ERCIM News No.40 - January 2000

Vladimir KuceraAutomatic Control: Past, Present and Future

Automatic control is a driver of technology progress and society industrialization. Control systems have a truly impressive legacy of innovation. Control engineering has made possible space travel and communication satellites, has assisted in the design of safe and efficient aircraft, ships, trains and cars, has helped in developing cleaner chemical processes while addressing environmental concerns. Automatic control constantly improves the quality of human life.

Control theory and engineering have witnessed dramatic achievements throughout this century. Recall the stability theory of Lyapunov from the beginning of the century, the conception of three-term or PID controllers in the 1910s, electronic and pneumatic feedback amplifiers in the 1920s, Nyquist and Bode charts of the 1930s, and Wiener’s cybernetics of the 1940s. Then came the 1950s and Bellman’s principle of optimality, Kalman filter of the 1960s, adaptive control in the 1970s, robust control in the 1980s, and the hybrid control systems of the current decade.

The milestones of this development were the introduction of negative feedback amplifiers, field adjustable PID controllers, and especially digital computers. These technologies have had a tremendous impact on control theory and its application. The origin of classical control theory dates back to the conception of negative feedback and the subsequent development of frequency domain techniques, while modern control theory coincides with the introduction of state space methods related to the use of computers in space applications.

Today, as a result of this evolution, it is possible to implement advanced control methodologies. We have smart sensors and smart actuators. The most dramatic impact of electronic processing occurs in controllers. In times past, computational demands of adaptive, optimal and robust control techniques could not be easily performed. With modern electronics, such operations are possible. Modern electronic implementations are also more immune to aging effects, system noise and disturbances.

Control theory, on the other hand, is looking for new solutions. There is a strong influence of computer science and engineering. Analytic methods are giving way to synthetic ones. A search for closed-form solutions was typical in the classical era. The modern control theory considers a problem solved when it is reduced to an equation; an algorithmic solution is left to a computer. This trend will continue. The control of complex systems, or systems of systems, will be formulated as a mathematical program. Much of the analysis will be replaced by interactive, computer-aided design procedures. The solution will have to address the issues of hierarchy, interaction and the possibility of system evolution.

The truly exciting developments in automatic control will occur where there is a confluence of application drivers and disciplinary development of the subject. Changes in control education, adjustments in research directions, and more emphasis on implementation may provide the foundations and tools to meet the challenge of the next century and keep automatic control a dynamic and fascinating field.


return to the ERCIM News 40 contents page