Hardware & control
Coordinate frame
A coordinate frame is a defined origin and set of oriented axes used to express positions, orientations, motions, forces, or other spatial quantities. A value has no complete geometric meaning until its frame and convention are known. Transformations relate measurements expressed in frames such as world, robot base, camera, end effector, object, or sensor.
Also known as: reference frame, frame of reference, coordinate frames
Updated
Spatial numbers need an origin and axes
The vector (0.1, 0, 0) could mean ten centimetres forward from a robot base, rightward from a camera or along an object's local axis. The numeric value alone cannot resolve the difference. A coordinate frame defines the origin, axis directions and handedness needed to interpret it.
Modern Robotics uses rotation and homogeneous transformation matrices to represent rigid-body configurations and change the frame in which a vector or pose is expressed. A transform between two frames contains both their relative translation and orientation.
Robotics systems use many connected frames
Common frames include a fixed world or map frame, a moving robot base frame, link and joint frames, an end-effector or tool frame, camera optical frames and object frames. ROS REP 120 adds conventions for humanoid locations such as wrists, hands, feet and soles.
Axis conventions are not universal. REP 103 recommends right-handed body frames with x forward, y left and z up, while optical camera frames use a different named convention. A frame identifier must therefore travel with the values.
Frame errors can look numerically plausible
Applying a pose or action in the wrong frame can rotate directions, shift origins or reverse signs while leaving every number within a reasonable range. This makes frame mistakes dangerous in cross-robot datasets.
Dataset documentation should name the parent and child frames, units, axis convention, transform direction, calibration source and whether actions are absolute or relative. Time-varying transforms also need timestamps because the relationship between a moving camera, hand and world changes throughout an episode.
Sources
Related terms
Hardware & control
Robot kinematics
Robot kinematics describes the geometric relationship between a robot’s joint configuration and the position, orientation and velocity of its links or end-effector, without modelling the forces that cause the motion. Forward kinematics computes pose from joint values; inverse kinematics searches for joint values that achieve a requested pose.
Models & learning
Pose estimation
Pose estimation is the process of inferring the position and orientation of a body, object, camera, hand, or robot relative to a specified coordinate frame. In three-dimensional robotics this is often called 6D or 6-DoF pose estimation because the result has three translational and three rotational degrees of freedom, even when orientation is stored with more than three numbers.
Models & learning
Action space
An action space is the set and representation of commands that an agent or robot policy is allowed to choose. In robotics, actions may be discrete choices or continuous values such as joint targets, motor torques, end-effector pose changes, base velocities, or gripper commands. The action space defines what the policy outputs, not necessarily the motion the hardware ultimately executes.
Data & collection
Sensor calibration
Sensor calibration is the estimation and documentation of parameters that map raw sensor readings into physically meaningful values and known spatial relationships. In robotics it can include scale, bias, distortion, intrinsic camera parameters, sensor-to-sensor or sensor-to-robot transforms, and timing offsets. Calibration does not remove all noise or drift.
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.