policy performance and training times. In addition, we present a novel game-inspired curriculum that is well suited for training with thousands of simulated robots in parallel. We evaluate the approach by training the quadrupedal robot ANYmal to walk on challenging terrain. The parallel approach allows training policies for flat terrain in under four minutes, and in twenty minutes for uneven terrain. This represents a speedup of multiple orders of magnitude compared to previous work. Finally, we transfer the policies to the real robot to validate the approach. We open-source our training code to help accelerate further research in the field of learned legged locomotion.
This paper has not been read by Pith yet.
Forward citations
Cited by 9 Pith papers
-
LeLaR: The First In-Orbit Demonstration of an AI-Based Satellite Attitude Controller
First in-orbit demonstration of a DRL-trained AI satellite attitude controller that performs robust inertial pointing after sim-to-real transfer.
-
Unleashing Infinite Motion: Scaling Expressive Quadrupedal Motion via Generative Video Priors
Uni-Mo generates 7,488 language-annotated quadruped motions via LLM prompts and video diffusion, lifts them to 3D trajectories, and trains policies achieving 96.7% real-robot success on 392 sampled motions.
-
Real-IKEA: Physical Fidelity is the Prerequisite for Robust Manipulation
Real-IKEA supplies 1,079 physically accurate articulated asset configurations from real IKEA parts together with resistance-calibrated simulation parameters that enable RL policies to discover robust hooking and lever...
-
Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters
A hybrid neural policy operating in impulse space enables physics-based characters to track exaggerated, dynamically infeasible motions that standard DRL methods cannot stabilize.
-
Do as I Do: Dexterous Manipulation Data from Everyday Human Videos
DO AS I DO reconstructs and retargets hand-object interactions from in-the-wild monocular RGB videos to produce dexterous robot manipulation trajectories, outperforming prior methods on ground-truth and online video datasets.
-
SigLoMa: Learning Open-World Quadrupedal Loco-Manipulation from Ego-Centric Vision
SigLoMa enables dynamic loco-manipulation on quadrupeds from ego-centric 5 Hz vision alone by using Sigma Points for scalable exteroception, an ego-centric Kalman Filter for high-rate state estimation, and an active s...
-
Learning to Act Through Contact: A Unified View of Multi-Task Robot Learning
A single goal-conditioned RL policy trained on contact plans performs multiple gaits and bimanual manipulation tasks on quadruped and humanoid robots.
-
Efficient On-policy Visual-RL via Stochastic Decoupled Policy Gradient
SDPG is a new on-policy visual RL algorithm that estimates gradients via stochastic perturbations of rollouts, achieving faster training and lower memory use than baselines on visual MuJoCo tasks while adding new robo...
-
Tracking the Effective Surface Area of Non-Convex Satellites
Backstepping control tracks effective surface area of non-convex satellites for drag-based orbital control, with asymptotic stability proofs and an extension for solar panel exposure.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.