Humanoid Production Update - 2026-07-24 - Article - Figure reaches 1,000 robots, EngineAI activates a second production base and Humanoid funds a partner-led manufacturing plan.
Humanoid production — June 2026 - 2026-06-29 - Article - Company developments across robotics, humanoid production, deployments, supply chains and robot learning.
Robotics capital moves — 8–14 June 2026 - 2026-06-08 - Article - Funding rounds, acquisitions and other capital transactions across robotics, autonomous systems and their enabling technologies.
Research
Vision as Proprioception: Estimating Humanoid Robot Pose from Video - 2026-10-18 - Research Publication - HumanoidPose estimates humanoid joint configurations and root pose from ordinary video using pretrained human-vision features and a kinematics-conditioned decoder. The ETH Zürich team trains across16 synthetic embodiments… - source: arxiv.org
Learning Agile Perceptive Traversal of Sparse 3D Structures for Humanoids - 2026-08-30 - Research Publication - The system combines an attention-based lidar encoder, recurrent memory and phase-scheduled teacher-student reinforcement learning, with hardware-aware models for lidar noise, battery-voltage sag and actuator thermal limits. - source: arxiv.org
ENGINEAI ROS2 interfaces and simulation examples - Research Publication - ENGINEAI’s ROS2 workspace provides robot interfaces, MuJoCo simulation and control examples. The current source tree contains PM01, T800 and T800 Pro configurations, with an included PM01 walking-policy file. - source: github.com
EngineAI robot description collection - Research Publication - EngineAI publishes separate PM01, T800 and T800 Pro robot descriptions and mesh assets. The collection distinguishes PM01 lidar and EDU revisions, T800, and 25/43-DoF T800 Pro configurations. - source: github.com
EngineAI native control SDK - Research Publication - EngineAI releases a modular control runtime with task orchestration, simulation and hardware-deployment tooling. Named configurations include PM01 EDU and T800, with robot models and actual MNN walking, dance and transition policies. - source: github.com
EngineAI RL Lab - Research Publication - EngineAI RL Lab supplies Isaac Lab training environments for PM01 and T800 whole-body tracking and adversarial-motion-prior locomotion. Source includes task registrations, replay and deployment workflows plus exported ONNX policies. - source: github.com
EngineAI PM01 navigation stack - Research Publication - EngineAI’s navigation stack trains an SRU-based policy from ordered lidar range images and executes velocity commands through a walking controller. It contains separate PM01 Gazebo and physical ROS 2 deployment trees, preprocessing and… - source: github.com