ABB ACU-01B 3HNA024871-001 Robot controller

Robot controller is a device that controls the robot to complete certain actions or tasks according to instructions and sensing information. It is the heart of the robot and determines the performance of the robot. From the processing mode of robot control algorithm, it can be divided into two structural types: serial and parallel.ABB ACU-01B 3HNA024871-001 Robot controller

Serial processing structure

The so-called serial processing structure means that the control algorithm of the robot is processed by a serial machine. For this type of controller, it can be divided into the following types according to the computer structure and control mode:

(1) Single CPU structure, centralized control mode, all control functions are realized by a powerful computer. In early robots, such as Hero-I and Robot-I, this structure was adopted, but many calculations (such as coordinate transformation) were needed in the control process, so this control structure was slow.

(2) Two-stage CPU structure and master-slave control mode. The first-stage CPU is the host, which plays the functions of system management, robot language compilation and man-machine interface. At the same time, it also uses its computing power to complete coordinate transformation and trajectory interpolation, and regularly sends the computing results to the public memory as the increment of joint motion for the second-stage CPU to read; The secondary CPU completes the digital control of all joint positions. There is basically no connection between the two CPU buses of this kind of system, and it only exchanges data through common memory, which is a loosely coupled relationship. It is very difficult to further disperse the functions by using more CPUs. The computer system of M otoma n robot (5 joints, driven by DC motor) produced in Japan in the 1970s belongs to this master-slave structure.

(3) Multi-CPU structure and distributed control mode

At present, this two-level distributed structure of upper and lower computers is widely used, and the upper computer is responsible for the whole system management, kinematics calculation, trajectory planning and so on. The lower computer is composed of multi-C PU, and each CPU controls a joint movement. These CPUs are closely coupled with the main controller through the bus, so the working speed and control performance of the controller with this structure are obviously improved. However, the common feature of these multi-CPU systems is the functional distributed structure adopted for specific problems, that is, each processor undertakes a fixed task, which is the structure of most commercial robot controllers in the world at present.

The position control part of the controller computer control system adopts digital position control almost without exception.

The above-mentioned controllers all use serial computers to calculate the robot control algorithm, and they have a common weakness: heavy calculation burden and poor real-time performance. Therefore, offline planning and feedforward compensation decoupling are mostly used to reduce the calculation burden in real-time control, and when the robot is disturbed in operation, its performance will be affected, making it more difficult to ensure the required accuracy index in high-speed movement.

Parallel processing structure

Parallel processing technology is an important and effective means to improve the calculation speed, which can meet the real-time requirements of robot control. According to the literature, the parallel algorithm of robot kinematics and dynamics and its realization are studied more. In 1982, J. Y. S. Luh first proposed the parallel processing problem of robot dynamics. This is because the dynamic equation of articulated robot is a set of nonlinear and strongly coupled second-order differential equations, and the calculation is very complicated. Improving the calculation speed of robot dynamic algorithm also lays the foundation for realizing complex control algorithms such as calculating torque method, nonlinear feedforward method and adaptive control method. One of the ways to develop parallel algorithms is to transform serial algorithms to parallelize them, and then map them to parallel structures. Generally, there are two ways: one is to consider the given parallel processor structure and develop the parallelism of the algorithms according to the computing model supported by the processor structure; The second is to develop the parallelism of the algorithm first, and then design the parallel processor structure supporting the algorithm to achieve the best parallel efficiency.