Robotics

Robotics is the interdisciplinary field devoted to the design, construction, operation and use of robots: machines able to perceive their environment, process that information and act physically on the world. It brings together mechanical, electrical and electronic engineering with computer science and artificial intelligence.

The sense-plan-act loop

An autonomous robot is usually organized around a classic cycle, that of sense-plan-act: perceive the environment with sensors to build a model of the world, plan the best sequence of movements over that model, and act by executing it with the actuators. This paradigm was born in the late 1960s with pioneering robots such as Shakey, of the Stanford Research Institute, and works especially well in structured environments. Documentation: institutional Shakey archive.

Robotics and artificial intelligence

It is worth not confusing the two disciplines. Robotics provides the physical body; artificial intelligence is one of the technologies that govern it, the one that gives it “intelligence”: computer vision to perceive, planning to decide, and reinforcement or imitation learning to learn to move. Not all robotics uses AI, nor is all AI robotics.

Where the words come from

The word “robot” was born in 1920 in the play R.U.R., by the Czech writer Karel Čapek—though it was his brother Josef who suggested it—and comes from the Czech robota, “forced labor.” Writer Isaac Asimov introduced “robotics” and the Three Laws through his early-1940s stories; the Computer History Museum timeline places that cluster in 1941. They are fictional rules devised to explore conflicts among orders, safety, and self-preservation, not a technical standard governing real robots. Documentation: history documented by the Czech foreign service.

Where it is heading

Robotics spans industrial arms, mobile and collaborative robots, drones, and autonomous vehicles. According to the International Federation of Robotics' World Robotics 2025, the operational stock of industrial robots reached 4,664,000 units in 2024. This counts industrial robots operating in factories, not every robot or its degree of autonomy. In 2025, two manufacturers introduced models combining vision, language, and action for robotics. Google DeepMind described Gemini Robotics as a vision-language-action model that adds physical actions to Gemini 2.0; NVIDIA presented GR00T N1 for humanoids and released selected weights and data. Those are issuer descriptions, not independent proof of a general policy. Evaluating “generality” requires the robots, tasks, environments, success rate, and task-specific adaptation behind each demonstration.