Application of Intelligent Control in Robot Technology
Abstract: With the progress of science and technology, the intelligence of robots has grown from scratch, from low to high. With the development of computer technology, network technology, artificial intelligence, new materials and MEMS technology, the development trend of intelligence, networking and miniaturization is prominent. This paper mainly discusses the application of intelligent control in robot technology.
Keywords: intelligent control robot technology
1, Introduction
Industrial robot is a complex nonlinear, strongly coupled and multivariable dynamic system, and its operation is often uncertain. However, it is often difficult to establish an accurate mathematical model by using the prior knowledge of the existing robot dynamics model. Even if the model is established, it is very complicated and has a large amount of calculation, which can not meet the requirements of real-time control of robots. The emergence of intelligent control provides a new way to solve some problems in robot control. Because intelligent control has the characteristics of global optimization, independence from object model, self-learning and self-adaptation, it can achieve good results in solving complex control problems such as robots.
2. Overview of intelligent robots
When it comes to intelligent robots, it is easy to think of artificial intelligence. There are three different schools of artificial intelligence: biological simulation school, psychology school and behaviorism school. In the mid-1950s, the behaviorism school was always dominant. Behaviorist scholars believe that most of human knowledge can not be accurately described by mathematical methods, and put forward a symbolic reasoning system, that is, an expert system, which uses symbols to express knowledge on computers. Expert systems use rules or semantic web to represent knowledge rules. However, some human knowledge cannot be described by clear rules, so the expert system was once in trouble. In recent years, some breakthroughs have been made in neural network technology, which makes the school of biological simulation active. Intelligent robot is the carrier of artificial intelligence research, but there are great differences between them. For example, for the intelligent assembly robot, it is required to obtain the assembly information on the drawings through the vision system, find and find the required workpieces through analysis, and assemble the workpieces one by one according to the correct assembly order. Therefore, intelligent robots need to have knowledge expression and acquisition technology, and make plans for assembly. At the same time, when looking for and looking for artifacts, it is necessary to use pattern recognition technology to find artifacts on patterns. Assembly is a complex process, which may require the mixed control technology of force and position, or the robot body may need to be equipped with flexible wrist to complete the task, which is another mechanical problem. Intelligent robot involves a wide range and has high technical requirements. It is a high-tech complex. So, what is an intelligent robot? So far, there is still no unified definition of intelligent robot in the world. It is generally believed that intelligent robot is a machine with perception, thinking and action. The so-called perception refers to the ability to discover, identify and describe the external environment and its own state. For example, in an assembly operation, if it wants to find and identify the required workpiece, it needs to use a visual sensor to perceive the workpiece. At the same time, in order to get close to the workpiece, intelligent robots need to know the obstacles and realize obstacle avoidance in unstructured environment. These all depend on the sensory system of intelligent robots, that is, various sensors. The so-called thinking is that the robot itself has the ability to solve problems. For example, the assembly robot can find the assembly method and sequence of parts for a complex machine according to the design requirements, and command the actuator, that is, the action part, to assemble the machine. Action means that the robot has a mechanism and a driving device that can complete the work. Therefore, intelligent robot is a complex combination of software and hardware. Although there is no unified definition of intelligent robots, there is still a perceptual knowledge through the investigation of specific intelligent robots.
3. Intelligent robot architecture.
The architecture of intelligent robot mainly includes hardware system and software system.
Three aspects. Because of the different uses of intelligent robots, the composition of hardware systems is also different. The structure is designed according to the human prototype. The system mainly includes vision system, walking mechanism, manipulator, control system and man-machine interface. As shown in figure 1:
2. 1 vision system
Intelligent robots use artificial vision system to simulate human eyes. Vision system can be divided into three parts: image acquisition, image processing and image understanding. A vision sensor is a device that converts the light signal of a scene into an electrical signal. Early intelligent robots used photoconductive cameras as their visual sensors. In recent years, solid-state vision sensors, such as charge-coupled devices CCD, metal oxide semiconductor CMOS devices. Compared with TV cameras, solid-state vision sensors are small in size and light in weight, so they are widely used. The electrical signals obtained by the vision sensor are converted into digital signals by A/D, that is, digital images. A single vision sensor can only obtain plane images, but cannot obtain depth or distance information. At present, binocular stereo vision or distance sensor is being studied to obtain three-dimensional stereo vision information. But so far there is no simple and practical device. The digital image is processed to extract features, and then the external scene is recognized by the image understanding part.
2.2 Walking mechanism
The walking mechanisms of intelligent robots are wheeled, crawler or crawling and humanoid biped. At present, most intelligent robots adopt wheeled, crawler or crawling walking mechanism, which is simple and convenient to realize. 1987 biped robots began to appear, and then quadruped and hexapod robots were developed one after another. It is a dream that scientists have been pursuing to make robots walk like humans.
2.3 manipulator
Intelligent robots can borrow the manipulator structure of industrial robots. However, the degree of freedom of the hand needs to be increased, and sensors such as touch, pressure, strength and sliding should be equipped to produce soft, flexible and reliable movements and complete complex operations.
2.4 control system
Multi-sensor information fusion, motion planning, environment modeling and intelligent reasoning of intelligent robots need a lot of memory and high-speed, real-time processing ability. How is Feng now? As the controller of intelligent robot, Norman structure still has some shortcomings. With the emergence of photonic computers and parallel processing structures, the processing capacity of intelligent robots will be higher. Robots will have higher potassium energy.
2.5 man-machine interface
The man-machine interface of intelligent robot includes the robot's speaking, listening and network connection, microphone, speaker, speech synthesis and recognition system, which enables the robot to understand human instructions and communicate with people in natural language. Robots also need network connection, and people can control and operate machines and people through network and communication technology.
With the in-depth research of intelligent robots, sensors are used more and more, and unit technologies such as information fusion, planning, problem solving, kinematics and dynamics calculation are constantly improved, which makes the overall intelligent ability of intelligent robots continuously enhanced and the system structure increasingly complex. Intelligent robot is a complex system with multiple CPUs, which must be divided into several modules or hierarchical structures. In this structure, how to decompose functions, how to determine the time relationship and how to allocate space resources are the key issues that directly affect the intelligent ability of the whole system. At the same time, in order to ensure the expansion of the intelligent system and facilitate the technical update, the structure of the system is required to be open, to ensure the continuous enhancement of intelligent capabilities, the entry of new or more sensors, and the combination of various algorithms, which makes the architecture itself a complex problem that needs to be studied and solved. The architecture of intelligent robot is to define the relationship and function distribution among all parts of an intelligent robot system, and to determine the information circulation relationship and logical calculation structure of an intelligent robot or multiple intelligent robot systems. For a specific robot, it can be said that it is the overall structure of the robot information processing and control system, excluding the mechanical structure of the robot. In fact, any robot has its own architecture. At present, the control system of most industrial robots has a two-layer structure. The upper layer is responsible for kinematics calculation and human-computer interaction, and the lower layer is responsible for servo control of each joint.
References:
Zuo Min, Zeng Guangping. Research on intelligent control of robots based on parallel evolution [J]. Computer simulation, 20 1 1, 08: 15- 16.
Chen Wei, Si Guangshu. Design of intelligent control system for autonomous humanoid robot [J]. Servo control, 2009,02: 76-78.
Conway. Intelligent control of mobile robot motor [J]. Equipment manufacturing technology, 2010.08:102-103.
Did you get a look at him? Robot control technology paper? People still see:
1. Technical papers on handling robots
2. Mechanical and electrical control technology paper
3. Scientific papers on robots
4. Model essay on industrial robot technology (2)
5. Papers on Robotics Science and Technology