1、 What is a parallel robot

A parallel robot, also known as a parallel mechanism in English, is a multi loop control structure compared to a series mechanism arranged in an open-loop structure. The parallel robot is formed by connecting two or more independent motion branches between the moving platform and the fixed platform. Its characteristics include: compact structure, high stiffness, and large load-bearing capacity; No cumulative error, high accuracy; Small footprint; Fast speed, excellent sports performance, minimal component wear, and long service life. Parallel robots emerged as early as the 1990s and are now a new generation of industrial robots.

2、 The Development History of Parallel Robots

The earliest concept of parallel robots can be traced back to 1931, when Gwinnett proposed an entertainment device based on spherical parallel robots, as shown in Figure 1. Subsequently, in industrial applications, Pollard proposed a spatial parallel robot for automotive spraying in 1940, as shown in Figure 2. Gough invented a six degree of freedom parallel robot for partial tire inspection in the automotive industry in 1962, as shown in Figure 3. Subsequently, Stewart proposed a parallel robot similar to Gough, known as the classic Stewart mechanism. Due to the excessive similarity between this institution and Gough's institution, some scholars also refer to this institution as the Stewart Gough institution. The Stewart organization consists of six identical independent branches connecting the moving platform and the fixed platform, and each branch is composed of a ball joint connecting the moving platform, a moving pair, and a ball joint connecting the fixed platform, as shown in Figure 4. These early parallel robot designs were designed to meet specific entertainment or industrial needs, laying the foundation for the later more widely applied parallel robot technology.

In 1991, the research on parallel robots in China took the first step, and Professor Huang Zhen from Yanshan University successfully developed a six degree of freedom parallel robot prototype. In 1997, Tsinghua University and Tianjin University collaborated to develop the VAMTIY parallel robot for large-scale boring machines. Northeastern University launched the DSX5-70 three-bar parallel machine tool with five axis linkage in 1998. In 1999, Tianjin University and Tianjin First Machine Tool Factory successfully collaborated to develop a commercial prototype of a three coordinate parallel machine tool, LINAPOD. At the same time, Harbin Institute of Technology also developed a six degree of freedom parallel machine tool prototype. Although parallel robots started relatively late in China and there are some technological gaps compared to foreign countries, research in this field has been continuously developing in recent years. The application level of domestic parallel robots is gradually improving and has been recognized and widely used by enterprises. This development trend indicates that the domestic parallel robot field is gradually catching up and has made significant progress in technological innovation and industrial application.

3、 Classification of parallel robots

According to the structure and motion mode of robots during task execution, they can be classified into non redundant parallel robots and redundant parallel robots.

Non redundant parallel robots have the same input and output degrees of freedom. Common parallel robots such as Delta robots (Figure 5) and Stewart robots belong to non redundant parallel robots. Non redundant parallel robots have been widely used in industrial production due to their characteristics of high speed, high stiffness, high load-bearing capacity, and high precision. However, there are also some shortcomings, such as small workspace and multiple singular configurations. Through research, scholars have found that drive redundancy or motion redundancy can be used to improve this problem.

Figure 5 Delta Robot

[Explanation on redundancy: When the input degrees of freedom and output degrees of freedom of a parallel robot are inconsistent, it is called redundancy. Redundancy can be divided into two types: if the driving joint increases the mechanism's own degrees of freedom, it is called motion redundancy; if the mechanism's degrees of freedom remain unchanged, it is called driving redundancy.]

Kinematic Redundancy:

In motion redundancy, the structural degrees of freedom of the robot exceed the minimum degrees of freedom required to perform the task. In other words, robots have extra joints that can move end effectors in different ways while still completing the same task.

Features:

The degree of structural freedom increases.

Realize the same end effector effect through different combinations of joint movements.

Advantages:

Greater flexibility and adaptability.

Specific performance can be optimized during task execution, such as avoiding obstacles or optimizing motion trajectories.

A typical motion redundant planar parallel robot is shown in Figure 6, proposed by the Gosselin team and published in the TRO journal. This robot has (3+1) degrees of freedom, 1 redundant degree of freedom, no singularity in the workspace, and infinite rotational ability.

Figure 6 3+1 DOFs planar robot

Ø Redundant Execution:

In drive redundancy, the structural degrees of freedom of the robot remain unchanged, but drive joints are added. This means that robots have excess driving capabilities and can achieve the same end effector effect through different joint movements. Generally speaking, drive redundancy can be divided into two forms, as shown in Figure 7: (1) intra branch redundancy, which converts one or more passive joints within a branch into drive joints; (2) Additional branch redundancy refers to adding additional driving branches to the minimum number of branches required for the degree of freedom of the mechanism. Researchers usually use the latter because this method can improve static distribution and increase stiffness.

Features:

The structural degrees of freedom remain unchanged, but the driving joints are added.

Multiple driving methods achieve the same structural end position and direction.

Advantages:

Performance can be optimized through different driving strategies during task execution, such as energy consumption, speed, etc.

Figure 7: Two Different Methods of Driving Redundancy

Conclusion: The design and control of parallel robots involve multiple fields such as mechanical engineering, electrical engineering, and computer science, and are an important component of modern automation technology. Due to its high load, low inertia, and high precision, it has been widely used in industrial automation, assembly lines, medical surgery, precision machining, and other fields to improve production efficiency, accuracy, and adaptability. With the continuous development of technology, it is believed that parallel robots will continue to play an important role in the future, promoting further progress in automation technology.