Reference
Robotics glossary
473 terms, each in one sentence. Click any term for the full explanation, with how it works, where you will see it, and sources.
A
- A* (A-Star) Pathfinding
- A* finds the shortest path between two points on a grid or graph.
- Absolute Encoder
- An absolute encoder reports a shaft's exact angle the instant it powers on — no homing needed — which makes it the safe, preferred position sensor…
- Accelerometer in Robotics
- An accelerometer measures linear acceleration along an axis.
- Ackermann Steering
- Ackermann steering is the car-like geometry where the front wheels turn at slightly different angles so all four roll cleanly around a shared cente…
- Actor-Critic
- Actor-critic methods pair a policy that acts with a value estimator that critiques — combining the strengths of policy-gradient and value-based RL,…
- Actuator
- An actuator is the part of a machine that turns a control signal into physical movement — force, torque or displacement.
- Adaptive Control in Robotics
- Adaptive control automatically adjusts its parameters as the robot operates — handling changing payloads, friction, or environments without manual re-tuning.
- Additive Manufacturing (3D Printing) for Robotics
- Additive manufacturing builds robot parts layer by layer.
- Admittance Control
- Admittance control lets a stiff, position-controlled robot behave as if it were soft — it senses contact force and moves in response, so a heavy in…
- Agricultural robot
- An agricultural robot — or agribot — is a machine that automates farm tasks such as planting, monitoring, weeding, and harvesting, reducing the nee…
- Agricultural Robotics in India
- Agricultural robots in India tackle labour shortages and yield gaps — drones spray pesticides, robots harvest tomatoes, AI sorts produce.
- Agricultural Spraying Drone
- Spraying drones cover Indian fields with pesticide and foliar nutrients 20× faster than manual labour.
- AMCL (Adaptive Monte Carlo Localisation)
- AMCL is ROS2's standard particle-filter localiser.
- Angular Momentum
- Angular momentum is the rotational version of momentum — conserved, and the key to how robots balance, spin, flip, and control their orientation, f…
- Angular Velocity
- Angular velocity is how fast and about which axis something is spinning — the rotational twin of linear velocity, and the quantity a robot's gyrosc…
- Antipodal Grasp
- An antipodal grasp pinches an object between two contacts pressing directly opposite each other — the simplest reliable two-finger grip, and the pa…
- Arduino
- Arduino is an open-source electronics platform whose boards each carry a microcontroller — one chip that runs a single program on its own, with no…
- Arduino-Based Robot Control
- Arduino boards are the most common microcontroller platform for beginner robotics.
- Asimov's Three Laws of Robotics
- Isaac Asimov's Three Laws of Robotics define a safety hierarchy for fictional robots.
- Augmented Reality in Robotics
- AR overlays virtual content on the real world.
- Automated Guided Vehicle (AGV)
- An automated guided vehicle follows fixed, pre-defined routes to move materials through factories and warehouses — the simple, reliable, decades-ol…
- Autonomous Mobile Robot (AMR)
- An autonomous mobile robot (AMR) is a robot that drives itself around a building or a street, working out its own route as it goes — with nobody st…
- Autonomous vehicle
- An autonomous vehicle is a car, truck, or other ground vehicle that can navigate roads and traffic without a human at the controls, using sensors,…
- Autonomy
- Autonomy is a robot's ability to make decisions and take actions on its own, without moment-to-moment human instruction.
B
- Backdrivability
- Backdrivability is how easily you can move a robot's joint by pushing on it from the outside — a property that decides whether a robot can feel and…
- Backlash
- Backlash is the tiny bit of slack between gear teeth that makes a joint lag when it reverses direction — a small mechanical imperfection that quiet…
- Ball Screw in Robotics
- Ball screws convert rotary motion into linear motion with very low friction.
- Bang-Bang Control in Robotics
- Bang-bang (on/off) control switches the actuator between two extremes — fully on or fully off — based on whether the output is above or below the setpoint.
- Battery and power systems
- A robot's power system converts stored energy into the right voltages and currents for every component on board — motors, computers, and sensors al…
- Battery Management System (BMS) in Robotics
- A BMS monitors and protects multi-cell battery packs — measuring voltage, current, temperature, and cutting off on faults.
- Bearing
- A bearing lets one part rotate or slide against another with minimal friction — the humble component that makes every robot joint, wheel, and spinn…
- Behavior Cloning
- Behavior cloning teaches a robot a skill by training it to copy a human's demonstrations — the simplest form of imitation learning, powerful and di…
- Behaviour cloning
- Behaviour cloning is a machine-learning technique where a robot is trained to copy an expert's actions as closely as possible by treating recorded…
- Behaviour Trees
- A behaviour tree is a way of organising a robot's decisions as a tree of small tasks, where a parent node repeatedly asks its children to run and e…
- Bellman Equation
- The Bellman equation is the recursive rule that says the value of a situation equals the reward now plus the value of what comes next — the single…
- Bin Picking
- Bin picking is a robot grabbing objects one by one from a jumbled pile in a container — the hard, high-value manipulation problem behind e-commerce…
- Bipedal robot
- A bipedal robot is a robot that walks on two legs, like a person.
- Bomb-Disposal Robot
- A bomb-disposal robot lets an operator inspect and neutralize explosives from a safe distance — a teleoperated machine that has saved countless liv…
- Brain-Computer Interface (BCI) in Robotics
- Brain-Computer Interfaces let people control robots and prosthetics with thoughts.
- Bricklaying Robot
- A bricklaying robot lays bricks or blocks automatically — buttering mortar and placing units with precision — one of the first serious attempts to…
- Brushless DC Motor (BLDC) in Robotics
- A BLDC motor uses electronic commutation instead of mechanical brushes.
- Bundle Adjustment
- Bundle adjustment is the optimization that refines camera poses and 3D points together so they best explain every image observation — the accuracy…
- Buoyancy Control
- Buoyancy control lets an underwater robot rise, sink, and hover by adjusting its effective density — the depth-keeping mechanism behind AUVs, glide…
C
- C++ for Robotics
- C++ is the language for performance-critical robotics — drivers, real-time control, and ROS2 nodes that can't afford Python's overhead.
- Cable-Driven Mechanism
- A cable-driven mechanism transmits force through tensioned cables instead of rigid gears — letting motors sit far from the moving parts, so robots…
- Camera (machine vision)
- A machine-vision camera is a sensor that turns light into digital images a robot can interpret.
- Camera Calibration
- Camera calibration measures a camera's exact lens and geometry so a robot can turn pixels into real directions and distances — the unglamorous firs…
- CAN Bus
- CAN bus is the rugged two-wire network that lets a robot's many microcontrollers, motors, and sensors talk over a single shared line — the reliable…
- Canadarm — The ISS Robotic Arm
- Canadarm and Canadarm2 are the most famous space manipulators.
- Capacitive Sensor
- A capacitive sensor detects nearby objects — especially a human hand — by sensing tiny changes in an electric field, giving robots touchless proxim…
- Capsule Endoscopy
- Capsule endoscopy uses a swallowable camera robot to image the GI tract.
- Capture Point
- The capture point is where a legged robot must step to come to a complete stop — the modern balance concept behind push recovery and dynamic walkin…
- Carbon-Fibre Composites in Robotics
- Carbon-fibre composites give robots incredibly high strength-to-weight ratios.
- Cascade Control in Robotics
- Cascade control uses two nested feedback loops — an inner fast loop for the actuator and an outer slow loop for the high-level goal — to deliver fa…
- CE Marking for Robots
- CE marking is the European Union conformity mark that allows a robot or robotic product to be sold in the EU.
- CE Marking for Robots
- CE marking is required to sell most robots and machinery into the European Union.
- Center of Mass
- The center of mass is the single point where a robot's weight effectively acts — the thing a legged or humanoid robot must constantly manage over i…
- Center of Pressure
- The center of pressure is the single point on a robot's foot where the ground reaction force effectively acts — a measurable balance indicator that…
- Central Pattern Generator (CPG)
- A central pattern generator is a network of coupled oscillators that produces rhythmic motion — the biology-inspired engine that gives legged and s…
- Chassis (robot)
- A robot's chassis is its structural skeleton — the rigid frame that holds all components in their correct positions and transmits forces between th…
- Cleaning Robots
- Cleaning robots autonomously sweep, mop, scrub, or vacuum floors at home and in commercial spaces.
- Closed-loop control
- Closed-loop control uses sensor feedback to continuously compare what a robot is doing with what it should be doing, then adjusts its output to clo…
- Cobot (collaborative robot)
- A cobot is a robotic arm designed to share workspace safely with humans.
- colcon
- colcon is the build tool that compiles a ROS 2 workspace — resolving dependencies between many packages and building them in the right order, the e…
- Collaborative robot
- A collaborative robot is any robot designed to work alongside humans in the same space, sensing their presence and adjusting its behaviour to stay…
- Collaborative Robot (Cobot) Safety
- Cobots work alongside humans without cages.
- Collision Checking
- Collision checking tests whether a robot in a given configuration hits itself or the world — the fast, constantly-called routine that keeps every m…
- Companion Robot
- A companion robot provides social interaction and emotional support rather than physical work — engaging people through conversation, presence, and…
- Complementary Filter
- A complementary filter blends two sensors that fail in opposite ways: it trusts a gyroscope for fast changes and an accelerometer for the long-term…
- Compliant Assembly
- Compliant assembly lets a robot fit parts together by being deliberately soft — yielding to contact so pieces self-align instead of jamming — the k…
- Compliant Mechanism
- A compliant mechanism gets its motion by flexing rather than from joints and bearings — a single elastic part that bends to move, giving robots hin…
- Computed Torque Control
- Computed torque control uses a robot's own physics model to cancel its nonlinear dynamics — gravity, inertia, and coupling — so a simple linear con…
- Computer vision (for robots)
- Computer vision is how a robot makes sense of what its camera sees.
- Configuration Space
- Configuration space is the abstract space of all the poses a robot can be in — the arena where motion planning actually happens, turning "don't hit…
- Construction Robots
- Construction robots automate bricklaying, 3D printing, surveying, and demolition.
- Contact Mechanics
- Contact mechanics is the physics of what happens when surfaces touch — the forces, deformation, and friction at the interface — and the single hard…
- Convolutional Neural Network (CNN) in Robotics
- CNNs are the workhorse neural architecture for robot vision.
- Coriolis & Centrifugal Forces
- Coriolis and centrifugal forces are the velocity-dependent effects that make a robot's joints tug on each other as they move — the coupling term in…
- Costmap in ROS2 Navigation
- A costmap is a 2D grid where each cell holds a cost reflecting how desirable it is to traverse.
- Costmap Layers
- A layered costmap builds a robot's picture of where it's safe to drive by stacking sources — static map, live obstacles, and safety inflation — int…
- Coulomb Friction
- Coulomb friction is the classic model where sliding resistance is proportional to the pressing force — the simple rule (and its 'friction cone') th…
- Curriculum Learning
- Curriculum learning trains a robot on progressively harder tasks — easy first, then building up — the way people learn, which makes hard skills lea…
D
- D* (Dynamic A*)
- D and D Lite are replanning algorithms that efficiently repair a path when the map changes — instead of planning from scratch each time an obstacle…
- da Vinci Surgical Robot
- The da Vinci Surgical System is the most-used surgical robot worldwide — 10M+ operations across cardiac, urological, gynaecological, and abdominal surgery.
- Damping Ratio
- Damping ratio measures how quickly a system's oscillations die out — the single number that decides whether a robot joint settles crisply, overshoo…
- DC motor
- A DC motor converts direct-current electricity into continuous spinning motion.
- DDS Middleware
- DDS is the industrial-grade messaging layer under ROS 2 — a decentralized, real-time publish–subscribe system with no central master, built-in disc…
- Dead Reckoning
- Dead reckoning is estimating where a robot is by adding up how far it has moved from a known start — cheap, always-available, and the reason robots…
- Deep learning
- Deep learning is a branch of machine learning that stacks many layers of neural-network calculations on top of each other, allowing computers to le…
- Deep Q-Network (DQN)
- The Deep Q-Network fused Q-learning with deep neural networks, learning to act straight from raw pixels — the 2013 breakthrough that kicked off the…
- Degrees of freedom
- Degrees of freedom (DOF) count the independent ways a robot or object can move.
- Delivery Robot
- Delivery robots autonomously transport goods between locations — indoors through hospitals and hotels, or outdoors across campuses and cities — aut…
- Denavit-Hartenberg Parameters
- Denavit-Hartenberg (DH) parameters are a compact way to describe robot arm geometry.
- Depth Camera in Robotics
- A depth camera captures the distance to objects in addition to colour.
- Depth sensor
- A depth sensor measures the distance to every visible point in a scene simultaneously, producing a 2D image where each pixel encodes depth rather than colour.
- Dexterous Manipulation
- Dexterous manipulation is the frontier skill of robot hands — not just holding an object but skillfully controlling it with the fingers, turning, r…
- Differential Drive in Robotics
- Differential drive uses two independently driven wheels and a passive caster to steer a mobile robot.
- Diffusion Models in Robotics
- Diffusion models generate data — including robot trajectories — by learning to reverse a noising process.
- Diffusion Policy
- A diffusion policy is a robot control approach that borrows image-generation AI to generate actions — turning noise into smooth, multi-modal motion.
- Digital twin
- A digital twin is a live virtual replica of a physical object or system — updated continuously by real-world sensor data — that engineers use to si…
- Docker for Robotics
- Docker containerises ROS2 stacks for reproducible robot deployment.
- Domain Randomization
- Domain randomization trains robots in simulation with wildly varied physics and visuals, so the real world looks like just another variation — the…
- DRDO Robotics
- DRDO builds India's defence robots — Daksh bomb-disposal, MUNTRA unmanned tanks, surveillance UAVs, and more.
- Drone (UAV)
- A drone, or unmanned aerial vehicle (UAV), is an aircraft that flies without a pilot on board — guided by remote control, onboard autopilot, or a c…
- Drone (UAV)
- A drone is a flying robot — usually four or more electric rotors arranged around a central body, controlled by a tiny computer doing math 1,000 tim…
- Dynamic Window Approach (DWA)
- The dynamic window approach is a fast local planner that picks a robot's next velocity by simulating short candidate motions and scoring them — the…
- Dynamics (robotics)
- Dynamics extends kinematics by adding forces and torques to the picture — it explains why a robot moves, not just how.
E
- Economics of Automation
- Automation economics studies the cost-benefit of robots vs human labour, the impact on jobs and wages, and the macro effect on national productivity.
- Edge AI
- Edge AI means running artificial-intelligence computations directly on the device — a robot, a camera, a smartphone — rather than sending data to a…
- Edge Computing
- Running computation near the data source — on the robot itself, not in the cloud.
- Educational robotics
- Educational robotics uses robots as learning tools — teaching programming, engineering, problem-solving, and teamwork by giving students a physical…
- EKF-SLAM
- EKF-SLAM was the original solution to simultaneous localization and mapping — one big Kalman filter tracking the robot's pose and every landmark at once.
- Elder-Care Robot
- Elder-care robots help older adults live independently — reminding, monitoring, assisting with mobility, and providing companionship — a socially v…
- Electromagnetic Interference (EMI)
- Electromagnetic interference is the electrical noise that motors and switching power dump onto a robot's signals — the invisible gremlin behind gli…
- Electronic Speed Controller (ESC)
- An electronic speed controller is the driver that turns a battery's DC power into the precisely timed pulses a brushless motor needs — the small bo…
- Embedded system
- An embedded system is a computer built into a device to perform one specific job — invisible to the user but essential to the machine's function.
- Embedded Systems in Robotics
- Embedded systems are tiny, dedicated computers inside robots that run real-time code — motors, sensors, safety logic.
- Embodied AI
- Embodied AI is artificial intelligence that learns and acts through a physical body — a robot that understands the world not just by processing dat…
- Emergency Stop (E-Stop)
- An emergency stop is the big red button that halts a robot instantly and reliably — the last line of safety defense, engineered to work even when t…
- Encoder (rotary)
- A rotary encoder is a sensor that counts how far a shaft has turned.
- End effector
- The end effector is the business end of a robot — the gripper, welder, suction cup, or any other tool attached to the tip of the arm.
- End-Effector Safety in Robotics
- The end-effector — gripper, tool, or weld torch — is often the highest-risk component on a robot.
- Epipolar Geometry
- Epipolar geometry is the math relating two camera views of the same scene — it constrains where a point in one image can appear in the other, the f…
- Equations of Motion
- The equations of motion are the physics that link a robot's joint torques to how it actually accelerates — the model behind realistic simulation, m…
- ESP32 for Robotics
- The ESP32 is a Wi-Fi/Bluetooth-enabled microcontroller.
- EtherCAT
- EtherCAT is a real-time industrial Ethernet protocol that synchronizes dozens of motors and sensors with microsecond precision — the high-speed ner…
- Euler Angles
- Euler angles describe an orientation as three simple rotations — roll, pitch, and yaw.
- Event Camera
- An event camera reports only the pixels that change, microsecond by microsecond, instead of full frames — giving robots blur-free vision at extreme…
- Exoskeleton
- An exoskeleton is a wearable robotic structure that attaches to the outside of a human body, augmenting the wearer's strength, endurance, or moveme…
- Extended Kalman Filter in Robotics
- The Extended Kalman Filter (EKF) handles nonlinear motion and sensor models by local linearisation.
F
- Factor Graph
- A factor graph is the flexible mathematical structure behind modern state estimation — it represents unknowns and the measurements that constrain t…
- Feature Extraction in Robotics Vision
- Feature extraction reduces raw sensor data — especially images — into compact, useful descriptors.
- Feature Matching
- Feature matching finds which points in one image correspond to points in another — the fundamental step that links images together for SLAM, 3D rec…
- Feedback
- Feedback is the mechanism by which a system uses information about its own output to correct its future behaviour.
- Feedback Linearization
- Feedback linearization is a control trick that cancels a system's nonlinearities with math, transforming a hard nonlinear robot into an easy linear…
- Feedback Loop in Robotics
- A feedback loop continuously measures a robot's output and adjusts its input to reach a desired target.
- Feedforward Control in Robotics
- Feedforward control predicts the disturbance or load on a robot and pre-compensates for it, instead of waiting to react.
- Few-Shot Learning
- Few-shot learning lets a model pick up a new task or object from just a handful of examples — the ability robots need to adapt on the fly to new it…
- Field-Oriented Control (FOC) in Robotics
- FOC is the dominant control technique for BLDC motors — turning a 3-phase motor into a smooth, torque-controllable actuator.
- Finite State Machine
- A finite state machine is a model in which a robot is always in exactly one state out of a fixed list — charging, delivering, idle — and moves to a…
- Firefighting Robots
- Firefighting robots enter burning buildings, fight chemical fires, and protect human firefighters.
- Fixed-Wing UAV
- A fixed-wing UAV flies like an airplane — using wings for efficient lift rather than spinning rotors — trading a drone's hover for far greater rang…
- Flexure Joint
- A flexure joint provides rotation or motion by bending a thin elastic section instead of using a pin and bearing — giving robots and precision inst…
- Force Closure
- Force closure is the condition for a truly secure grasp — the fingers can resist a push or twist from any direction using friction and contact forc…
- Force Sensor in Robotics
- Force sensors measure pushing, pulling, or twisting forces.
- Force-torque sensor
- A force-torque sensor measures the pushing, pulling, and twisting forces acting on a robot's wrist or tool.
- Form Closure
- Form closure is a grasp so secure it holds an object purely by geometry — the contacts cage it so it can't move even if the surfaces were frictionl…
- Formation Control
- Formation control keeps a group of robots moving in a coordinated shape — a drone show, a convoy, a search line — using simple local rules so the w…
- Forward kinematics
- Forward kinematics calculates where a robot's end-effector ends up in space given a specific set of joint angles — the straightforward direction of…
- Foundation models in robotics
- Foundation models are large AI systems trained on vast, diverse datasets that can be adapted to many tasks — and researchers are now applying them…
- Four-Bar Linkage
- A four-bar linkage is the simplest and most useful mechanical linkage — four rigid bars and four joints arranged so one input motion produces a spe…
- FPGA
- An FPGA is a chip whose digital circuitry you can reconfigure — giving robots custom, massively parallel hardware for tasks like motor control, sen…
- FPGA
- An FPGA is a chip you can rewire in software — instead of a fixed set of circuits, it contains thousands of programmable logic blocks that can be c…
- Friction Model
- A friction model is the math a robot uses to predict how surfaces resist sliding — deceptively hard to get right, and a major reason grasps, walkin…
- Functional Safety in Robotics
- Functional safety ensures a robot's safety-related electronics fail in a safe state.
- Future of Work with Robots
- How robots reshape what humans do at work.
- Fuzzy Logic in Robotics
- Fuzzy logic replaces strict true/false rules with degrees of truth.
G
- Gain Scheduling
- Gain scheduling runs different controller tunings for different operating conditions and blends between them — a practical way to control a robot w…
- Gait
- A gait is the pattern and timing of how a legged robot moves its feet — walk, trot, gallop — and choosing and generating the right gait is the foun…
- Gazebo Simulator
- Gazebo is the physics simulator that lets roboticists test a full robot — sensors, motors, and world — in software before touching hardware, saving…
- Gear Ratio
- Gear ratio is the trade a robot makes between speed and force — gearing a fast, weak motor down to turn slower but push far harder, which is why al…
- Gear Train
- A gear train is a series of meshing gears that transmits and transforms rotation — multiplying torque, changing speed, or redirecting motion, the m…
- Gimbal Lock
- Gimbal lock is the orientation glitch where two rotation axes line up and a system loses a degree of freedom — the failure that plagues Euler angle…
- GPS / GNSS
- GPS is the American satellite navigation system; GNSS is the broader term for all such systems worldwide.
- GPS Module in Robotics
- A GPS module gives a robot its global position via satellites.
- Graceful degradation
- Graceful degradation is the ability of a system to continue functioning at a reduced level when part of it fails, rather than failing completely.
- Grasp Planning
- Grasp planning is how a robot decides where and how to grip an object so it holds securely — the make-or-break step between seeing an object and ac…
- Grasp Quality
- Grasp quality is a score that rates how good a grip is — not just whether it holds, but how much disturbance it can survive — letting a robot rank…
- Gripper
- A gripper is the end-effector a robot uses to grasp, hold, and release objects.
- Ground Loop
- A ground loop happens when a circuit has more than one path to ground at different voltages, letting stray currents flow through signal lines — a s…
- Ground Reaction Force
- The ground reaction force is the push the ground exerts back on a robot's foot — the only external force a legged robot can use to move and balance…
- Gyroscope in Robotics
- A gyroscope measures angular velocity around an axis.
H
- H-Bridge Circuit in Robotics
- An H-bridge is a circuit of four transistors that lets a DC motor spin in either direction.
- Hall-Effect Sensor in Robotics
- Hall-effect sensors detect magnetic fields.
- Haptic Feedback in Robotics
- Haptic feedback lets robots transmit touch and force to humans (or vice versa).
- Hardware-in-the-Loop (HIL)
- Hardware-in-the-loop testing connects a robot's real controller to a simulated world — so the actual electronics and software can be tested safely…
- Harmonic Drive in Robotics
- Harmonic drive gears deliver huge reduction with near-zero backlash in a compact package.
- Harvesting Robot
- A harvesting robot picks fruit and vegetables autonomously — using vision to find ripe produce and gentle grippers to pick it — tackling farm labor…
- Hexapod Robot
- A hexapod robot walks on six legs — insect-style — which makes it naturally stable, able to keep moving even if a leg fails, and a favorite for rou…
- Holonomic Constraint
- A holonomic constraint restricts a robot's positions directly — and a holonomic robot is one that can move in any direction it pleases.
- Holonomic Drive in Robotics
- Holonomic drives allow a robot to translate in any direction without first rotating.
- Home robotics
- Home robotics covers robotic devices designed for domestic environments — cleaning, lawn mowing, companionship, and household assistance — bringing…
- Homogeneous Transformation
- A homogeneous transformation is the 4×4 matrix that packs rotation and translation into one object — the single tool robots use to describe full po…
- Human-robot interaction (HRI)
- Human-robot interaction is the study of how people and robots communicate, collaborate, and affect each other — encompassing interfaces, trust, saf…
- Humanoid robot
- A humanoid robot is a robot built in the rough shape of a human body — two legs, two arms, a head with cameras.
- Hybrid A*
- Hybrid A* is a path planner for car-like robots that searches over feasible, drivable motions instead of grid cells — producing smooth paths a vehi…
- Hydraulic actuator
- A hydraulic actuator uses pressurised liquid — usually oil — to produce force far beyond what any electric motor of the same size could manage, mak…
I
- I2C
- I2C is a simple two-wire bus that lets a microcontroller talk to many small sensors and chips using just two shared lines — the everyday plumbing c…
- IIT Bombay Robotics
- IIT Bombay's robotics ecosystem spans CSRE, electrical engineering, mechanical, and the Wadhwani Centre.
- IIT Madras Robotics
- IIT Madras leads Indian robotics research in surgical, agricultural, and humanoid robotics.
- IIT Robotics Research — Indian Institutes of Technology
- An overview of robotics research labs at India's leading engineering institutes — IIT Bombay, IIT Madras, IIT Delhi, IIT Kanpur, IISc.
- Image Segmentation in Robotics
- Image segmentation labels every pixel in an image with a class — essential for robot grasping, autonomous driving, and crop-disease detection.
- Imitation learning
- Imitation learning is a training approach where a robot learns to behave by watching demonstrations of a task performed by a human or another exper…
- Impedance Control in Robotics
- Impedance control makes a robot behave like a programmable spring-damper — essential for safe contact, manipulation, and human-robot interaction.
- IMU (Inertial Measurement Unit)
- An IMU is a chip that measures how fast something is accelerating and rotating.
- IMU Sensor in Robotics
- An IMU (Inertial Measurement Unit) combines accelerometer, gyroscope, and often a magnetometer to estimate orientation and motion in 3D space.
- In-Hand Manipulation
- In-hand manipulation is repositioning an object within the hand without putting it down — the finger gymnastics that let a robot reorient a part to…
- India Robot Density
- India's industrial robot density is far below leaders such as South Korea, which has over 1,000 robots per 10,000 manufacturing workers.
- India's National Robotics Mission
- The National Robotics Mission (NRM) is a mechanism proposed in India's draft National Strategy on Robotics to coordinate robotics research, testing…
- Industrial robot
- An industrial robot is a fixed robotic arm bolted to a factory floor that does one job — welding a car door, soldering a chip, assembling a pen — m…
- Inertia Tensor
- The inertia tensor describes how a robot's mass is distributed, and therefore how hard it is to spin about each axis — the rotational counterpart o…
- Infrared sensor
- An infrared sensor detects invisible light just beyond the red end of the visible spectrum.
- Ingress Protection (IP) Rating
- An IP rating is the two-digit code that says how well a robot's enclosure keeps out dust and water — the spec that decides whether a robot can work…
- Inspection Robot
- Inspection robots go where people can't or shouldn't — inside tanks, up towers, through tunnels — carrying sensors to check for damage, corrosion,…
- Instance Segmentation
- Instance segmentation outlines each individual object in an image pixel-by-pixel and tells them apart — not just 'these pixels are cups' but 'this…
- Inverse kinematics
- Inverse kinematics is the math a robot uses to figure out which joint angles will put its hand exactly where you want it.
- Inverse Reinforcement Learning (IRL)
- Inverse reinforcement learning flips the usual problem: instead of learning behavior from a reward, it infers the hidden reward from expert behavio…
- ISO 10218 — Industrial Robot Safety Standard
- ISO 10218 is the international standard for industrial robot safety.
- ISO/TS 15066 — Cobot Safety Standard Guide
- ISO/TS 15066 defines safe collaborative-robot operation, including biomechanical force/pressure limits humans can tolerate during contact.
- ISRO Chandrayaan Pragyan Rover
- Pragyan is ISRO's lunar rover from Chandrayaan-3 — India's first robot on the Moon.
- ISRO Robotics — India's Space Robotics Programme
- How the Indian Space Research Organisation (ISRO) builds space robots — Pragyan rover, Vyommitra humanoid, the Gaganyaan mission, and on-orbit servicing.
- Iterative Closest Point (ICP)
- ICP is the algorithm that aligns two 3D scans by repeatedly matching nearest points and nudging one cloud onto the other — the core of lidar odomet…
J
- Jacobian
- The Jacobian is the matrix that links how a robot's joints move to how its hand moves — the bridge between joint space and the real world, and the…
- Jerk
- Jerk is the rate of change of acceleration — the quantity that decides whether a robot's motion feels smooth or slams.
- Joint (robot)
- A robot joint is the connection point between two rigid segments where controlled movement occurs — it is the mechanical equivalent of a knee, shou…
- Joint degrees of freedom
- A joint's degrees of freedom is the count of independent ways it can move — one for a simple hinge, up to six for an unconstrained connection — and…
K
- Kalman Filter
- The Kalman filter optimally fuses noisy sensor data to estimate a robot's true state.
- Kinematic Coupling
- A kinematic coupling is a way to join two parts so they mate in the exact same position every time — using precisely six contact points to remove a…
- Kinematics
- Kinematics describes the geometry of a robot's motion — where each part is and where it will be — without asking why it moves.
L
- Lagrangian Mechanics
- Lagrangian mechanics derives a robot's equations of motion from energy alone — a systematic recipe that turns any complicated linkage into its dyna…
- Laparoscopic Robot
- A laparoscopic robot lets a surgeon operate through tiny incisions with tremor-filtered, wristed instruments guided from a console — the technology…
- Large Language Models for Robotics
- LLMs let robots understand natural-language instructions and reason about tasks.
- Last-Mile Delivery Robot
- Last-mile delivery robots carry packages and food the final stretch to your door — small sidewalk rovers and drones tackling the most expensive, ha…
- Last-Mile Delivery Robots
- Last-mile delivery robots deliver food, groceries, and packages from a local hub to your door.
- Lawn-Mowing Robot
- A robotic lawn mower cuts grass on its own — historically wandering within a buried boundary wire, now increasingly navigating by GPS and vision —…
- Levels of Autonomy
- Levels of autonomy classify how much a machine does on its own versus how much a human does — from full manual control to full self-driving — a fra…
- Lidar
- Lidar is a sensor that measures distance by firing invisible laser pulses and timing how long they take to bounce back.
- LIDAR SLAM in Robotics
- LIDAR SLAM uses laser-scan data to map an environment and localise a robot.
- Light Curtain in Robotics Safety
- A light curtain is an array of safety-rated optical beams.
- Linear actuator
- A linear actuator converts energy into straight-line push-or-pull motion rather than rotation, giving a robot the ability to extend, retract, lift,…
- Link (robot mechanism)
- A link is the rigid segment between two joints in a robot — the structural member that transmits forces and defines the geometry of the kinematic c…
- LiPo Battery in Robotics
- Lithium polymer (LiPo) batteries deliver high energy density and high discharge rates — the standard for drones, racing robots, and high-performanc…
- Load Cell
- A load cell is a force sensor that turns weight or push into a precise electrical reading — how robots weigh payloads, verify a grip, and detect co…
- Localisation
- Localisation is the process by which a robot figures out where it is — its position and orientation — within a known or partially known environment…
- Loop Closure
- Loop closure is the moment a robot recognizes "I've been here before" — the single most important event in SLAM, because it lets the system cancel…
- LQR Controller (Linear Quadratic Regulator)
- An LQR controller (linear quadratic regulator) is a feedback controller that computes the optimal gains for the weights you choose: you say how muc…
M
- Machine Learning
- A family of algorithms that learn patterns from data instead of following explicit instructions — the foundation of modern AI in robotics.
- Make in India for Robotics
- How the Make in India policy supports the Indian robotics industry — tax breaks, PLI schemes, and domestic manufacturing of robots.
- Manipulability Ellipsoid
- A manipulability ellipsoid is a visual measure of how easily a robot arm can move or exert force in each direction at a given pose — showing where…
- Manufacturing automation
- Manufacturing automation is the use of machines, software, and control systems to perform production tasks with little or no direct human input, ma…
- Mapping (robotics)
- Mapping is the process by which a robot builds a representation of its environment from sensor data — turning raw readings from lidar, cameras, or…
- Markov Decision Process (MDP)
- A Markov Decision Process is the mathematical frame for sequential decision-making — states, actions, rewards, and transitions — and the formal pro…
- Mecanum Wheel in Robotics
- Mecanum wheels allow a robot to translate sideways without turning.
- Mechanical Metamaterial (in Robotics)
- Mechanical metamaterials are engineered structures whose behavior comes from their geometry, not their material — enabling robots and structures th…
- Mechatronics
- Mechatronics is the engineering discipline that integrates mechanics, electronics, and computing into a single unified design — rather than bolting…
- Medical robotics
- Medical robotics is the application of robotic systems to surgery, rehabilitation, diagnostics, and patient care, letting doctors operate with grea…
- micro-ROS
- micro-ROS brings ROS 2 down onto tiny microcontrollers — the small, cheap chips that run motors and sensors — so the smallest parts of a robot can…
- Microcontroller
- A microcontroller is a tiny self-contained computer — processor, memory, and input/output pins all on a single chip — designed to run one specific…
- Mining Robot
- Mining robots automate digging, hauling, and drilling underground and in open pits — from driverless haul trucks to autonomous drills — making one…
- Mobile robot
- A mobile robot is any robot that can move through its environment under its own power — on wheels, tracks, or legs — rather than staying bolted to one spot.
- Model Predictive Control (MPC) in Robotics
- MPC predicts future robot behaviour over a horizon and solves an optimisation problem to choose the best control input — used in self-driving cars…
- Model-Based Reinforcement Learning
- Model-based reinforcement learning learns a model of how the world works and uses it to plan or train — far more sample-efficient than learning by…
- Momentum
- Momentum is mass times velocity — a conserved quantity that governs impacts, throws, and why a moving robot can't stop instantly, central to legged…
- Monte Carlo Localization
- Monte Carlo localization figures out where a robot is on a known map by scattering thousands of guesses and letting sensor readings vote the wrong…
- Motion planning
- Motion planning determines the precise sequence of joint angles, wheel velocities, or body configurations a robot must pass through to move safely…
- Motor Driver in Robotics
- A motor driver is an electronic circuit that lets a low-power microcontroller control a high-power motor.
- MoveIt (Motion Planning for Robot Arms)
- MoveIt is free, open-source software for ROS that moves robot arms: you give it a target for the gripper and it works out the joint angles, plans a…
- MoveIt 2 in ROS 2: Architecture, Planners and Setup
- MoveIt 2 is the ROS 2 version of MoveIt, the motion-planning framework for robot arms.
- MuJoCo
- MuJoCo is a fast, accurate physics simulator built for robotics and machine learning — prized for its stable contact modeling, and now free and ope…
N
- Natural Frequency
- Natural frequency is the rate at which a system wants to oscillate when disturbed — the number that decides how fast a robot joint responds, when i…
- Natural-language robotics
- Natural-language robotics is the field of research and engineering that enables robots to understand and act on instructions given in ordinary spok…
- Nav2
- Nav2 is the ROS 2 navigation stack — the batteries-included framework that takes a mobile robot from "go to that point" to actually driving there s…
- Nav2 (ROS2 Navigation Stack)
- Nav2 is the ROS2 navigation stack for mobile robots — planners, controllers, behaviour trees, recovery.
- Neural network
- A neural network is a mathematical system loosely inspired by the brain, made of layers of simple calculations that together learn to recognise pat…
- Neural Networks for Robotics
- How robots use neural networks for perception, control, and decision-making — from object detection to end-to-end driving policies.
- Neural Radiance Field (NeRF)
- A NeRF learns a 3D scene from a set of photos by training a neural network to render it from any viewpoint — a breakthrough in 3D reconstruction no…
- NITI Aayog and Indian Robotics Policy
- NITI Aayog is the Government of India's policy think tank — its reports and recommendations shape India's robotics, AI, and electronics-manufacturing strategy.
- Nonholonomic Constraint
- A nonholonomic constraint limits how a robot can move without limiting where it can ultimately go — the reason a car can reach any parking spot but…
- Nonprehensile Manipulation
- Nonprehensile manipulation moves objects without grasping them — pushing, sliding, toppling, or throwing — extending what a robot can do beyond wha…
- NVIDIA Isaac Sim
- Isaac Sim is NVIDIA's GPU-accelerated, photorealistic robot simulator — able to run thousands of robots in parallel and generate lifelike sensor da…
- NVIDIA Jetson
- NVIDIA Jetson is a family of compact computers with built-in GPUs that run AI and computer vision on the robot itself — the go-to onboard 'brain' f…
- NVIDIA Jetson Nano in Robotics
- The Jetson Nano is a $99 GPU-equipped single-board computer made for robotics AI workloads — perfect entry point for vision and deep learning.
- NVIDIA Jetson Orin in Robotics
- The Jetson Orin family delivers up to 275 TOPS of AI compute on a robot-friendly module.
O
- Object detection
- Object detection is an AI task where a computer identifies what objects are present in an image and draws a bounding box around each one — enabling…
- Occupancy grid
- An occupancy grid is a map of the environment divided into a regular grid of cells, where each cell stores a probability value indicating whether t…
- Odometry
- Odometry is a method of estimating a robot's current position by tracking how far and in what direction it has moved from a known starting point, u…
- Offline Reinforcement Learning
- Offline reinforcement learning trains a policy purely from a fixed dataset of past experience — no live trial-and-error — letting robots learn from…
- Omnidirectional Robot
- An omnidirectional robot can move in any direction — forward, sideways, diagonally — and rotate at the same time, thanks to special wheels.
- Open-loop control
- Open-loop control means a robot sends a command and assumes it worked, with no sensor feedback to check.
- Optical Flow
- Optical flow is the pattern of apparent motion between video frames — the per-pixel arrows showing which way things moved.
- Optical Flow Sensor
- An optical flow sensor is a small downward-facing camera that measures how fast the ground slides past it.
- ORB Features
- ORB is a fast, free way to find and describe distinctive points in an image — the feature detector behind real-time visual SLAM and tracking on mod…
P
- Palletizing Robot
- A palletizing robot stacks boxes onto pallets in tight, stable patterns — a high-throughput, heavy-duty application that automates one of the most…
- Panoptic Segmentation
- Panoptic segmentation labels every pixel of a scene at once — separating countable objects individually while classifying background regions — givi…
- Parallel Robot
- A parallel robot connects its end-effector to the base through multiple parallel kinematic chains.
- Parallel-Jaw Gripper
- A parallel-jaw gripper is the two-finger pincer that most robot arms use to grab things — simple, reliable, and precise, it's the default end-effec…
- Particle Filter in Robotics
- Particle filters represent uncertainty with thousands of weighted samples — perfect for nonlinear, multi-modal robot localisation problems like AMCL.
- Path planning
- Path planning is the process of finding a route through space from a starting point to a goal, avoiding obstacles along the way — the geometric que…
- PCB (printed circuit board)
- A PCB is a flat board of insulating material threaded with copper tracks that connect electronic components without wires — the backbone of almost…
- Peg-in-Hole
- Peg-in-hole is the classic robotic assembly benchmark — inserting a part into a tight opening — that looks trivial but demands force sensing and co…
- Physics Engine
- A physics engine is the software that simulates how objects move, collide, and respond to forces — the core of every robot simulator, and the diffe…
- Pick and Place
- Pick and place is the fundamental robot task — grab an object here, put it there — that underlies most industrial automation, from assembly lines t…
- PID controller
- A PID controller is a feedback algorithm that continuously corrects a robot's behaviour by measuring the gap between where it is and where it shoul…
- Piezoelectric Actuator in Robotics
- Piezoelectric actuators convert voltage into nanometre-precise motion.
- Pipeline Inspection Robot
- Pipeline inspection robots travel inside pipes — from city sewers to oil and gas lines — surveying for cracks, corrosion, and blockages in spaces n…
- Planetary Gearbox in Robotics
- Planetary gearboxes deliver high torque and compact size for robot motors.
- Planetary Landers
- Planetary landers safely deliver spacecraft to the surface of moons, planets, and asteroids.
- Planetary Rover
- A planetary rover explores other worlds — driving across Mars or the Moon to study terrain and search for signs of life — the pinnacle of autonomou…
- Pneumatic actuator
- A pneumatic actuator uses compressed air to produce fast, forceful movement.
- Point cloud
- A point cloud is a collection of millions of individual 3D coordinates, each one marking a surface that a laser or depth camera measured, together…
- Pole Placement
- Pole placement is a control design method that chooses feedback gains to put a system's poles exactly where you want them — directly dialing in how…
- Policy Gradient
- Policy-gradient methods learn a robot's behavior directly — nudging the policy toward actions that earned more reward — and are the family behind t…
- Pose Estimation in Robotics
- Pose estimation predicts the 3D position and orientation of objects, people, or the robot itself.
- Pose-Graph Optimization
- Pose-graph optimization is how modern SLAM builds a consistent map: treat every robot pose as a node and every measurement as a constraint, then so…
- Potential Field Method
- The potential field method steers a robot by pretending the goal pulls it like gravity and obstacles push it away like magnets — a fast, intuitive…
- Power and Force Limiting
- Power and force limiting makes a robot inherently gentle — capped in speed and force so that even if it touches a person, it can't hurt them — the…
- Precision Agriculture
- Precision agriculture uses sensors, GPS, and AI to apply seed, water, fertiliser, and herbicide only where needed — maximising yield while cutting cost.
- Probabilistic Roadmap (PRM)
- A probabilistic roadmap pre-builds a network of collision-free connections through a space, so a robot can quickly find paths between many start-go…
- Prosthetic Hand
- A robotic prosthetic hand restores grasping to people who've lost a hand — reading signals from the residual muscles to drive motorized fingers, a…
- Proxemics (in Robotics)
- Proxemics is the study of personal space — and in robotics, teaching robots to respect it, so they move and position themselves in ways that feel c…
- Proximal Policy Optimization (PPO)
- PPO is the workhorse reinforcement-learning algorithm of modern robotics — it improves a policy in stable, bounded steps, which is why it reliably…
- Publish–Subscribe
- Publish–subscribe is the messaging pattern where senders broadcast data on named channels and receivers listen without either knowing about the oth…
- Push Recovery
- Push recovery is how a legged robot keeps its balance when shoved — the ankle, hip, and stepping reflexes that let a humanoid take a hit and stay u…
- PWM (Pulse Width Modulation)
- PWM is a trick for making a digital chip — which can only turn things fully on or fully off — behave as if it's dimming a light or running a motor at half-speed.
- PWM Signal in Robotics
- Pulse-Width Modulation (PWM) is the standard technique for controlling motor speed, LED brightness, and servo position.
- PyBullet
- PyBullet is a free, easy-to-use physics simulator with a Python interface — the accessible on-ramp to robot simulation and reinforcement learning,…
- Python for Robotics
- Python is the most popular language for robotics prototyping, ROS2 development, AI integration, and rapid scripting.
Q
- Q-Learning
- Q-learning is a reinforcement-learning method that learns the value of each action in each state purely from trial and error — no model of the worl…
- Quadrotor Dynamics
- Quadrotor dynamics explain how a four-rotor drone flies — how varying four propeller speeds produces thrust, tilt, and turn — and why a drone is in…
- Quadruped Locomotion
- Quadruped locomotion is how four-legged robots like Spot walk, trot, and climb — balancing on shifting sets of legs across terrain that wheels can'…
- Quadruped robot
- A quadruped robot has four legs. Four legs is the easiest legged form (stable when standing still) but the most complex for high-speed running — think galloping.
- Quality of Service (QoS)
- Quality of Service is the set of policies that tune how ROS 2 delivers each message stream — reliable or best-effort, latched or not — and the numb…
- Quaternion
- A quaternion is a four-number way to represent 3D rotation that robots and drones prefer over Euler angles because it never suffers gimbal lock and…
R
- Radar Sensor in Robotics
- Radar sensors use radio waves to measure distance and velocity even in rain, fog, or dust.
- Rapidly-exploring Random Tree (RRT)
- RRT is a motion-planning algorithm that finds a path through complex, high-dimensional spaces by randomly growing a tree of reachable states — the…
- Raspberry Pi
- The Raspberry Pi is a small, inexpensive computer that runs a full Linux operating system and has a row of pins for wiring to sensors and motors —…
- Real-time computing
- Real-time computing means a system that must produce a correct result within a guaranteed time deadline — not just eventually, but by a specific moment.
- Real-Time Operating System (RTOS)
- A real-time operating system guarantees that critical tasks run exactly on time — the deterministic software foundation robots need for control loo…
- Redundancy (robotics)
- Redundancy in robotics means having more capability than the minimum strictly required — extra joints, extra sensors, extra computers — so that if…
- Redundant Manipulator
- A redundant manipulator has more joints than a task strictly needs — the extra freedom lets it reach the same point many ways, so it can dodge obst…
- Rehabilitation Robots
- Rehabilitation robots help patients regain motor function after stroke, spinal-cord injury, or amputation.
- Reinforcement learning (for robots)
- Reinforcement learning is how you teach a robot a skill by letting it try millions of times, rewarding it when it gets closer and punishing it when…
- Reinforcement Learning for Robotics
- Reinforcement learning teaches robots to do things by trial and error, optimising rewards.
- Remote Center of Compliance (RCC)
- A remote center of compliance is a clever mechanical wrist that lets a gripped part tilt and shift around its tip — so pegs self-align into holes w…
- Resolver
- A resolver is a rugged, analog position sensor that survives heat, shock, and radiation where optical encoders fail — the angle sensor of choice fo…
- Reward Shaping
- Reward shaping is the art of designing the reward signal so a robot actually learns what you intended — the make-or-break, deceptively hard part of…
- Risk Assessment (Robotics)
- Risk assessment is the systematic process of identifying what could go wrong with a robot and reducing each hazard to an acceptable level — the req…
- Robot Arm (Manipulator)
- An articulated mechanical arm with multiple powered joints, designed to position a tool — gripper, welder, spray nozzle — in 3D space.
- Robot ethics
- Robot ethics examines the moral questions raised by autonomous and semi-autonomous machines — who is responsible when a robot causes harm, how robo…
- Robot Programming Basics — Complete Beginner Guide
- A beginner's overview of how to program a robot — pseudocode, control loops, sensors, actuators, debugging — for students new to robotics in India.
- Robot Risk Assessment
- Risk assessment systematically identifies hazards in a robot system and the controls needed to reduce risk to acceptable levels.
- Robot ROI — Calculating Return on Investment
- Robot ROI quantifies whether a robotic system pays back its cost in saved labour, quality, and uptime.
- Robot Safety in Industrial and Service Robotics
- Robot safety covers the engineering and procedural measures that prevent injury during robot operation.
- Robot Singularity
- A singularity is a robot pose where the arm loses a degree of freedom — joints can't produce certain end-effector motions.
- robot_state_publisher
- robot_state_publisher is the ROS node that turns a robot's joint angles into a live tree of coordinate frames — the quiet bridge that lets the rest…
- Robot-as-a-Service (RaaS)
- RaaS lets customers pay monthly to use robots instead of buying.
- Robotic Prosthetic Limb
- Robotic prosthetic limbs restore function for amputees using motors, sensors, and EMG/EEG control.
- Robotics
- Robotics is the field that designs, builds, and programs machines capable of carrying out physical tasks automatically or semi-automatically — blen…
- Robotics on the ISS
- The International Space Station hosts Canadarm2, Dextre, JEMRMS, free-flying CubeSats, and the Astrobee free-flyers — a robotic ecosystem in orbit.
- Robotics System Integrator
- System integrators design, install, and maintain robotic systems for customers.
- ROS (Robot Operating System)
- ROS isn't really an operating system — it's the toolkit that lets the dozens of programs inside a robot talk to each other.
- ROS Action
- A ROS action handles long-running goals — navigate somewhere, execute a trajectory — with live progress feedback and the ability to cancel, filling…
- ROS Node
- A ROS node is a single program that does one job in a robot's software — a camera driver, a planner, a controller — and talks to the others over to…
- ROS Parameter
- A ROS parameter is a named configuration value a node exposes — a gain, a frame name, a topic — that can be set at launch and changed at runtime, s…
- ROS Service
- A ROS service is a request–response call between nodes — one node asks a question and waits for an answer, the right tool for quick, occasional com…
- ROS Topic
- A ROS topic is a named channel that carries a continuous stream of messages between nodes — how a robot's sensor data, commands, and state flow thr…
- ROS2
- Robot Operating System 2 — the DDS-based middleware that runs nearly every modern research and commercial robot.
- ROS2 Actions
- A ROS 2 action is how one piece of a robot's software asks another to carry out a job that takes a while — drive across a room, rotate to face a di…
- ROS2 Bags (rosbag2)
- rosbag2 records and replays ROS2 topics for offline analysis and regression testing.
- ROS2 Humble
- ROS2 Humble Hawksbill is a long-term-support release running on Ubuntu 22.04.
- ROS2 Launch
- ROS2 launch files start, configure, and monitor multiple nodes from a single command.
- ROS2 Nodes
- A ROS2 node is a single-process unit of computation that publishes and subscribes to topics, services, and actions.
- ROS2 Parameters
- ROS2 parameters let you configure nodes at runtime without recompiling.
- ROS2 Services
- A ROS 2 service is a short question one piece of a robot's software asks another and waits for a single answer to — "save the map", "spawn a turtle…
- ROS2 TF (tf2)
- tf2 in ROS2 tracks every coordinate frame on a robot and lets nodes transform points and poses between them.
- ROS2 Topics
- ROS2 topics are typed, asynchronous message buses that let robot nodes publish and subscribe data — like sensor readings or control commands — anyw…
- ROS2 URDF
- URDF (Unified Robot Description Format) describes a robot's links, joints, and visuals as XML.
- rosbag
- rosbag records a robot's message streams to a file and plays them back later — the tool that lets engineers capture real runs, debug offline, and t…
- Rotary Encoder
- A rotary encoder measures how far and how fast a shaft has turned — the sense of position that lets a robot know where its joints and wheels are, a…
- Rotation Matrix
- A rotation matrix is a 3×3 grid of numbers that encodes an orientation in 3D — the workhorse representation robots use to turn a direction in one f…
- RRT (Rapidly-exploring Random Tree)
- RRT is a sampling-based motion-planning algorithm that finds paths in complex high-DOF spaces.
- RRT*
- RRT* is the optimal version of the RRT motion planner — it rewires its search tree as it grows so the path it finds converges toward the shortest o…
- RTK-GPS
- RTK-GPS sharpens ordinary satellite positioning from meters down to centimeters using a nearby reference station — the precision-outdoor-localizati…
- RViz
- RViz is ROS's 3D visualization tool — it shows what a robot sees and thinks, from lidar scans and camera feeds to planned paths and coordinate fram…
- RViz2
- RViz2 is the ROS2 3D visualisation tool. It shows your robot, sensor data, planned paths, and tf frames in real time — the daily window into your robot.
S
- Safety Laser Scanner
- A safety laser scanner is a safety-rated 2D LIDAR that defines virtual safety zones.
- Safety Light Curtain
- A safety light curtain is an invisible wall of light beams that stops a machine the instant someone reaches through it — a standard guarding device…
- Safety-Rated Monitored Stop
- A safety-rated monitored stop halts a robot — with power still on — whenever a person enters its shared space, and lets it resume automatically whe…
- Satellite-Servicing Robot
- Satellite-servicing robots repair, refuel, and reposition spacecraft in orbit — extending satellites' lives and cleaning up debris, an emerging fie…
- Scan Matching
- Scan matching estimates how a robot moved by aligning one lidar scan to the next — the front-end that turns raw range data into motion estimates an…
- Screw Theory
- Screw theory describes every movement of a rigid object as one screw motion (turning about a single axis while sliding along it) and every force on…
- Search-and-rescue robotics
- Search-and-rescue robotics is the use of ground, aerial, and aquatic robots to locate and assist survivors in disaster zones — collapsed buildings,…
- Self-Supervised Learning
- Self-supervised learning trains models on unlabeled data by inventing the labels from the data itself — the technique that lets robots learn rich r…
- Semantic Mapping
- Semantic mapping builds a map that knows not just the shape of a space but the meaning of what's in it — 'this is a chair, that's a door, this room…
- Semantic segmentation
- Semantic segmentation is a computer-vision technique that classifies every single pixel in an image into a named category — road, pedestrian, sky —…
- Sense-Think-Act loop
- The Sense-Think-Act loop is the repeating cycle at the heart of every robot — gather information from the environment, process it, then do something physical.
- Sensor
- A sensor is a device that detects something physical — light, sound, distance, heat, motion, force — and turns it into an electrical signal a computer can read.
- Sensor fusion
- Sensor fusion is the process of combining data from multiple sensors — cameras, lidar, radar, IMU — into a single, more reliable picture of the wor…
- Series Elastic Actuator (SEA)
- A series elastic actuator deliberately puts a spring between the motor and the load — turning a stiff, hard-to-control drive into one that can sens…
- Service robot
- A service robot is any robot that performs tasks useful to humans outside of industrial manufacturing — in places like hospitals, hotels, homes, an…
- Servo motor
- A servo motor is a small motor that knows its own angle — you tell it where to point, and it goes there and holds.
- Shape-memory alloy
- A shape-memory alloy is a metal that remembers a shape it was trained to hold, snapping back to that shape when heated — giving roboticists a wire…
- Shared Autonomy
- Shared autonomy blends human intent with robot autonomy — the person gives high-level guidance while the robot handles the fine details — combining…
- SIFT Features
- SIFT finds image features that stay recognizable despite changes in scale, rotation, and lighting — the landmark algorithm that made robust image m…
- Sim-to-real
- Sim-to-real is the challenge — and the set of techniques — involved in training a robot in a computer simulation and then deploying that trained be…
- Sim-to-Real Transfer in Robotics
- Sim-to-Real bridges robot policies trained in simulation to the real world.
- Single-board computer
- A single-board computer is a full computer — processor, RAM, storage, and ports — built onto a single circuit board the size of a credit card, powe…
- Singularity Analysis
- Singularity analysis finds the configurations where a robot arm loses the ability to move or exert force in some direction — mapping the 'no-go' po…
- Skid-Steer
- Skid-steer drives a robot by running the wheels or tracks on each side at different speeds and letting them scrub sideways to turn — a simple, rugg…
- SLAM
- SLAM is the technique a robot uses to build a map of an unfamiliar place — while figuring out where it is on that map.
- Sliding Mode Control in Robotics
- Sliding mode control drives a robot onto a chosen "sliding surface" with strong disturbance rejection.
- Slip Ring
- A slip ring carries power and signals across a continuously rotating joint — the component that lets a spinning lidar, a turret, or a rotating robo…
- Social Robotics
- Social robots are designed to interact with humans as companions, teachers, or service agents.
- Soft actuator
- A soft actuator produces movement using flexible, compliant materials — silicone, fabric, hydrogels — rather than rigid metal, letting robots bend,…
- Soft robot
- A soft robot is a robot built primarily from flexible, compliant materials — silicone, rubber, hydrogels — that can deform, stretch, and squeeze th…
- Solenoid in Robotics
- A solenoid is a linear actuator that converts electrical current into linear motion using a magnetic coil.
- SOTIF (Safety of the Intended Functionality)
- SOTIF addresses a safety problem unique to AI-driven robots and self-driving cars — being unsafe even when nothing is 'broken,' because the system'…
- Space robotics
- Space robotics is the design and operation of robotic systems that work in orbit, on planetary surfaces, or in deep space — environments so extreme…
- Speech Recognition for Robots
- Speech recognition lets robots understand spoken commands.
- Speed and Separation Monitoring
- Speed and separation monitoring keeps a robot and a person at a safe distance by watching where the human is and slowing or stopping as they get cl…
- SPI
- SPI is a fast four-wire bus for high-speed communication between a controller and chips — used when a robot needs to move data quickly to displays,…
- Spring-Mass-Damper
- The spring-mass-damper is the simplest model of a vibrating, settling system — three elements that capture how almost any robot joint or structure…
- State Lattice Planner
- A state lattice planner searches over a pre-built set of feasible motion primitives arranged in a regular grid of states — giving vehicles drivable…
- State Observer
- A state observer estimates the internal states a robot can't directly measure — like velocity or load — by running a model in parallel and correcti…
- Stepper motor
- A stepper motor moves in precise, discrete jumps rather than continuous rotation, letting a robot know exactly where it is without needing a position sensor.
- Stereo Vision
- Stereo vision gives a robot depth from two cameras the same way two eyes give you depth — by measuring how far each object shifts between the left…
- STM32 in Robotics
- STM32 ARM Cortex microcontrollers are the workhorse MCU for serious robotics — motor control, sensor sampling, real-time loops at high frequency.
- Strain Gauge
- A strain gauge is a tiny sensor whose resistance changes when it's stretched — the basic element that lets robots feel force and weight, from a gri…
- Structure from Motion (SfM)
- Structure from motion reconstructs 3D scenes and camera positions from a set of ordinary 2D photos — the classic pipeline behind photogrammetry, 3D…
- Structured Light
- Structured light measures 3D shape by projecting a known pattern and watching how it bends over surfaces — the high-detail depth tech behind the or…
- Suction Gripper
- A suction gripper picks things up with vacuum through a soft cup — the fast, gentle, one-touch way robots grab flat or smooth objects, and the quie…
- Supercapacitor
- A supercapacitor stores energy that it can release and absorb in bursts far faster than a battery — powering a robot's sudden high-current demands,…
- Surgical robot
- A surgical robot is a system that allows a surgeon to perform operations with greater precision, smaller incisions, and less physical fatigue — the…
- Swarm robotics
- Swarm robotics is the study and practice of coordinating large numbers of simple robots to achieve complex tasks collectively — taking inspiration…
- Swerve Drive
- Swerve drive gives a robot omnidirectional motion using normal wheels that can each both steer and drive independently — the powerful, rugged answe…
- Synthetic Data
- Synthetic data is computer-generated training data — rendered images, simulated sensor readings — with perfect labels for free, letting robots lear…
- Systemantics — India's Industrial Robotics Pioneer
- Systemantics is a Bengaluru-based Indian industrial-robotics company building affordable SCARA, articulated, and cartesian robots for Indian factories.
T
- Tactile sensor
- A tactile sensor gives a robot spatially distributed touch — not just how hard it is pressing in total, but exactly where across a surface contact is happening.
- Teleoperation
- Teleoperation means controlling a robot from a distance, with a human in the loop making the decisions.
- Tendon-Driven Hand
- A tendon-driven hand moves its fingers with cables pulled by motors placed in the forearm — the design that gives robot hands slim, light, human-li…
- TF (Coordinate Transforms)
- TF is the system robots use to keep track of where every part and sensor is relative to everything else — a live tree of coordinate frames that ans…
- Thermal Camera in Robotics
- Thermal cameras image infrared heat instead of visible light.
- Time-of-Flight Sensor
- A time-of-flight sensor measures distance by timing how long light takes to bounce back — the compact ranging tech behind depth cameras, phone face…
- Time-of-Flight Sensor in Robotics
- Time-of-Flight (ToF) sensors measure distance by timing reflected light pulses.
- Timed Elastic Band (TEB)
- The timed elastic band is a local planner that treats a robot's path like a stretchy band pulled taut by the goal and pushed by obstacles — optimiz…
- Timing Belt
- A timing belt is a toothed belt that transmits motion between pulleys without slipping — a light, quiet, cheap way to move power around a robot, us…
- Tool Center Point (TCP)
- The tool center point is the exact spot on a robot's tool that it actually controls and moves — the tip of the welding torch or the center of the g…
- Torque
- Torque is turning force — the twist a motor applies to move a joint.
- Tracked Robot
- A tracked robot drives on continuous belts instead of wheels — spreading its weight and gripping loose or broken ground, which makes tracks the go-…
- Trajectory
- A trajectory is a path through space with a time dimension added — specifying not just where a robot should go, but exactly when it should be at ea…
- Trajectory Optimization
- Trajectory optimization computes a robot's motion by minimizing a cost — smoothness, effort, time — subject to physics and constraints, producing e…
- Transfer learning
- Transfer learning is a machine-learning technique where a model trained on one task or dataset is reused as the starting point for a different but…
- Transformer Architecture
- Transformers use self-attention to handle sequences and multimodal inputs.
- Trapezoidal Velocity Profile
- A trapezoidal velocity profile moves a robot joint by accelerating, cruising, then decelerating — the simplest smooth motion plan, and the default…
- Turing Test in Robotics & AI
- Alan Turing's 1950 thought experiment defines machine intelligence as indistinguishability from human conversation.
U
- UART
- UART is the simple point-to-point serial link that connects two devices with just a transmit and receive wire — the humble, universal way robots ta…
- Ultrasonic sensor
- An ultrasonic sensor measures distance by sending out a burst of sound too high-pitched for humans to hear and timing how long the echo takes to return.
- Underactuated Hand
- An underactuated hand has more finger joints than motors — so its fingers passively conform around whatever they touch, grabbing many shapes secure…
- Underwater robot (AUV / ROV)
- An underwater robot is a vehicle that operates beneath the water's surface — either remotely piloted by a human via a tether (ROV) or navigating au…
- Unimate — The First Industrial Robot
- Unimate, installed at General Motors in 1961, was the first industrial robot.
- URDF
- URDF (Unified Robot Description Format) is the XML file that tells a robot's software what its body looks like — every link, joint, and sensor — so…
V
- Vacuum Gripper in Robotics
- Vacuum grippers pick up flat, smooth, or porous objects using suction.
- Variable Stiffness Actuator (VSA)
- A variable stiffness actuator can change how springy it is on command — stiff for precision, soft for safety and efficiency — letting a robot adapt…
- Visual Grasp Detection
- Visual grasp detection uses a neural network to predict where to grip an object straight from a camera or depth image — the learning-based approach…
- Visual Odometry in Robotics
- Visual odometry estimates a robot's motion from camera images alone.
- Visual Servoing
- Visual servoing closes the control loop with a camera — the robot moves to make what it sees match what it wants to see, letting arms and drones tr…
- Visual SLAM in Robotics
- Visual SLAM uses only camera input to simultaneously map an environment and localise the robot.
- Voltage Regulator in Robotics
- Voltage regulators step a battery's variable voltage down (or up) to a fixed, clean voltage for sensors and microcontrollers.
- VTOL
- VTOL — vertical takeoff and landing — lets an aircraft rise straight up like a helicopter, then fly efficiently like a plane, giving drones both ho…
W
- Warehouse Robot
- Warehouse robots move goods, shelves, and orders through fulfillment centers — the fastest-growing application of mobile robotics, and the reason s…
- Warehouse robotics
- Warehouse robotics is the use of automated machines — mobile robots, robotic arms, and conveyor systems — to move, sort, and store goods inside a f…
- Waypoint
- A waypoint is a specific named position — and optionally orientation — that a robot is instructed to reach on its way to a final goal, breaking a l…
- Weeding Robots
- Weeding robots use computer vision to spot weeds and remove them mechanically or with targeted spray.
- What is a robot?
- A robot is a machine that can sense the world, decide what to do, and act on its own — without a human guiding every move.
- Wheels vs Tracks for Robots
- When to use wheels, tracks, or legs on a mobile robot.
- Whole-Body Control
- Whole-body control coordinates every joint of a complex robot at once to juggle many goals — balance, reach, and contact — turning a humanoid's 30-…
- Window-Cleaning Robot
- A window-cleaning robot climbs and scrubs glass — clinging with suction or magnets while it wipes — automating a tedious chore at home and a danger…
- Work-Energy Theorem
- The work-energy theorem links force, motion, and energy — the reasoning behind how much power a robot needs, how legged robots store and return ene…
- Workspace of a Robot
- A robot's workspace is the full set of positions its end-effector can reach.
- World Model
- A world model is a neural network that learns to predict how the world changes — letting a robot 'imagine' the outcomes of its actions and plan or…
Z
- Zero-Moment Point (ZMP)
- The zero-moment point is the spot on the ground a walking robot must keep its dynamic forces balanced over to avoid tipping — the criterion that ma…
Numbers & symbols
- 3D Gaussian Splatting
- 3D Gaussian splatting represents a scene as millions of tiny colored blobs, rendering photorealistic 3D in real time — the fast successor to NeRF t…
- 6D Pose Estimation
- 6D pose estimation figures out an object's full position and orientation in 3D from a camera — not just where it is, but which way it faces — the p…
Missing a term? Every entry is written by one person and reviewed against its sources; tell us what to add at hello@r2bot.in. For the AI-specific vocabulary (VLA models, world models, sim-to-real) see the AI + Robotics glossary.