humanoidsdata.com

Search

Search datasets, articles, and glossary terms for humanoids and embodied AI.

← All glossary terms

Hardware & control

Locomotion

Robot locomotion is the process by which a robot transports its body from one place to another through physical interaction with its environment. Robots may walk, roll, crawl, hop, fly, or swim. For a humanoid, locomotion usually means coordinated legged movement such as walking, turning, stepping, climbing, or recovering balance.

Also known as: robot locomotion, robotic locomotion

Updated

Locomotion is physical movement, not navigation

Locomotion concerns how a robot generates motion through contact with its environment. Navigation concerns where the robot should go and how it plans a route. A navigation system may choose a destination while a locomotion controller turns that decision into wheel motion, footsteps or another mode of travel.

The MIT Press introduction to mobile robots treats locomotion mechanisms, sensing, localisation and planning as connected but distinct parts of mobility. Wheels are efficient on suitable ground; legs can place contacts across gaps, steps and uneven terrain at the cost of greater mechanical and control complexity.

Humanoid locomotion is a contact-switching problem

Bipedal walking repeatedly makes and breaks foot contact. Impacts, friction limits, centre-of-mass motion, angular momentum, joint limits and actuator capability all affect whether a step remains stable. The MIT notes on legged robots explain why humanoid control often uses simplified centre-of-mass dynamics and footstep planning despite the full body's complexity.

A gait describes a recurring pattern of contacts and body motion, such as walking or running. Locomotion is broader: it includes starting, stopping, turning, climbing, transition between gaits and recovery after a disturbance.

Useful locomotion data extends beyond joint angles

A motion clip can show the visible pose while omitting the forces and state that kept the robot upright. Training records may need joint position and velocity, commanded and measured torque, base pose, IMU data, foot contacts, ground reaction forces, terrain geometry, target velocity and intervention or fall labels.

The robot model, footwear, payload, control rate and support surface are also part of the context. Those fields help distinguish a reusable movement pattern from a trajectory that worked only under one embodiment and floor condition.

Sources