What is mechatronics?

The term mechatronics was ‘invented’ by a Japanese engineer in 1969, as a combination of ‘mecha’ from mechanisms and ‘tronics’ from electronics. The word now has a wider meaning, being used to describe a philosophy in engineering technology in which there is a co-ordinated, and concurrently developed, integration of mechanical engineering with electronics and intelligent computer control in the design and manufacture of products and processes. As a result, mechatronic products have many mechanical functions replaced with electronic ones. This results in much greater flexibility, easy redesign and reprogramming, and the ability to carry out automated data collection and reporting.

A mechatronic system is not just a marriage of electrical and mechanical systems and is more than just a control system; it is a complete integration of all of them in which there is a concurrent approach to the design. In the design of cars, robots, machine tools, washing machines, cameras and very many other machines, such an integrated and interdisciplinary approach to engineering design is increasingly being adopted. The integration across the traditional boundaries of mechanical engineering, electrical engineering, electronics and control engineering has to occur at the earliest stages of the design process if cheaper, more reliable, more flexible systems are to be de- veloped. Mechatronics has to involve a concurrent approach to these dis- ciplines rather than a sequential approach of developing, say, a mechanical system, then designing the electrical part and the microprocessor part. Thus mechatronics is a design philosophy, an integrating approach to engineering.

Mechatronics brings together areas of technology involving sensors and measurement systems, drive and actuation systems, and microprocessor systems ( Figure 1.1 ), together with the analysis of the behaviour of systems and control systems. That essentially is a summary of this book. This chapter is an introduction to the topic, developing some of the basic concepts in order to give a framework for the rest of the book in which the details will be developed.

Examples of mechatronic systems

Consider the modern autofocus, auto-exposure camera. To use the camera all you need to do is point it at the subject and press the button to take the picture. The camera can automatically adjust the focus so that the subject is in focus and automatically adjust the aperture and shutter speed so that the correct exposure is given. You do not have to manually adjust focusing and aperture or shutter speed controls. Consider a truck smart suspension. Such a suspension adjusts to uneven loading to maintain a level platform, adjusts to cornering, moving across rough ground, etc., to maintain a smooth ride. Consider an automated production line. Such a line may involve a number of production processes which are all automatically carried out in the correct sequence and in the correct way with a reporting of the outcomes at each stage in the process. The automatic camera, the truck suspension and the automatic production line are examples of a marriage between electronics, control systems and mechanical engineering.

The term embedded system is used where microprocessors are embedded into systems and it is this type of system we are generally concerned with in mechatronics. A microprocessor may be considered as being essentially a collection of logic gates and memory elements that are not wired up as individual components but whose logical functions are implemented by means of software. As an illustration of what is meant by a logic gate, we might want an output if input A AND input B are both giving on signals. This could be implemented by what is termed an AND logic gate. An OR logic gate would give an output when either input A OR input B is on. A microprocessor is thus concerned with looking at inputs to see if they are on or off, processing the results of such an interrogation according to how it is programmed, and then giving outputs which are either on or off. See Chapter 10 for a more detailed discussion of microprocessors.

Digital communications

An external bus is a set of signal lines that interconnects microprocessors, microcontrollers, computers and programmable logic controllers (PLCs) and also connects them with peripheral equipment. Thus a computer needs to have a bus connecting it with a printer if its output is to be directed to the printer and printed. Multiprocessor systems are quite common. For example, in a car there are likely to be several microcontrollers with each controlling a different part of the system, e.g. engine management, braking and instrument panel, and communication between them is necessary. In automated plant not only is there a need for data to pass between programmable logic controllers, displays, sensors and actuators and allow for data and programs to be inputted by the operator, but there can also be data communications with other computers. There may, for example, be a need to link a PLC to a control system involving a number of PLCs and computers. Computer integrated manufacturing (CIM) is an example of a large network which can involve large numbers of machines linked together. This chapter is a consideration of how such data communications between computers can take place, whether it is just simply machine-to-machine or a large network involving large numbers of machines linked together, and the forms of standard communication interfaces.

Centralised computer control involves the use of one central computer to control an entire plant. This has the problem that failure of the computer results in the loss of control of the entire plant. This can be avoided by the use of dual computer systems. If one computer fails, the other one takes over. Such centralised systems were common in the 1960s and 1970s. The development of the microprocessor and the ever reducing costs of computers have led to multi-computer systems becoming more common and the development of hierarchical and distributed systems. With the hierarchical system, there is a hierarchy of computers according to the tasks they carry out. The computers handling the more routine tasks are supervised by computers which have a greater decision-making role. For example, the computers which are used for direct digital control of systems are subservient to a computer which performs supervisory control of the entire system. The work is divided between the computers according to the function involved. There is specialisation of computers with some computers only receiving some information and others different information. With the distributed system, each computer system carries out essentially similar tasks to all the other computer systems. In the event of a failure of one, or overloading of a particular computer, work can be transferred to other computers. The work is spread across all the computers and not allocated to specific computers according to the function involved. There is no specialisation of computers. Each computer thus needs access to all the information in the system. In most modern systems there is generally a mixture of distributed and hierarchical systems. For example, the work of measurement and actuation may be distributed among a number of microcontrollers/computers which are linked together and provide the database for the plant. These may be overseen by a computer used for direct digital control or sequencing and this in turn may be supervised by one used for supervisory control of the plant as a whole. Typical levels in such a scheme are: Level 1 Measurement and actuators Level 2 Direct digital and sequence control Level 3 Supervisory control Level 4 Management control and design Distributed/hierarchical systems have the advantage of allowing the task of measurement scanning and signal conditioning in control systems to be carried out by sharing it between a number of microprocessors. This can involve a large number of signals with a high frequency of scanning. If extra measurement loops are required, it is a simple matter to increase the capacity of the system by adding microprocessors. The units can be quite widely dispersed, being located near the source of the measurements. Failure of one unit does not result in failure of the entire system.

EParallel and serial data transmission

Parallel data transmission Within computers, data transmission is usually by parallel data paths. Parallel data buses transmit 8, 16 or 32 bits simultaneously, having a separate bus wire for each data bit and the control signals. Thus, if there are eight data bits to be transmitted, e.g. 11000111, then eight data wires are needed. The entire eight data bits are transmitted in the same time as it takes to transmit one data bit because each bit is on a parallel wire. Handshaking (see Section 13.3.2) lines are also needed, handshaking being used for each character transmitted with lines needed to indicate that data is available for transmission and that the receiving terminal is ready to receive. Parallel data transmission permits high data transfer rates but

is expensive because of the cabling and interface circuitry required. It is thus normally only used over short distances or where high transfer rates are essential. 2 Serial data transmission This involves the transmission of data which, together with control signals, is sent bit by bit in sequence along a single line. Only two conductors are needed, to transmit data and to receive data. Since the bits of a word are transmitted sequentially and not simultaneously, the data transfer rate is considerably less than with parallel data transmission. However, it is cheaper since far fewer conductors are required. For example, with a car when a number of microcontrollers are used, the connections between them are by serial data transmission. Without the use of serial transmission the number of wires involved would be considerable. In general, serial data transmission is used for all but the shortest peripheral connections. Consider the problem of sending a sequence of characters along a serial link. The receiver needs to know where one character starts and stops. Serial data transmission can be either asynchronous or synchronous. Asynchronous transmission implies that both the transmitter and receiver computers are not synchronised, each having its own independent clock signals. The time between transmitted characters is arbitrary. Each character transmitted along the link is thus preceded by a start bit to indicate to the receiver the start of a character, and followed by a stop bit to indicate its completion. This method has the disadvantage of requiring extra bits to be transmitted along with each character and thus reduces the efficiency of the line for data transmission. With synchronous transmission there is no need for start and stop bits since the transmitter and receiver have a common clock signal and thus characters automatically start and stop always at the same time in each cycle. The rate of data transmission is measured in bits per second. If a group of n bits form a single symbol being transmitted and the symbol has a duration of T seconds then the data rate of transmission is n/T. The baud is the unit used. The baud rate is only the same as the number of bits per second transmitted if each character is represented by just one symbol. Thus a system which does not use start and stop pulses has a baud rate equal to the bit rate, but this will not be the case when there are such bits.

Network standards

There are a number of network standards, based on the OSI layer model, that are commonly used. The following are examples. In the United States, General Motors realised that the automation of its manufacturing activities posed a problem of equipment being supplied with a variety of non-standard protocols. GM thus developed a standard communication system for factory automation applications. The standard is referred to as the Manufacturing Automation Protocol (MAP) (Figure 15.5). The choice of protocols at the different layers reflects the requirement for the system to fit the manufacturing environment. Layers 1 and 2 are implemented in hardware electronics and levels 3 to 7 using software. For the physical layer, broadband transmission is used. The broadband method allows the system to be used for services in addition to those required for MAP communications. For the data link layer, the token system with a bus is used with logical link control (LLC) to implement such functions as error checking, etc. For the other layers, ISO standards are used. At layer 7, MAP includes manufacturing message services (MMS), an application relevant to factory floor communications which defines interactions between programmable logic controllers and numerically controlled machines or robots.

The Technical and Office Protocol (TOP) is a standard that was developed by Boeing Computer Services. It has much in common with MAP but can be implemented at a lower cost because it is a baseband system. It differs from MAP in layers 1 and 2, using either the token with a ring or the CSMA/CD method with a bus network. Also, at layer 7, it specifies application protocols that concern office requirements, rather than factory floor requirements. With the CSMA/CD method, stations have to listen for other transmissions before transmitting. TOP and MAP networks are compatible and a gateway device can be used to connect TOP and MAP networks. This device carries out the appropriate address conversions and protocol changes. Systems Network Architecture (SNA) is a system developed by IBM as a design standard for IBM products. SNA is divided into seven layers; it, however, differs to some extent from the OSI layers (Figure 15.6). The data link control layer provides support for token ring for LANs. Five of the SNA layers are consolidated in two packages: the path control network for layers 2 and 3 and the network addressable units for layers 4, 5 and 6. With PLC systems, it is quite common for the system used to be that marketed by the PLC manufacturer. For example, Allen-Bradley has the Allen-Bradley data highway which uses token passing to control message transmission; Mitsubishi has Melsec-Net and Texas Instruments has TIWAY. A commonly used system with PLC networks is the Ethernet. This is a single-bus system with CSMA/CD used to control access and is widely used with systems involving PLCs communicating with computers. The problem with using CSMA/CD is that, though this method works well when traffic is light, as network traffic increases the number of collisions and corresponding back-off of transmitters increases. Network throughput can thus slow down quite dramatically.

Aurther

William Bolton

Downlaod Book