Robot learning frameworks
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | AI territory |
|---|---|
| Categories | 6 |
| Companies | 31 |
| Map changed | 2026-11-19 |
| Map | open the map |
Robot learning frameworks is a territory of the AI industry on map of ai. In short: humanoid robot simulation benchmarking provides standardized test environments and evaluation metrics for simulating humanoid robot behaviors in physics simulators. The map lists 31 products and companies in 6 categories.
The map of robot learning frameworks has 6 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.
robot foundation models7 products and companies · its own article →
Robot foundation models are large pretrained neural networks trained on diverse robot data that transfer to new tasks and embodiments. Roboticists, researchers, and robot manufacturers use them to reduce training time for specific manipulation, navigation, and perception tasks.
Products and companies
Papers, last 30 days
- ScaleVLA: Scaling Vision-Language-Action Models to 552 Billion Parameters 2026-11-16
- Synthesizing Dexterous Robot Demonstrations from Human Videos via Virtual Try-On 2026-11-03
- Sensorimotor Alignment: A Platonic Proxy for Embodied Policy Success 2026-10-22
- Trajectory-Calibrated Merging of Embodied Policies 2026-11-19
- $Z$-sight: Unifying Vision-Language-Action and Latent World Modeling 2026-11-18
Open ground
Google DeepMind, OpenAI, Boston Dynamics, and NVIDIA each release foundation models trained on collected or simulated robot trajectories; none of them yet publishes models trained on a unified dataset across five or more distinct robot morphologies in production use.
open source robot learning libraries5 products and companies
Libraries and frameworks that enable robots to learn from data through reinforcement learning, imitation learning, or other machine learning techniques. Roboticists, researchers, and engineers use these to train policies for robot control tasks.
Products and companies
simulation and training environments5 products and companies
Simulation and training environments are software platforms where robots learn policies and behaviors in virtual worlds before deployment on physical hardware. Roboticists, researchers, and engineers use these to accelerate learning cycles and test control strategies without hardware costs.
Products and companies
teleoperation and data collection3 products and companies
Teleoperation and data collection involves remote control of robots and systematic capture of operational data to train learning systems. Roboticists, industrial operators, and ML engineers use these tools to gather supervised demonstrations and real-world interaction logs.
Products and companies: scale ai, standard bots, cortex robotics.
humanoid robot platforms6 products and companies · its own article →
Humanoid robot platforms are physical robots with two arms, a torso, and a head designed to operate in human environments. Researchers, roboticists, and companies use them to develop and test manipulation skills, navigation, and human-robot interaction.
Products and companies: tesla optimus, 1x technologies, agility robotics, apptronik, unitree, boston dynamics.
Papers, last 30 days:- DyCraft: An Agentic System for Long-Horizon 4D Scene Generation 2026-10-29
- HuGe: Scaling Humanoid Generalist Vision-Language-Action Models with Egocentric Experience 2026-10-25
- ODE Policy: Robot Learning via Action Dynamics, Not Actions 2026-11-01
- Reference-Free Humanoid Skateboarding via Termination-Shaped Reinforcement Learning 2026-11-19
- Controlling the Humanoid Robot H1-2 With a Simulated Biologically Plausible Network 2026-11-19
warehouse and industrial manipulation5 products and companies
Systems and software that enable robots to learn manipulation tasks through interaction with warehouse and factory environments. Manufacturing plants, logistics hubs, and research institutions use these to train robotic arms and mobile manipulators on picking, placing, sorting, and assembly operations.
Products and companies: covariant, dexterity, osaro, plus one robotics, mujin.
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong.