Robotics is the interdisciplinary field concerned with the design, construction, programming, control, sensing and application of robots and robotic systems. It combines mechanical engineering, electrical and electronic engineering, computer science, control theory, artificial intelligence, materials science and increasingly areas such as human-computer interaction and cognitive science. A robot is generally understood as a programmable physical machine capable of sensing its environment, processing information and producing physical action, although definitions vary according to context. Robotics therefore encompasses both the physical machinery of robots and the computational methods used to make them perform useful tasks. It includes industrial manipulators, mobile robots, autonomous vehicles, drones, medical systems, agricultural machines, domestic robots and specialised research platforms.
The basic architecture of a robotic system usually involves sensing, computation, actuation and mechanical structure. Sensors provide information about the robot's internal condition and external environment, including position, velocity, acceleration, force, torque, temperature, distance, images and sound. A controller interprets sensor information and calculates appropriate commands, while actuators convert electrical, hydraulic, pneumatic or other forms of energy into movement. Mechanical components such as joints, links, wheels, tracks, grippers and end-effectors determine how physical movement is produced. Software coordinates these components and may implement low-level motor control, trajectory generation, perception, planning, navigation, manipulation or high-level task management.
Robot motion can be described through concepts such as degrees of freedom, kinematics, dynamics and workspace. Forward kinematics calculates the position and orientation of a robot's end-effector from known joint positions, while inverse kinematics determines joint configurations required to achieve a desired position and orientation. Dynamics concerns the forces and torques required to produce movement, taking account of mass, inertia, gravity and other physical effects.
A robot's workspace describes the region that its mechanism can reach, while its payload specifies the load it can safely manipulate under defined conditions. These characteristics are fundamental when selecting or designing a robot because a system that can reach a location may not necessarily have sufficient strength, speed, precision or stability to perform the intended task.
Control is central to robotics because a robot must continuously translate desired behaviour into physical movement. Open-loop control issues commands without relying on feedback, whereas closed-loop control measures the resulting state and adjusts commands to reduce error. Proportional-integral-derivative controllers are widely used for regulating variables such as motor position or speed, while more advanced systems can use model-based control, adaptive control, optimisation or nonlinear control. Motion planning determines feasible paths while considering obstacles, mechanical constraints and task requirements. In autonomous robots, planning can operate at several levels, from low-level trajectory generation to high-level decisions about which task should be performed next.
Perception allows robots to construct useful information about their surroundings. Cameras can provide visual information, while lidar, radar, ultrasonic sensors, inertial measurement units, force sensors and tactile sensors provide complementary information. Computer vision techniques can identify objects, estimate poses, segment scenes or track movement, while simultaneous localisation and mapping methods can allow mobile robots to estimate their position while constructing maps. Artificial intelligence and machine learning can improve recognition, prediction and decision-making, although robotic systems often combine learned components with deterministic control because physical actions require predictable constraints and safety behaviour. Sensor fusion can combine several imperfect measurements to produce a more robust estimate of the robot's state or environment.
Robotics developed from earlier mechanisms, automata, numerical control systems and programmable machinery, with modern industrial robotics emerging particularly strongly during the twentieth century. The word "robot" derives from the Czech word "robota", associated with compulsory labour, and became widely known through Karel Čapek's 1920 play "R.U.R." The concept subsequently evolved from fixed industrial manipulators toward mobile, autonomous and increasingly adaptive machines. Industrial robots transformed manufacturing by making repetitive manipulation and welding tasks programmable, while advances in computing, sensors and artificial intelligence expanded robotics into logistics, healthcare, agriculture, exploration and consumer applications. Contemporary robotics increasingly combines physical autonomy with networked computing, simulation and machine-learning techniques.
Robotics has major applications where physical interaction is repetitive, hazardous, highly precise, difficult for humans or economically advantageous to automate. Manufacturing robots can weld, paint, assemble and handle materials; logistics robots can move goods; agricultural robots can inspect crops or perform field operations; medical robots can assist clinicians; and domestic robots can perform cleaning or other household tasks.
Robots can also operate in environments that are dangerous or inaccessible to people, including contaminated areas, deep underwater locations, disaster sites and certain space environments. However, automation is constrained by reliability, cost, maintenance, environmental uncertainty and the difficulty of reproducing general human dexterity. A robot that performs exceptionally well in a structured environment may perform poorly when confronted with unexpected objects, variable lighting, deformable materials or ambiguous instructions.
Robotics is closely related to automation, artificial intelligence, mechatronics and cyber-physical systems, but these terms are not interchangeable. Automation can occur without robots, such as a software process or fixed mechanical control system, while robotics specifically involves programmable physical machines. Artificial intelligence can exist entirely in software and therefore does not necessarily constitute robotics. Mechatronics integrates mechanical, electrical and control engineering and provides much of the engineering foundation for robotic machines.
The field's broader significance comes from bringing computation into direct physical interaction with the world: robots do not merely process information but use that information to move, manipulate, navigate and alter physical environments. This makes reliability, safety, physical design and real-time control as important as algorithms and software.
