Agile
Agile ONE is a bipedal industrial humanoid robot by Agile Robots SE (Munich, Germany), designed for manufacturing and logistics. Features the world's most dexterous five-fingered robotic hands (21 joints per hand, force-torque sensors at every joint). Powered by Agile Robots' layered Robotic Foundation Models. Height 174cm, weight ~69kg, 40 DoF, payload ~20kg.
Primary use case: Industrial manufacturing and logistics: material handling, machine tending, fine assembly
Height
174 cm
Weight
69 kg
DOF
71
Payload
20 kg
Specifications
Physical
Height
174 cm
Weight
69 kg
Degrees of Freedom
71
Locomotion
bipedal
Manipulation
Hand Type
test
Payload
20 kg
Max Lift/Arm
3 kg
Power & Autonomy
Autonomy Level
semi-autonomous
Commercial
Availability
pre-production
Primary Use Case
Industrial manufacturing and logistics: material handling, machine tending, fine assembly
Related news
Swiss startup Embodied AI launches AIR, a humanoid robot that learns from factory workers
Toyota Robotics: Toyota just established a dedicated robotics division
Inside Neura Robotics' massive physical AI strategy: more than humanoids
The engineer who taught 1X's NEO to walk just raised $30 million to give robots a sense of touch
Agile Robots and Audi piloting humanoids ... on a task done 1 million times a day
Agile and Generalized Legged Locomotion via Attention-Based Neural Map Encoding
Achieving agile and generalized legged locomotion across terrains requires tight integration of perception and control, especially under occlusions and sparse footholds. Existing methods have demonstrated agility on parkour courses but often rely on end-to-end sensorimotor models with limited generalization and interpretability. By contrast, methods targeting generalized locomotion typically exhibit limited agility and struggle with visual occlusions. We introduce a unified reinforcement learning (RL) framework for agile and generalized locomotion that incorporates a novel attention-based map encoder in the control policy. This encoder extracts local and global mapping features and uses attention mechanisms to focus on salient regions, producing an interpretable and generalized embedding for RL-based control. We further propose a learning-based mapping pipeline that provides fast, uncertainty-aware terrain representations robust to noise and occlusions, serving as policy inputs. It use
NVIDIA expands Isaac tools for humanoid robot development
NVIDIA expands Isaac tools for humanoid robot development
Nvidia Expands Physical AI Stack Across Robotics, Humanoids and Factory Automation
Nvidia Expands Physical AI Stack Across Robotics, Humanoids and Factory Automation