London-based robotics startup Humanoid has unveiled HMND 01 Alpha, a wheeled robot with a human-like torso, head and arms.
Its use cases include goods handling, picking and packing, and parts handling for manufacturing, which could include feeding machines in automotive, electronics and other manufacturing facilities.
The robot could also be deployed in warehouses, logistics hubs and retail facilities where there is a need for repetitive or physically demanding tasks.
Humanoid founder Artem Sokolov said that he was inspired to create the company after seeing the monotonous work of his grandparents, who worked their entire lives in jewellery production.
Solokov said that, instead of taking people’s jobs as many fear, increasing automation could “free people from routine and repetitive tasks, allowing them to engage in more creative and meaningful work”.
HMND 01 Alpha is 2.2 metres tall (7 foot 3 inches) and weighs 300kg (661 lb).
It has a maximum speed of 7.2 kph (around 4.4 mph) and an average runtime of four hours.
When using both arms, it can carry payloads of up to 15kg (33.1 lb), with the ability to lift even more when objects are closer to its body.
Humanoid robot can use five-fingered hands guided by AI
Its end-effectors, the robotic equivalent of hands, can be deployed as actual five-fingered hands or a parallel gripper, depending on whether the task needs manual dexterity or simple lifting and handling.
The head, meanwhile, features 360-degree colour cameras and double depth sensors for comprehensive perception.
The robot also features AI-powered end-to-end reasoning, incorporating NVIDIA’s newly released Jetson Thor platform.
This helps it to adapt to people and environments in real time and make decisions as it runs multiple AI workflows and large-scale generative AI models simultaneously.
While many humanoid robots in development use human-like legs to walk, HMND 01 Alpha combines the versatility of a humanoid torso and limbs with a more stable wheeled base.
A bipedal version is planned for late 2026, however, with roles aimed at home and service rather than industrial environments.
Humanoid said that its development included a unique combination of 360-degree simulation training and real-world data flywheels.
These are feedback loops in which data collected from processes or interactions can be used to continuously refine AI models.
Today’s news was brought to you by TD SYNNEX – the UK’s number one solutions distributor.
Read more of our latest Industry Updates stories
