Hardware & control
Robot
A robot is a programmed, actuated mechanism with some degree of autonomy that performs locomotion, manipulation or positioning. The term includes the mechanism and its control system, but it does not imply a human-like body, machine learning, general intelligence or fully independent operation.
Updated
A robot combines mechanism, actuation and control
ISO 8373:2021 defines a robot as a programmed actuated mechanism with a degree of autonomy for locomotion, manipulation or positioning. Its control system is part of that system.
This scope covers industrial manipulators, mobile platforms, humanoids and other mechanical structures. A software agent without a physical actuated mechanism is not a robot under this definition, even when its interface uses a robot character.
Autonomy does not mean no human involvement
The ISO definition requires a degree of autonomy, not complete independence. Teleoperation alone does not settle the classification: a platform may still qualify as a robot when low-level functions such as stabilisation or command execution provide some autonomy. A mechanism whose task motion is wholly and continuously determined by an operator may instead be classed as a robotic device under the ISO vocabulary.
A more autonomous system may perceive, plan and act for longer without intervention, but that difference should be described at the task level rather than inferred from appearance.
The paper describing the ISO vocabulary revision explains that the broader robot definition was designed to apply across industrial and service robots while retaining that distinction. A “robot system” is also broader than one robot: it can include the robot and supporting equipment, so the terms are not synonyms.
“Robot” is not a capability claim
The label alone says nothing about morphology, payload, safety, reliability, intelligence or task range. A fixed arm and a biped humanoid are both robots but expose very different observations, actions and physical constraints.
Robot data should therefore identify the exact embodiment, controller, sensors, actuators, end effectors, operating mode and environment. A policy or trajectory recorded on one robot is not automatically executable on another merely because both systems share the same top-level label.
Sources
Related terms
Hardware & control
Humanoid robot
A humanoid robot is a robot whose body is modelled on the human form, typically with a torso, head and limbs, so it can move through human spaces or use human-scale tools and interfaces. The term describes morphology, not intelligence or autonomy; wheeled, partial-body and simplified-hand designs are also sometimes classed as humanoids, so the robot's actual configuration should be stated.
Hardware & control
Actuator
An actuator is a robot’s power mechanism that converts electrical, hydraulic, pneumatic or other supplied energy into controlled mechanical force, torque or motion. In a humanoid, actuators drive joints, often through transmissions; they are distinct from the joints and sensors, and their arrangement need not correspond one-to-one with the robot’s degrees of freedom.
Hardware & control
Autonomy
Robot autonomy is the ability to perform an intended task from the robot’s current state and sensor information without human intervention during that task. Autonomy is contextual and can differ by function, environment and operating phase; it is not a single permanent capability level for the whole robot.
Hardware & control
Robot embodiment
A robot embodiment is the particular body and sensorimotor interface through which a robot perceives and acts. It includes morphology and kinematics, actuators, end effectors, sensors, physical limits, and the observation and action conventions exposed to a controller or learned policy. Two robots can perform the same task while having different embodiments.
Models & learning
Policy
A policy is the decision rule that maps a robot’s current observations or estimated state, and sometimes a task instruction, to an action or probability distribution over actions. It can be hand-designed or learned from demonstrations, rewards or both. In humanoid robotics, its outputs may be joint targets, torques, end-effector changes or higher-level skills.