There is no unified international standard for robot classification; different classifications can be made from different perspectives.
Development Stages
① First Generation Robots: Teach-Put Robots. In 1947, Oak Ridge National Laboratory in the United States developed the world's first remotely controlled robot for handling nuclear fuel. In 1962, the United States successfully developed the PUMA general-purpose teach-put robot. This type of robot uses a computer to control a multi-degree-of-freedom machine. It stores programs and information through teaching, retrieves the information during operation, and then issues commands. In this way, the robot can repeatedly reproduce the actions based on the human's teaching. For example, a spot welding robot for automobiles, once the spot welding process is taught, will always repeat this task.
② Second Generation Robots: Sensory Robots Teach-and-playback robots lack perception of their external environment. They are unaware of the magnitude of the manipulating force, the existence of the workpiece, or the quality of the weld. Therefore, in the late 1970s, research began on second-generation robots called sensory robots. These robots possess sensory abilities similar to humans, such as force, touch, slip, vision, and hearing. They can perceive and identify the shape, size, and color of workpieces through these senses.
③ Third-generation robots: Intelligent robots. These robots, invented since the 1990s, are equipped with multiple sensors and can perform complex logical reasoning, judgment, and decision-making. They autonomously determine their behavior in changing internal states and external environments.