Pith. sign in

REVIEW 31 cited by

Robot Parkour Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2309.05665 v2 pith:DK2DA6WN submitted 2023-09-11 cs.RO cs.AIcs.CVcs.LG

classification cs.ROcs.AIcs.CVcs.LG
keywords parkourskillsvision-baseddiverselearningrobotscomplexdata
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Parkour is a grand challenge for legged locomotion that requires robots to overcome various obstacles rapidly in complex environments. Existing methods can generate either diverse but blind locomotion skills or vision-based but specialized skills by using reference animal data or complex rewards. However, autonomous parkour requires robots to learn generalizable skills that are both vision-based and diverse to perceive and react to various scenarios. In this work, we propose a system for learning a single end-to-end vision-based parkour policy of diverse parkour skills using a simple reward without any reference motion data. We develop a reinforcement learning method inspired by direct collocation to generate parkour skills, including climbing over high obstacles, leaping over large gaps, crawling beneath low barriers, squeezing through thin slits, and running. We distill these skills into a single vision-based parkour policy and transfer it to a quadrupedal robot using its egocentric depth camera. We demonstrate that our system can empower two different low-cost robots to autonomously select and execute appropriate parkour skills to traverse challenging real-world environments.

Discussion (0). Sign in to comment.

Forward citations

Cited by 31 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Light-Loco-Parkour: Versatile Perceptive Whole-Body Locomotion via Multi-Skill Distillation

    cs.RO 2026-08 conditional novelty 7.0 of 10

    A single neural-network policy, trained in simulation, makes a humanoid climb, vault, and traverse uneven terrain from onboard depth and a velocity command, with no skill labels or runtime motion graphs.

  2. Transformer Transformer: A Unified Model for Motion-Conditioned Robot Co-design

    cs.RO 2026-07 conditional novelty 7.0 of 10

    A single diffusion transformer trains on tokenized robot bodies and motions to generate and optimize robot designs for unseen rewards and trajectories, outpacing evolutionary search in speed and often in reward.

  3. StairMaster: Learning to Conquer Risky Hollow Stairs for Agile Quadrupedal Robots

    cs.RO 2026-06 conditional novelty 6.0 of 10

    A three-stage RL framework with cross-attention, spatial memory, and depth-noise augmentation lets a Unitree Go2 climb 55° hollow stairs zero-shot from simulation.

  4. StairMaster: Learning to Conquer Risky Hollow Stairs for Agile Quadrupedal Robots

    cs.RO 2026-06 unverdicted novelty 6.0 of 10

    StairMaster trains an RL policy that lets a Unitree Go2 quadruped climb hollow stairs up to 55 degrees via zero-shot sim-to-real transfer using cross-attention, SRU memory, and active-perception rewards.

  5. SWAP: Symmetric Equivariant World-Model for Agile Robot Parkour

    cs.RO 2026-06 unverdicted novelty 6.0 of 10

    SWAP embeds symmetry equivariance into world models and policies, enabling a quadruped to leap 2.13m gaps and climb 1.63m platforms with robust generalization to mirrored and outdoor terrains.

  6. GS-Playground: A High-Throughput Photorealistic Simulator for Vision-Informed Robot Learning

    cs.RO 2026-04 unverdicted novelty 6.0 of 10

    GS-Playground delivers a high-throughput photorealistic simulator for vision-informed robot learning via parallel physics integrated with batch 3D Gaussian Splatting at 10^4 FPS and an automated Real2Sim workflow for ...

  7. Learning Task-Invariant Properties via Dreamer: Enabling Efficient Policy Transfer for Quadruped Robots

    cs.RO 2026-04 unverdicted novelty 6.0 of 10

    DreamTIP adds LLM-identified task-invariant properties as auxiliary targets in Dreamer's world model plus a mixed-replay adaptation step, delivering 28.1% average simulated transfer gains and 100% real-world climb suc...

  8. PTLD: Sim-to-real Privileged Tactile Latent Distillation for Dexterous Manipulation

    cs.RO 2026-03 conditional novelty 6.0 of 10

    A sim-to-real method that distills a privileged camera-based teacher policy into a tactile student policy, improving in-hand rotation and reorientation over proprioception-only policies.

  9. PTLD: Sim-to-real Privileged Tactile Latent Distillation for Dexterous Manipulation

    cs.RO 2026-03 unverdicted novelty 6.0 of 10

    PTLD distills real privileged tactile data into a state estimator to boost sim-to-real performance of proprioceptive dexterous manipulation policies, yielding 182% improvement on in-hand rotation and 57% on reorientat...

  10. Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching

    cs.RO 2026-02 unverdicted novelty 6.0 of 10

    A modular system uses motion matching to compose long-horizon human skill chains, trains RL experts, and distills them into a depth-based policy that lets a Unitree G1 humanoid autonomously climb, vault, and roll over...

  11. Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

    cs.RO 2025-11 unverdicted novelty 6.0 of 10

    MSDP pretrains a transformer encoder via masked multisensory reconstruction and feeds the embeddings into an asymmetric actor-critic RL setup, yielding faster learning and high real-robot success rates with only 6,000...

  12. Sampling Strategies for Robust Universal Quadrupedal Locomotion Policies

    cs.RO 2025-10 conditional novelty 6.0 of 10

    A particle-filter-based adaptive sampling of morphologies and wide PD-gain randomization yields a single quadruped locomotion policy that transfers zero-shot to ANYmal hardware.

  13. Benchmarking Massively Parallelized Multi-Task Reinforcement Learning for Robotics Tasks

    cs.RO 2025-07 conditional novelty 6.0 of 10

    The paper introduces MTBench, a GPU-accelerated benchmark for massively parallel multi-task RL, and reports experiments suggesting on-policy methods outperform off-policy baselines while value learning limits MTRL per...

  14. DreamPolicy: A Unified World-model Policy for Scalable Humanoid Locomotion

    cs.RO 2025-05 unverdicted novelty 6.0 of 10

    DreamPolicy integrates an autoregressive diffusion world model with policy learning to produce a single scalable policy that generalizes to unseen composite terrains for humanoid locomotion.

  15. RAVEN: Reinforcement-Adaptive Visibility-Graph Planning for Robust Humanoid Navigation with Collision-Free MPC

    cs.RO 2026-07 conditional novelty 5.0 of 10

    Reinforcement learning that adjusts obstacle-inflation radii in a visibility-graph planner improves humanoid navigation robustness under delay and noise, beating a static MPC baseline and an end-to-end RL policy in a ...

  16. Mind Your Steps: A General Learning Framework for Accurate Humanoid Foothold Tracking

    cs.RO 2026-06 unverdicted novelty 5.0 of 10

    A lightweight RL framework trains terrain-agnostic 3D foothold-tracking policies for humanoids that transfer directly to real-world use as standalone low-level controllers.

  17. LadderMan: Learning Humanoid Perceptive Ladder Climbing

    cs.RO 2026-06 unverdicted novelty 5.0 of 10

    A hybrid motion-tracking and imitation-reinforcement pipeline produces a depth-based visuomotor policy that lets humanoids climb varied ladders zero-shot on hardware and perform teleoperated manipulation while climbing.

  18. RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting

    cs.RO 2026-04 unverdicted novelty 5.0 of 10

    RPG trains a single policy with transition and timing randomization for stable multi-skill fighting on humanoids, integrated with locomotion for arbitrary-duration combat.

  19. Now You See That: Learning End-to-End Humanoid Locomotion from Raw Pixels

    cs.RO 2026-02 unverdicted novelty 5.0 of 10

    An end-to-end policy learns robust humanoid locomotion directly from noisy depth images via high-fidelity sensor simulation, vision-aware distillation from privileged maps, and terrain-specific multi-critic reward shaping.

  20. Coupled Local and Global World Models for Efficient First Order RL

    cs.RO 2026-02 conditional novelty 5.0 of 10

    Coupled local/global world models let first-order RL train image-space robot policies inside a learned diffusion simulator, outperforming PPO and a DreamerV3-only ablation on two tasks.

  21. UniCon: A Unified System for Efficient Robot Learning Transfers

    cs.RO 2026-01 unverdicted novelty 5.0 of 10

    UniCon standardizes states and control logic into modular execution graphs for efficient transfer of learning controllers across heterogeneous robots, with lower latency than ROS.

  22. Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input

    cs.RO 2025-12 conditional novelty 5.0 of 10

    A four-stage RL system with teacher-student distillation and online constrained adaptation enables humanoid robots to achieve robust ball-kicking accuracy under noisy perception in simulation and on physical hardware.

  23. Self-Supervised Multisensory Pretraining for Contact-Rich Robot Reinforcement Learning

    cs.RO 2025-11 conditional novelty 5.0 of 10

    MSDP pre-trains a transformer encoder with masked multisensory autoencoding, then uses an asymmetric actor-critic bridge (cross-attention for critic, pooling for actor) to accelerate and robustify contact-rich RL acro...

  24. DPL: Depth-only Perceptive Humanoid Locomotion via Realistic Depth Synthesis and Cross-Attention Terrain Reconstruction

    cs.RO 2025-10 conditional novelty 5.0 of 10

    Combining a blind-backbone policy, cross-attention terrain reconstruction from depth plus proprioception, and realistic synthetic depth with noise enables depth-only full-sized humanoid locomotion over stairs, slopes,...

  25. First Order Model-Based RL through Decoupled Backpropagation

    cs.RO 2025-08 conditional novelty 5.0 of 10

    By computing gradients through a learned dynamics model while unrolling trajectories in the real simulator, DMO achieves SHAC-level sample efficiency with standard simulators and deploys on a real quadruped.

  26. End-to-End Humanoid Robot Safe and Comfortable Locomotion Policy

    cs.RO 2025-08 conditional novelty 5.0 of 10

    An end-to-end humanoid locomotion policy maps raw LiDAR point clouds to motor commands using P3O with CBF-inspired safety costs and comfort rewards, with sim-to-real tests on a Unitree G1.

  27. WARL: Wrench-Augmented Reinforcement Learning for Task-Agnostic Learning in Legged Robots

    cs.RO 2026-07 conditional novelty 4.0 of 10

    Adding a simulated torso wrench during early RL training and gradually removing it lets a quadruped learn six locomotion tasks with a shared reward, yielding a joint-only policy in simulation.

  28. Learning Perceptive Platform Adaptive Locomotion Controllers for Quadrupedal Robots

    cs.RO 2026-06 unverdicted novelty 4.0 of 10

    Empirical comparison of blind, critic-perceptive, and fully perceptive variants of morphology-aware RL locomotion controllers shows critic-only perception improves robustness over blind baselines while remaining more ...

  29. RPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting

    cs.RO 2026-04 unverdicted novelty 4.0 of 10

    RPG trains a unified humanoid robot policy using motion and temporal randomization to achieve smooth, stable transitions between fighting skills and locomotion.

  30. Quadruped Parkour Learning: Sparsely Gated Mixture of Experts with Visual Input

    cs.RO 2026-04 unverdicted novelty 4.0 of 10

    Sparsely gated MoE policies double the success rate of a real Unitree Go2 quadruped on large-obstacle parkour versus matched-active-parameter MLP baselines while cutting inference time compared with a scaled-up MLP.

  31. Ground-Aware Octree-A* Hybrid Path Planning for Memory-Efficient 3D Navigation of Ground Vehicles

    cs.RO 2025-09 conditional novelty 3.0 of 10

    An A*-on-octree planner with a height penalty plans ground-hugging 3D routes for ground vehicles in two simulations using under a tenth of the memory and computation time of uniform-grid A*, at nearly equal path length.

Pith tools