The so - called collaborative robot is actually a kind of robot that can work in a shared space with humans and interact closely with them. In short, it can work together with humans. A collaborative robot is mainly composed of a robot body, hollow torque motors, harmonic reducers, servo drives, and a controller.

Characteristics of collaborative robots:

Collaborative robots with different structures: They are mainly single - arm and dual - arm robots. A single - arm collaborative robot has only one arm and is a common form of robotic arm. Due to its flexible layout and simple control, it is widely used in various industrial scenarios. A dual - arm collaborative robot has redundant motion space, is highly efficient and flexible, and is suitable for load applications that require interactive cooperation or for scenarios where multiple actions need to be completed in a limited space, such as the cooperative action of unscrewing a bottle cap with both hands. However, the dual - arm design will lead to an increase in cost and control complexity.

Collaborative robots with different load capacities: Limited by application scenarios and structural characteristics, the load capacity of collaborative robots generally does not exceed 20 kg. Usually, we define a light - load collaborative robot as having a load capacity of 3 - 10 kg, a heavy - load collaborative robot as having a load capacity of over 10 kg, and a desktop collaborative robot as having a load capacity of less than 3 kg. Light - load collaborative robots are the most common type of collaborative robots at present and are mainly applied in the consumer electronics, testing, medical, and service industries. Collaborative robots are often used in industrial scenarios such as assembly, handling, palletizing, and packaging, while desktop collaborative robots are used in work scenarios such as sorting, dispensing, gluing, and automatic testing.

Collaborative robots with different force - sensing forms: Force - sensing is a basic and important function for collaborative robots to achieve interaction, mainly divided into current loops, joint torque sensors, six - axis force/torque sensors, and electronic skins. Among them, the sensing form based on the current loop is an early type. The motor current of each joint of the robot is directly obtained through the robot's servo system, and then the output torque of each joint is calculated using the torque coefficient provided by the motor manufacturer. This method is both economical and simple and is widely used by robot manufacturers. Since accurate external forces cannot be obtained, it is often used for collision detection and simple drag - teaching.

Collaborative robots based on joint torque sensors have been a research hotspot in recent years. A popular flexible joint is composed of a torque sensor, a multi - turn encoder, and a single - turn absolute encoder, and then the torque of each joint shaft is directly output by the torque sensor. Compared with existing feedback methods, this method has higher accuracy and feedback frequency, and it is easy to achieve compliant control of the robot. It has been applied in scenarios such as precision assembly, polishing, and surgery. However, the use of torque sensors will reduce the joint stiffness and trajectory accuracy of the robot, and structural integration will also increase the difficulty of design and manufacturing.