Mendy's Robotics and AI

Robotics Glossary

Robotics Glossary

A

  • Actuator – The component that converts electrical, pneumatic, or hydraulic energy into mechanical motion.
  • Adaptive Control – A control strategy that adjusts its own parameters in real time to maintain performance as conditions change.
  • AGV (Automated Guided Vehicle) – Floor-bound mobile robot that follows fixed guidance (magnetic tape, QR codes, lasers) in factories and warehouses.
  • AI (Artificial Intelligence) – Computational techniques enabling perception, reasoning and decision-making; in robots it powers vision, planning, learning and autonomy.
  • AMR (Autonomous Mobile Robot) – Vehicle that plans paths dynamically and avoids obstacles without fixed guides.
  • Anthropomorphic Robot – Manipulator or mobile robot designed with human-like proportions or movements.
  • Articulated Robot – A robot arm with rotary joints (2 – 10 + axes) that resemble a human arm’s shoulder-elbow-wrist layout.
  • Augmented Reality (AR) – Overlay of digital data on the real world; used to program or service robots via head-mounted displays.
  • Axis – A single degree of freedom in which a joint can move (rotational or translational).

B

  • Backlash – Mechanical “slop” between mating gear teeth or linkages, causing positioning error.
  • Bandwidth (Control) – Frequency range over which a control loop can faithfully track commands.
  • Battery Management System (BMS) – Electronics that monitor, balance and protect rechargeable cells in mobile robots.
  • Bipedal Robot – Legged robot with two feet, designed for human terrain and stairs.
  • BLDC Motor (Brushless DC) – High-efficiency motor using electronic commutation; dominant in drones and precision servos.
  • Bus (e.g., CAN, RS-485) – Shared communication line linking sensors, drives and controllers.

c

  • Calibration – Procedure that aligns a robot’s internal model with physical reality (e.g., tool-center-point or camera-to-arm alignment).
  • Capacitive Sensor – Touch or proximity sensor that detects changes in electric field.
  • Cartesian Robot (Gantry) – Three linear axes arranged at 90°; common in pick-and-place and 3-D printers.
  • Cobot (Collaborative Robot) – Robot designed to operate safely next to humans, typically with force-limited joints and advanced sensing.
  • Compliance – Measure of how much a structure deflects under load; intentional compliance protects against impacts.
  • Controller – The hardware–software unit that closes the control loop and executes motion plans.
  • CPG (Central Pattern Generator) – Biologically inspired oscillator network that drives rhythmic motions such as walking or flapping.
  • Cycloidal Drive – Compact gear reducer with high ratio and zero backlash, often used in industrial arms.

D

  • Dead Reckoning – Position estimation by integrating wheel or joint motion over time, prone to drift.
  • Degrees of Freedom (DoF) – Independent movements a mechanism can perform; six DoF describe full 3-D position + orientation.
  • Differential Drive – Mobile base steered by varying the speeds of left and right wheels.
  • Direct Drive – Actuator architecture with the motor shaft coupled directly to the load, eliminating gear reduction.
  • Drone (UAV) – Unmanned aerial vehicle; multirotor drones are a major robotics subfield.
  • Drive Amplifier – Power electronics module that supplies precise current/voltage to a motor.

E

  • Edge Computing – Running perception and control algorithms on-board the robot instead of in the cloud to cut latency.
  • Elastic Actuator (Series Elastic Actuator) – Motor output passes through a calibrated spring, permitting force control and safe contact.
  • Encoder – Sensor that converts rotary or linear displacement into digital signals; essential for position feedback.
  • End Effector – Tool attached to the robot’s wrist (gripper, welder, dispenser, etc.).
  • EtherCAT – Real-time industrial Ethernet fieldbus widely used for multi-axis servo control.
  • EKF (Extended Kalman Filter) – Non-linear state estimator used for SLAM and sensor fusion.
  • Exoskeleton – Wearable robot that augments human strength or assists rehabilitation.

F

  • Feedforward – Control term that predicts required actuation based on a model rather than sensor feedback.

  • Feedback – Sensor information returned to the controller to correct errors.
  • Force–Torque Sensor – Multi-axis transducer in wrists or grippers that measures interaction forces.
  • Forward Kinematics – Calculating end-effector pose from known joint angles.
  • Friction – Resistive force hampering motion; static, kinetic and viscous forms all influence control.
  • Fuzzy Logic – Rule-based control method handling imprecise inputs; once popular for mobile-robot navigation.

G

  • G-code – Text command language for CNC machines and many 3-D printers.
  • Gantry Robot – Large Cartesian robot spanning a workspace like an overhead crane.
  • Gear Ratio – Relationship between motor turns and output shaft turns; higher ratios boost torque, lower inertia.
  • Gripper – End-effector that grasps objects; can be mechanical, vacuum, magnetic or soft.
  • Gyroscope – Sensor that measures angular velocity for balance and navigation.
  • Gaussian Process – Probabilistic model used for learning robot dynamics or terrain properties.

H

  • H-bridge – Power circuit that drives DC motors forward or reverse.
  • Harmonic Drive – High-precision strain-wave gear reducer with zero backlash; common in joint servos
  • Holonomic – Robot whose controllable DoF equal its total DoF (e.g., omnidirectional Mecanum platform).
  • Human–Robot Interaction (HRI) – Study and design of communication, safety and ergonomics between people and robots.
  • Hydraulic Actuator – Cylinder or vane using pressurized fluid for very high force/weight ratios.

I

  • IMU (Inertial Measurement Unit) – Package combining accelerometers and gyros for 6-DoF motion sensing.
  • Inverse Kinematics – Computing joint angles that achieve a desired end-effector pose.
  • ISO 10218 / ISO TS 15066 – International safety standards for industrial and collaborative robots.
  • Intelligent Control – Umbrella term for AI-enhanced adaptive, learning or model-based controllers.
  • Interpolation – Generating intermediate joint or toolpath points for smooth motion.

J

  • Jacobian – Matrix relating joint velocities to end-effector linear/angular velocities; central to control and IK.
  • Jerk – Time derivative of acceleration; limiting jerk yields smoother motion and less vibration.
  • Joint – Mechanical connection allowing relative movement between links (revolute, prismatic, etc.).
  • Jogging – Manually commanding incremental robot moves with a teach pendant.

K

  • Kalman Filter – Optimal recursive estimator for linear Gaussian systems; backbone of many navigation stacks.
  • Kinematics – Study of motion without considering forces; contrasts with dynamics.
  • Kinesthetic Teaching – Programming method where an operator physically guides a compliant robot through a task.
  • KRL (KUKA Robot Language) – Proprietary language for KUKA industrial arms.

L

  • Lidar – Laser ranging sensor that produces dense 2-D/3-D point clouds.
  • Linear Actuator – Device that converts rotary motion into straight-line travel (screw, belt, rod-less).
  • Localization – Estimating a robot’s pose within a known map.
  • Loop Closure – Event in SLAM when the robot recognizes a previously visited place, reducing accumulated drift.
  • LQR (Linear Quadratic Regulator) – Optimal control law minimizing a quadratic cost on state and input.
  • LVDT (Linear Variable Differential Transformer) – Precision inductive position sensor.

M

  • Manipulator – Robot arm that interacts with objects in its workspace.
  • Mecanum Wheel – Wheel with angled rollers enabling omnidirectional translation.
  • Middleware – Software layer (e.g., ROS) that abstracts hardware and passes messages between modules.
  • Model Predictive Control (MPC) – Optimization-based controller that plans over a receding horizon.
  • Motion Planning – Algorithmic generation of a collision-free, dynamically feasible trajectory.
  • MTBF (Mean Time Between Failures) – Reliability metric critical for production robots.
  • Machine Vision – Cameras and algorithms enabling inspection, guidance and measurement.

N

  • NEMA Frame – Standardized motor mounting dimensions (e.g., NEMA 17 stepper).

  • Neural Network – Data-driven function approximator widely used for perception and decision-making.
  • Nonholonomic Constraint – Velocity constraint that cannot be integrated into a position constraint (e.g., car cannot move sideways).
  • Node (ROS) – Executable that publishes and subscribes to topics in a ROS graph.
  • Navigation Stack (ROS) – Collection of packages providing mapping, localization, path planning and control.

O

  • Odometry – Incremental pose estimate derived from encoders or IMU; foundation for mobile-robot navigation.
  • Omnidirectional Robot – Platform that can translate in any planar direction without rotation.
  • Operational Space Control – Control formulation that commands forces/accelerations directly in task space.
  • Opto-isolator – Device that electrically isolates high-power drive circuits from logic electronics.
  • Overcurrent Protection – Safety feature that cuts power when motor current exceeds limits.

P

  • PID Controller – Proportional-Integral-Derivative loop; default choice for low-level servo control.
  • PLC (Programmable Logic Controller) – Rugged industrial computer for discrete and analog I/O control.
  • Payload – Maximum mass a robot can handle without impaired accuracy or safety.
  • Path Planning – Generating a geometric route from start to goal (e.g., RRT*, A*).
  • Proprioception – Robot’s internal sensing of its own state (joint angles, battery level).
  • Proximity Sensor – Detects nearby objects without contact (inductive, capacitive, IR).
  • PWM (Pulse-Width Modulation) – Modulates average voltage/current to drive motors efficiently.

Q

  • Quadrature Encoder – Incremental encoder with two channels 90° out of phase, giving direction and high resolution.
  • Quaternion – Four-element representation of 3-D orientation without gimbal lock.
  • Quick-Change Tooling – Mechanical interface allowing rapid end-effector swaps without recalibration.
  • Q-learning – Reinforcement-learning algorithm that updates value tables to learn optimal actions.

R

  • Reach – Furthest distance the end-effector can extend from the base.
  • Repeatability – Ability to return to a taught point; key metric for industrial arms.
  • Resolution – Smallest distinguishable movement as limited by encoder counts or DAC steps.
  • Reactive Control – Fast sensor-to-actuator loops that handle disturbances without global planning.
  • Redundancy – Having more DoF than strictly needed, enabling obstacle avoidance or dexterous posture.
  • Reinforcement Learning (RL) – Trial-and-error method for training policies through reward signals.
  • ROS (Robot Operating System) / ROS 2 – Open middleware framework providing message passing, drivers and tools.
  • RRT (Rapidly-exploring Random Tree) – Sampling-based algorithm for high-DoF motion planning.
  • RViz – 3-D visualization tool in ROS for state, sensor and map data.

S

  • SCARA Robot – Selective Compliance Assembly Robot Arm; jointed-plane robot ideal for pick-and-place.
  • SEA (Series Elastic Actuator) – See Elastic Actuator; provides compliance and force sensing.
  • Sensor Fusion – Combining multiple sensor modalities to improve accuracy and robustness.
  • Servo – Closed-loop actuator (motor + feedback + controller).
  • SLAM (Simultaneous Localization and Mapping) – Building a map of an unknown environment while self-localizing in it.
  • Soft Robot – Robot built from highly compliant materials, capable of safe, adaptive manipulation.
  • Stepper Motor – Brushless motor that moves in discrete steps; open-loop but precise for low-speed tasks.
  • Stiffness – Opposite of compliance; high stiffness yields precise positioning but harder contacts.
  • Swarm Robotics – Coordinated behavior of large numbers of simple robots acting as a collective.

T

  • Teach Pendant – Handheld interface used to program and jog industrial robots.
  • Teleoperation – Human remotely controls a robot, often with force or visual feedback.
  • Time-of-Flight (ToF) Sensor – Measures distance by timing light or sound pulses.
  • Torque – Rotary equivalent of force; key spec for joint servos.
  • Trajectory – Time-parameterized path describing desired position, velocity and acceleration.
  • Transformation Matrix – 4 × 4 homogeneous matrix encoding rotation and translation between frames.
  • Turing Test (Robotics) – Philosophical benchmark for robot intelligence/human indistinguishability.

U

  • UAV / UGV / USV – Aerial, ground and surface robotic vehicles respectively.
  • UDP – Lightweight, connectionless network protocol used for high-rate streaming data.
  • Underactuated System – Robot with fewer actuators than DoF (e.g., passive walking robots).
  • URDF (Unified Robot Description Format) – XML schema for robot kinematic and visual models in ROS.
  • URDF (Unified Robot Description Format) – XML schema for robot kinematic and visual models in ROS.
  • Ultrasonic Sensor – Distance sensor that measures echo time of 40 kHz sound pulses.

V

  • Vision System – Integrated cameras, illumination and processing for inspection or guidance.
  • Velocity Control – Mode where the controller commands joint speeds rather than positions or torques.
  • V-REP / CoppeliaSim – Physics-based robot simulation environment with remote-API control.
  • Voltage Regulator – Circuit that maintains constant supply voltage to sensitive electronics.
  • Vibratory Feeder – Bowl or linear track that orients small parts for automated assembly.

W

  • Waypoint – Intermediate pose used to break a complex path into manageable segments.
  • Workspace / Work Envelope – Set of all points a robot can reach.
  • Wheeled Robot – Mobile robot using wheels for locomotion (differential, Ackermann, omnidirectional).
  • Worm Gear – Screw-like gear meshing with a toothed wheel, providing high reduction and self-locking.
  • Wireless Charging – Inductive or resonant energy transfer to recharge mobile robots without plugs.

X

  • X-axis – Horizontal axis in a Cartesian frame; robotics coordinate frames follow the right-hand rule.
  • XBee – Low-power radio module (IEEE 802.15.4) used for short-range robot networking.

Y

  • Yaw – Rotation about the vertical (Z) axis; crucial for drone and car heading control.
  • Y-askew Error – Small lateral alignment error when docking or palletizing.
  • YCB Object Set – Standardized collection of everyday items for benchmarking manipulation.

Z

  • Zero Moment Point (ZMP) – Point where tipping moment equals zero; stability metric for walking robots.
  • Zeroing – Calibration move driving an axis to a reference to establish absolute position.
  • Z-axis – Vertical axis in most right-hand frames; positive upward in ROS convention.