mlsanityData to 14 Oct 2026

OGMP: Oracle Guided Multi-mode Policies for Agile and Versatile Robot Control

arXiv 2403.04205 ↗First posted 7 Mar 2024cs.ROcited by 1

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

Abstract (arXiv)

The efficacy of reinforcement learning for robot control relies on the tailored integration of task-specific priors and heuristics for effective exploration, which challenges their straightforward application to complex tasks and necessitates a unified approach. In this work, we define a general class for priors called oracles that generate state references when queried in a closed-loop manner during training. By bounding the permissible state around the oracle's ansatz, we propose a task-agnostic oracle-guided policy optimization. To enhance modularity, we introduce task-vital modes, showing that a policy mastering a compact set of modes and transitions can handle infinite-horizon tasks. For instance, to perform parkour on an infinitely long track, the policy must learn to jump, leap, pace, and transition between these modes effectively. We validate this approach in challenging bipedal control tasks: parkour and diving using a 16 DoF dynamic bipedal robot, HECTOR. Our method results in a single policy per task, solving parkour across diverse tracks and omnidirectional diving from varied heights up to 2m in simulation, showcasing versatile agility. We demonstrate successful sim-to-real transfer of parkour, including leaping over gaps up to 105 % of the leg length, jumping over blocks up to 20 % of the robot's nominal height, and pacing at speeds of up to 0.6 m/s, along with effective transitions between these modes in the real robot.

Key facts. Cited by 1 paper in the map (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 31 references that are in the map. First posted on arXiv 7 Mar 2024.

Builds on

17 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Reinforcement Learning for Versatile, Dynamic, and Robust Bipedal Locomotion Control30 Jan

Other references · 16

  1. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  2. 2022Learning robust perceptive locomotion for quadrupedal robots in the wildcited by 382
  3. 2023Extreme Parkour with Legged Robotscited by 164
  4. 2023ANYmal Parkour: Learning Agile Navigation for Quadrupedal Robotscited by 152
  5. 2023Robot Parkour Learningcited by 149
  6. 2023Learning Agile Soccer Skills for a Bipedal Robot with Deep Reinforcement Learningcited by 133
  7. 2021Blind Bipedal Stair Traversal via Sim-to-Real Reinforcement Learningcited by 92
  8. 2023DTC: Deep Tracking Controlcited by 56
  9. 2022Advanced Skills by Learning Locomotion and Local Navigation End-to-Endcited by 52
  10. 2019Control Regularization for Reduced Variance Reinforcement Learningcited by 37
  11. 2022Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learningcited by 37
  12. 2019MPC-Net: A First Principles Guided Policy Searchcited by 34
4 more
  1. 2022Imitate and Repurpose: Learning Reusable Robot Movement Skills From Human and Animal Behaviorscited by 30
  2. 2020Guided Constrained Policy Optimization for Dynamic Quadrupedal Robot Locomotioncited by 25
  3. 2021Force-and-moment-based Model Predictive Control for Achieving Highly Dynamic Locomotion on Bipedal Robotscited by 4
  4. 2023Learning Multimodal Bipedal Locomotion and Implicit Transitions: A Versatile Policy Approachcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Accelerating and Scaling MPC-Guided Reinforcement Learning for Humanoid Locomotion and Manipulationcited by 3

Similar, not cited

Close in wording, no citation either way

  1. 2025Jump-Start Reinforcement Learning with Self-Evolving Priors for Extreme Monopedal Locomotion
  2. 2024SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experience
  3. 2026Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching
  4. 2026Actuator Dynamics Curricula for Narrow-Viability Tasks in Legged Robot Learning
  5. 2026BAT: Balancing Agility and Stability via Online Policy Switching for Long-Horizon Whole-Body Humanoid Control
  6. 2026AGILE: A Comprehensive Workflow for Humanoid Loco-Manipulation Learning
  7. 2025Parkour in the Wild: Learning a General and Extensible Agile Locomotion Policy Using Multi-expert Distillation and RL Fine-tuning
  8. 2025APEX: Action Priors Enable Efficient Exploration for Robust Motion Tracking on Legged Robots ↗