🤖 Awesome Loco-Manipulation

A curated list of resources on loco-manipulation — the joint problem of locomotion and manipulation for legged, humanoid, wheeled-legged, and mobile robots.

This list is organized by methodology (model-based control & planning → learning-based control → foundation models & high-level reasoning), with thematic sub-sections inside each. Entries are tagged by robot embodiment and annotated with the research group / institution. Contributions welcome — see Contributing.

📖 Two surveys anchor this taxonomy: Humanoid Locomotion and Manipulation (2501.02116, pipeline view) and A Survey of Behavior Foundation Models (2506.20487, foundation-model view).

---

🧭 Legend

Embodiment tags: 🧍 Humanoid · 🐾 Quadruped (+arm/leg) · 🛞 Wheeled-legged · 🚗 Mobile manipulator · 🦿 General legged

Other: ⭐ = seminal / highly-cited · 🏛️ research group in italics

---

📑 Table of Contents

• Surveys & Benchmarks
• 1. Model-Based Control & Planning
  • Whole-Body MPC
  • Multi-Contact Planning & Control
  • Trajectory Optimization & Contact-Implicit
  • Hierarchical Optimization for Heavy Objects
  • Force / Impedance / Interaction Control
  • Pattern Generation & Stabilization (Humanoid)
  • Wheeled-Legged & Quadruped-with-Arm
• 2. Learning-Based Control
  • Whole-Body RL Control
  • Motion Tracking & Retargeting
  • Visual / Perceptive Loco-Manipulation
  • Skill Blending / Hierarchical / Unified Policies
  • Teleoperation & Data Collection
  • Diffusion Policies
• 3. Foundation Models & High-Level Reasoning
  • Behavioral Foundation Models (BFM)
  • Vision-Language-Action Models (VLA)
  • VLM / LLM Planners
  • World Models / World-Action Models
• 🏛️ Research Groups & Lineages
• 🤖 Robot Models
• ✨ Contributing

---

📊 Surveys & Benchmarks

• ⭐ [2025 arXiv] Humanoid Locomotion and Manipulation: Current Progress and Challenges in Control, Planning, and Learning [paper] 🧍 Georgia Tech, USC, Stanford, NVIDIA, CMU, ...
• ⭐ [2025 TPAMI] A Survey of Behavior Foundation Model: Next-Generation Whole-Body Control System of Humanoid Robots [paper] [repo] 🧍 EIT Ningbo
• [2023 Front. Mech. Eng.] Legged Robots for Object Manipulation: A Review [paper] 🦿

---

1. ⚙️ Model-Based Control & Planning

Whole-Body MPC for Loco-Manipulation

• ⭐ [2021 RA-L/ICRA] A Unified MPC Framework for Whole-Body Dynamic Locomotion and Manipulation [paper] 🐾 ETH Zurich, RSL — canonical unified whole-body MPC on ALMA (ANYmal + arm).
• ⭐ [2019 RA-L] Whole-Body MPC for a Dynamically Stable Mobile Manipulator [paper] 🦿 ETH Zurich, RSL
• [2022 ICRA] A Collision-Free MPC for Whole-Body Dynamic Locomotion and Manipulation [paper] 🐾 ETH Zurich, RSL
• [2022 IROS] Articulated Object Interaction in Unknown Scenes with Whole-Body Mobile Manipulation [paper] 🐾 ETH Zurich RSL + NVIDIA
• [2024 RA-L] Tailoring Solution Accuracy for Fast Whole-Body MPC of Legged Robots [paper] 🧍 MIT Biomimetics — full-dynamics NMPC at 90 Hz on the MIT Humanoid.
• [2024 arXiv] Real-Time Whole-Body Control of Legged Robots with MPPI Control [paper] [project] 🦿 CMU, Robotic Exploration Lab
• [2025 RA-L] Whole-Body Inverse Dynamics MPC for Legged Loco-Manipulation [paper] 🐾 ETH Zurich, CRL — torque-level WB-MPC, 80 Hz on B2 + Z1.
• [2024 ICRA] Hierarchical Optimization-Based Control for Whole-Body Loco-Manipulation of Heavy Objects [paper] 🦿

Multi-Contact Planning & Control

• ⭐ [2023 Science Robotics] Versatile Multi-Contact Planning and Control for Legged Loco-Manipulation [paper] [journal] 🐾 ETH Zurich, RSL — lightly-guided TAMP auto-generating whole-body trajectories + contact schedules.
• [2024 CoRL] Guided Reinforcement Learning for Robust Multi-Contact Loco-Manipulation [paper] 🐾 ETH Zurich, RSL (Oral)
• [2022 arXiv] Multi-Contact MPC for Dynamic Loco-Manipulation on Humanoid Robots [paper] 🧍 USC, DRCL
• [2023 Humanoids] Whole-Body MPC for Highly Redundant Legged Manipulators (37-DoF CENTAURO) [paper] 🛞 IIT, HHCM

Trajectory Optimization & Contact-Implicit Methods

• ⭐ [2014 IJRR] A Direct Method for Trajectory Optimization of Rigid Bodies Through Contact [paper] 🦿 MIT — foundational contact-implicit trajectory optimization.
• ⭐ [2024 IJRR] Contact-Implicit MPC: Controlling Diverse Quadruped Motions Without Pre-Planned Contact Modes [paper] 🐾 KAIST
• [2021 IROS] Contact-Implicit Trajectory Optimization for Dynamic Object Manipulation [paper] 🐾 ETH Zurich, RSL
• [2025 T-RO] Inverse Dynamics Trajectory Optimization for Contact-Implicit MPC [paper] [project] [code] 🦿 Notre Dame / Toyota Research

Hierarchical Optimization for Heavy Objects

• ⭐ [2024 ICRA] Hierarchical Optimization-Based Control for Whole-Body Loco-Manipulation of Heavy Objects [paper] 🧍 USC, DRCL
• [2024 arXiv] Safety-Critical Motion Planning for Collaborative Legged Loco-Manipulation over Discrete Terrain [paper] 🦿 USC
• [2026 arXiv] Sumo: Dynamic and Generalizable Whole-Body Loco-Manipulation [paper] [project] [code] 🐾 RAI Institute + CMU — sample-based planner steering a whole-body policy (hybrid model-based).

Force / Impedance / Interaction Control

• ⭐ [2021 ICRA] Model Predictive Robot-Environment Interaction Control for Mobile Manipulation Tasks [paper] 🐾 ETH Zurich, RSL — force regulation against unknown environments (door opening).
• [2024 IROS] Physically Consistent Online Inertial Adaptation for Humanoid Loco-Manipulation [paper] 🧍
• [2023 IROS] Kinematically-Decoupled Impedance Control for Fast Visual Servoing and Grasping on Quadruped Manipulators [paper] 🐾
• [2023 Auton. Robots] RoLoMa: Robust Loco-Manipulation for Quadruped Robots with Arms [paper] 🐾

Pattern Generation & Stabilization (Humanoid)

• ⭐ [2003 ICRA] Biped Walking Pattern Generation by Using Preview Control of Zero-Moment Point [paper] 🧍 AIST — the seminal ZMP-preview walking pattern generator.
• ⭐ [2015 T-RO] Three-Dimensional Bipedal Walking Control Based on Divergent Component of Motion [paper] 🧍 DLR / TU Munich — seminal DCM formulation.
• ⭐ [2021 RA-L] Humanoid Loco-Manipulations Pattern Generation and Stabilization Control [paper] 🧍 AIST / CNRS-AIST JRL
• [2023 ICRA] Online Nonlinear Centroidal MPC for Humanoid Payload Carrying [paper] 🧍 IIT (iCub)
• [2023 arXiv] Dynamic Loco-Manipulation on HECTOR: Humanoid for Enhanced ConTrol and Open-source Research [paper] [code] 🧍 USC, DRCL
• [2025 arXiv] Whole-Body Control Framework for Humanoid Robots with Heavy Limbs [paper] 🧍 CUHK
• [2024 IROS] Physically Consistent Online Inertial Adaptation for Humanoid Loco-Manipulation [paper] 🧍

Wheeled-Legged & Quadruped-with-Arm

• ⭐ [2019 ICRA] ALMA — Articulated Locomotion and Manipulation for a Torque-Controllable Robot [paper] 🐾 ETH Zurich, RSL
• ⭐ [2021 IROS] Whole-Body MPC and Online Gait Sequence Generation for Wheeled-Legged Robots [paper] 🛞 ETH Zurich, RSL
• ⭐ [2019 RA-L] Keep Rollin' — Whole-Body Motion Control and Planning for Wheeled Quadrupedal Robots [paper] 🛞 ETH Zurich, RSL
• [2019 RA-L] Rolling in the Deep — Hybrid Locomotion for Wheeled-Legged Robots [paper] 🛞 ETH Zurich, RSL
• [2023 arXiv] Versatile Telescopic-Wheeled-Legged Locomotion of Tachyon 3 via Full-Centroidal NMPC [paper] 🛞 Sony Group
• [2024 IROS] Arm-Constrained Curriculum Learning for Loco-Manipulation of the Wheel-Legged Robot [paper] [code] 🛞
• [2022 RA-L] Combining Learning-Based Locomotion Policy With Model-Based Manipulation for Legged Mobile Manipulators [paper] 🐾
• [2020 TRO] A Motion Planning Approach for Nonprehensile Manipulation and Locomotion Tasks of a Legged Robot [paper] 🦿

---

2. 🧠 Learning-Based Control

Whole-Body RL Control

• ⭐ [2024 CoRL] Visual Whole-Body Control for Legged Loco-Manipulation (VBC) [paper] [code] 🐾 UC San Diego (Xiaolong Wang) (Oral)
• ⭐ [2025 ICRA] HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots [paper] [project] [code] 🧍 NVIDIA GEAR + CMU + UC San Diego + UT Austin
• [2024 RA-L] RoboDuet: Learning a Cooperative Policy for Whole-Body Legged Loco-Manipulation [paper] [project] [code] 🐾 Tsinghua / Shanghai AI Lab
• [2025 RSS] HOMIE: Humanoid Loco-Manipulation with Isomorphic Exoskeleton Cockpit [paper] [project] [code] 🧍 Shanghai AI Lab / OpenRobotLab + CUHK
• [2025 arXiv] FALCON: Learning Force-Adaptive Humanoid Loco-Manipulation [paper] [project] [code] 🧍 CMU LeCAR + NVIDIA
• [2025 CoRL] Learning Unified Force and Position Control for Legged Loco-Manipulation [paper] [project] [code] 🦿
• [2025 arXiv] Efficient Learning of a Unified Policy for Whole-Body Manipulation and Locomotion Skills [paper] 🦿
• [2025 arXiv] Versatile Loco-Manipulation through Flexible Interlimb Coordination (RELIC) [paper] [code] 🐾 RAI Institute
• [2025 arXiv] Embracing Bulky Objects with Humanoid Robots: Whole-Body Manipulation with RL [paper] 🧍 HKUST(GZ)
• [2024 CoRL] WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts [project] [code] 🧍 CMU LeCAR
• [2024 IROS] HiLMa-Res: A Hierarchical Framework via Residual RL for Quadrupedal Locomotion and Manipulation [paper] 🐾 UC Berkeley + SFU
• [2024 arXiv] Pedipulate: Enabling Manipulation Skills using a Quadruped Robot's Leg [paper] [project] 🐾 ETH Zurich, RSL
• ⭐ [2022 CoRL] Deep Whole-Body Control: Learning a Unified Policy for Manipulation and Locomotion [paper] [code] 🐾

Motion Tracking & Retargeting from Human Motion

• ⭐ [2024 IROS] H2O: Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation [paper] [project] 🧍 CMU
• ⭐ [2024 CoRL] OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning [paper] [project] 🧍 CMU + SJTU + Shanghai AI Lab
• ⭐ [2024 RSS] ExBody: Expressive Whole-Body Control for Humanoid Robots [paper] [project] [code] 🧍 UC San Diego (Xiaolong Wang) / MIT
• [2024 arXiv] ExBody2: Advanced Expressive Humanoid Whole-Body Control [paper] [project] 🧍 UC San Diego + UC Berkeley + MIT
• ⭐ [2025 arXiv] ASAP: Aligning Simulation and Real-World Physics for Agile Humanoid Whole-Body Skills [paper] [project] [code] 🧍 CMU + NVIDIA + UC Berkeley + UT Austin
• ⭐ [2025 arXiv] SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control [paper] [project] 🧍 NVIDIA GEAR — 42M-param, 700h-mocap motion-tracking foundation controller.
• [2025 arXiv] GMT: General Motion Tracking for Humanoid Whole-Body Control [paper] [project] 🧍 UC San Diego + SFU
• [2025 ICRA] Mobile-TeleVision: Predictive Motion Priors for Humanoid Whole-Body Control [paper] [project] [code] 🧍 UC San Diego + MIT
• [2025 arXiv] BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion [paper] [project] 🧍 UC Berkeley + Stanford
• [2025 arXiv] UniTracker: Learning a Universal Whole-Body Motion Tracker for Humanoid Robots [paper] [project] [code] 🧍 Shanghai AI Lab + ShanghaiTech + SJTU
• [2025 arXiv] ResMimic: From General Motion Tracking to Whole-Body Loco-Manipulation via Residual Learning [paper] [project] [code] 🧍 Amazon FAR + Stanford + UC Berkeley + CMU
• [2025 arXiv] OmniRetarget: Interaction-Preserving Data Generation for Humanoid Loco-Manipulation [paper] [project] [code] 🧍 Amazon FAR + Stanford + UC Berkeley + CMU
• [2025 CVPR] TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization [project] 🧍
• [2025 CVPR] Learning Physics-Based Full-Body Human Reaching and Grasping from Brief Walking References [project] 🧍
• [2025 CoRL] MaskedManipulator: Versatile Whole-Body Control for Loco-Manipulation [paper] [project] [code] 🧍 NVIDIA + SFU

Visual / Perceptive Loco-Manipulation

• ⭐ [2025 CVPR] WildLMa: Long-Horizon Loco-Manipulation in the Wild [paper] [project] 🐾 UC San Diego (Xiaolong Wang) + MIT
• [2024 CoRL] Visual Manipulation with Legs [paper] [project] 🐾 UC San Diego + CMU
• [2025 arXiv] VisualMimic: Visual Humanoid Loco-Manipulation via Motion Tracking and Generation [paper] [project] [code] 🧍 Stanford
• [2025 arXiv] VIRAL: Visual Sim-to-Real at Scale for Humanoid Loco-Manipulation [paper] 🧍 NVIDIA + CMU + UC Berkeley
• [2025 arXiv] Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer [paper] 🧍 NVIDIA + CMU + UC Berkeley
• [2024 CoRL] Learning to Manipulate Anywhere: A Visual Generalizable Framework for Reinforcement Learning [paper] 🦿
• [2024 IROS] Learning Visual Quadrupedal Loco-Manipulation from Demonstrations [paper] [project] 🐾 UC Berkeley + Tsinghua
• [2024 arXiv] Generalizable Humanoid Manipulation with 3D Diffusion Policies [paper] 🧍
• [2024 arXiv] Learning Generalizable Feature Fields for Mobile Manipulation [paper] 🚗
• [2024 Humanoids] Perceptive Pedipulation with Local Obstacle Avoidance [paper] 🐾 ETH Zurich, RSL
• [2024 CoRL] Learning to Open and Traverse Doors with a Legged Manipulator [paper] 🐾 ETH Zurich, RSL

Skill Blending / Hierarchical / Unified Policies

• [2025 arXiv] SkillBlender: Towards Versatile Humanoid Whole-Body Loco-Manipulation via Skill Blending [paper] [project] [code] 🧍 USC + UC Berkeley + Stanford/NVIDIA
• [2025 arXiv] DemoHLM: From One Demonstration to Generalizable Humanoid Loco-Manipulation [paper] [project] [code] 🧍 Peking University
• [2024 CoRL] UMI on Legs: Making Manipulation Policies Mobile with Whole-Body Controllers [project] 🐾
• [2024 IROS] HYPERmotion: Learning Hybrid Behavior Planning for Autonomous Loco-Manipulation [paper] [project] 🐾 IIT
• [2025 ICRA] Catch It! Learning to Catch in Flight with Mobile Dexterous Hands [project] 🐾
• [2024 arXiv] Helpful DoggyBot: Open-World Object Fetching using Legged Robots and VLMs [project] 🐾 UC San Diego + Stanford
• [2024 arXiv] Playful DoggyBot: Learning Agile and Precise Quadrupedal Locomotion [project] 🐾
• [2024 arXiv] LocoMan: Advancing Versatile Quadrupedal Dexterity with Lightweight Loco-Manipulators [paper] [project] [code] 🐾
• [2024 IROS] On Learning Scene-Aware Generative State Abstractions for Task-Level Mobile Manipulation Planning [paper] [code] 🚗 ETH Zurich
• [2024 IROS] BASENET: A Learning-Based Mobile Manipulator Base Pose Sequence Planning for Pickup Tasks [paper] 🚗

Teleoperation & Data Collection

• ⭐ [2024 CoRL] Mobile ALOHA: Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation [paper] [project] 🚗 Stanford
• ⭐ [2024 CoRL] HumanPlus: Humanoid Shadowing and Imitation from Humans [paper] [project] 🧍 Stanford
• ⭐ [2025 CoRL] TWIST: Teleoperated Whole-Body Imitation System [paper] [project] 🧍 Stanford + SFU
• [2025 arXiv] TWIST2: Scalable, Portable, and Holistic Humanoid Data Collection System [paper] [code] 🧍 Amazon FAR + Stanford + UC Berkeley + CMU
• ⭐ [2024 CoRL] Open-TeleVision: Teleoperation with Immersive Active Visual Feedback [paper] [project] [code] 🧍 UC San Diego + MIT
• ⭐ [2023 RSS] AnyTeleop: A General Vision-Based Dexterous Arm-Hand Teleoperation System [paper] [project] [code] 🦾 NVIDIA + UC San Diego
• [2024 CoRL] ACE: A Cross-Platform Visual-Exoskeletons System for Low-Cost Dexterous Teleoperation [paper] [project] [code] 🦾 UC San Diego
• [2024 arXiv] Bunny-VisionPro: Real-Time Bimanual Dexterous Teleoperation for Imitation Learning [paper] [project] [code] 🦾 UC San Diego + HKU
• [2024 IROS] GELLO: A General, Low-Cost, Intuitive Teleoperation Framework for Robot Manipulators [paper] [project] [code] 🦾 UC Berkeley
• [2024 RSS] DexCap: Scalable Portable Mocap Data Collection for Dexterous Manipulation [paper] [project] [code] 🦾 Stanford
• ⭐ [2024 RSS] DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset [paper] [project] [code] 🦾 multi-institution (Levine, Finn)
• [2022 arXiv] TeLeMan: Teleoperation for Legged Robot Loco-Manipulation using Wearable IMU-based Motion Capture [paper] 🐾

Diffusion Policies for Mobile / Legged Manipulation

• ⭐ [2023 RSS] Diffusion Policy: Visuomotor Policy Learning via Action Diffusion [paper] [project] [code] 🦾 Columbia + Stanford + TRI + MIT
• ⭐ [2024 RSS] 3D Diffusion Policy (DP3): Generalizable Visuomotor Policy Learning via Simple 3D Representations [paper] [project] [code] 🦾 Tsinghua / Shanghai Qi Zhi
• [2024 ICRA] NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration [paper] [project] [code] 🚗 UC Berkeley
• [2024 ICML] DiffuseLoco: Real-Time Legged Locomotion Control with Diffusion from Offline Datasets [paper] [project] [code] 🐾 UC Berkeley
• [2025 TPAMI] M2Diffuser: Diffusion-Based Trajectory Optimization for Mobile Manipulation in 3D Scenes [paper] [project] [code] 🚗 BIGAI + HUST
• [2024 arXiv] Combining Planning and Diffusion for Mobility with Unknown Dynamics [paper] 🚗

---

3. 🌐 Foundation Models & High-Level Reasoning

Behavioral Foundation Models (BFM)

• ⭐ [2021 NeurIPS] Learning One Representation to Optimize All Rewards [paper] Meta FAIR — original Forward-Backward representation.
• ⭐ [2023 ICLR] Does Zero-Shot Reinforcement Learning Exist? [paper] Meta FAIR
• ⭐ [2025 ICLR] Meta Motivo: Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models [paper] [project] [code] 🧍 Meta FAIR — first humanoid BFM (FB-CPR).
• [2025 arXiv] BFM-Zero: A Promptable Behavioral Foundation Model for Humanoid Control via Unsupervised RL [paper] [project] 🧍 CMU + Meta FAIR — sim-to-real on Unitree G1.
• [2025 arXiv] Behavior Foundation Model for Humanoid Robots [paper] 🧍 Shanghai AI Lab / InternRobotics
• [2024 NeurIPS] Finer Behavioral Foundation Models via Auto-Regressive Features and Advantage Weighting [paper] Meta FAIR

Vision-Language-Action Models (VLA)

• ⭐ [2025 arXiv] NVIDIA GR00T N1: An Open Foundation Model for Generalist Humanoid Robots [paper] [project] [code] 🧍 NVIDIA GEAR — open humanoid VLA (dual-system).
• [2025 release] GR00T N1.5 (Isaac GR00T) [blog] 🧍 NVIDIA GEAR
• [2025 blog] Helix: A Vision-Language-Action Model for Generalist Humanoid Control [blog] 🧍 Figure AI
• ⭐ [2023 CoRL] RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control [paper] 🦾 Google DeepMind
• ⭐ [2024 CoRL] OpenVLA: An Open-Source Vision-Language-Action Model [paper] [project] [code] 🦾 Stanford + UC Berkeley + Google DeepMind + TRI + MIT
• [2024 RSS] Octo: An Open-Source Generalist Robot Policy [paper] [project] [code] 🦾 UC Berkeley + Stanford + CMU + DeepMind
• ⭐ [2025 RSS] π₀ (Pi0): A Vision-Language-Action Flow Model for General Robot Control [paper] [project] [code] 🦾 Physical Intelligence
• [2025 arXiv] π₀.₅ (Pi0.5): Open-World Generalization for Mobile Manipulation [paper] [project] 🚗 Physical Intelligence
• [2025 ICLR] RDT-1B: A Diffusion Foundation Model for Bimanual Manipulation [paper] [project] [code] 🦾 Tsinghua TSAIL
• [2025 arXiv] Gemini Robotics: Bringing AI into the Physical World [paper] 🦾 Google DeepMind
• [2025 arXiv] LeVERB: Humanoid Whole-Body Control with Latent Vision-Language Instruction [paper] [project] 🧍 UC Berkeley + CMU + SFU
• [2025 RSS] NaVILA: Legged Robot Vision-Language-Action Model for Navigation [paper] [project] [code] 🦿 UC San Diego + NVIDIA + USC
• [2025 arXiv] Humanoid-VLA: Towards Universal Humanoid Control with Visual Integration [paper] 🧍 Westlake MiLAB
• [2024 ECCV] QUAR-VLA: Vision-Language-Action Model for Quadruped Robots [paper] [project] 🐾 Westlake MiLAB
• [2025 arXiv] Galaxea G0: A Dual-System Vision-Language-Action Model [paper] 🚗 Galaxea AI + Tsinghua

VLM / LLM as High-Level Planner

• ⭐ [2022 CoRL] SayCan: Do As I Can, Not As I Say — Grounding Language in Robotic Affordances [paper] 🚗 Google / Everyday Robots
• ⭐ [2023 ICML] PaLM-E: An Embodied Multimodal Language Model [paper] 🦾 Google + TU Berlin
• [2023 CoRL] VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models [paper] [project] [code] 🦾 Stanford SVL
• [2024 CoRL] ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints [paper] [project] [code] 🚗 Stanford SVL
• [2023 CoRL] SayTap: Language to Quadrupedal Locomotion [paper] 🐾 Google DeepMind
• ⭐ [2024 ICLR] Eureka: Human-Level Reward Design via Coding LLMs [paper] [project] [code] 🦿 NVIDIA + UPenn + Caltech + UT Austin
• [2024 RSS] DrEureka: Language Model Guided Sim-to-Real Transfer [paper] [project] [code] 🐾 UPenn + NVIDIA
• [2024 IROS] HYPERmotion: Learning Hybrid Behavior Planning for Autonomous Loco-Manipulation [paper] [project] 🐾 IIT (also listed under Skill Blending)

World Models / World-Action Models (WAM)

• ⭐ [2025 arXiv] Cosmos World Foundation Model Platform for Physical AI [paper] [code] NVIDIA
• [2025 arXiv] Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control [paper] NVIDIA
• [2025 arXiv] Cosmos-Reason1: From Physical Common Sense to Embodied Reasoning [paper] NVIDIA
• [2026 arXiv] MotionWAM: Foundation World-Action Models for Real-Time Humanoid Loco-Manipulation [paper] 🧍 Mondo Robotics + HKUST(GZ)
• [2024 tech report] Genie 2: A Large-Scale Foundation World Model [blog] Google DeepMind
• [2024 tech report] 1X World Model [blog] 🧍 1X Technologies

---

🏛️ Research Groups & Lineages

A few major threads run through the field — useful for tracing how ideas evolved:

💡 Note on "GRILITE": there is no NVIDIA model by that name — it most likely refers to GR00T (NVIDIA's humanoid VLA), possibly conflated with the Fourier GR-1 humanoid that GR00T is demonstrated on.

---

🤖 Robot Models

🐾 Quadruped robots with arm

🛞 Wheel-legged robots with arm

🚗 Mobile robots with arm

---

✨ Contributing

If you find any interesting papers or resources related to loco-manipulation, feel free to open an issue or submit a pull request!

📝 Format Guidelines

• Section: Place the entry under the matching methodology sub-section (model-based / learning-based / foundation model).
• Entry format: - **[Year Venue]** Title [[paper](url)] [[project](url)] [[code](url)] <embodiment emoji> *Group/Institution*
• Venue abbreviations: Use standard ones (CoRL, ICRA, IROS, RSS, CVPR, arXiv, ...).
• Embodiment: Tag with 🧍 humanoid · 🐾 quadruped · 🛞 wheeled-legged · 🚗 mobile · 🦿 general legged · 🦾 arm/manipulation-only.
• Seminal works: Prefix with ⭐ if foundational / highly-cited.
• Links: Include both paper and project/code links when available.

---

Last updated: June 2026 — all arXiv IDs verified to resolve; project/code links HTTP-checked.