Dear Ankur,

There are many reasons that lead to the development of modern control techniques (may be we can say advanced control methods). Regarding our experience in electric machines control, main reasons where concerned by robustness versus parameters variations and also models uncertainties. Enclosed you will fin some of our papers dealing with issues and using for example (Hinfinity and high-order sliding mode controls).

Hope this could be helpful.

Regards,

Mohamed

Modern Control Theory by William L brogan. The reason is straight forward i.e evolution of computer and digital devices had led to easier way of implementation of controllers that resulted into state space and discrete phenomenon. 

http://as.wiley.com/WileyCDA/WileyTitle/productCd-EHEP001820.html

good book which explains every things which you need related to control.

Absolutely I confirm the words of Mohamed Benbouzid on the question.

Some good reasons:

 - multivariable control: behaviors are coupled between modes (e.g. roll-yaw on a submersible or airplane) and can't be treated independently with decent results

 - digital implementation - frequency domain is still used for stability analyses, but implementation is in the time tomain

Oh,  very many reasons. First, and the most important, classical control theory is not quite suitable in the analysis of multiple-order control systems whereas modern control theory is. It reveals only the system output for a given input and output and provides no information regarding the internal state of the system. Modern control can do this. It only provides information about the present state of the system and tells nothing about the history (previous behaviour) or future behaviour of the system. Modern control can handle all that, and many more. I'm presently writing a book on control engineering that treats this subject.

the book that answers this question quite specifically is

Robert Skelton Dynamics and control of linear systems, by Wliey 1988

Dear Ankur,

I think you got some good answers. In classical control theory, the maintain of certain level of performance when the parameters of the plant are UNKNOWN And/Or CHANGE IN TIME was an inconvenience especially in industry, this is the reason and  not limited to (other reasons like improving robustness, time response, low steady state errors...)  for which the researcher in this field have tried to develop other/Modern control technics to eliminate these inconvenience. Modern/ADAPTIVE control technics are a set of technics for automatic adjustment of the controllers in real time in order to achieve or to maintain a desired performance level of the control system when the parameters of the plant are unknown and/or change in time. It is particularly used in systems with uncertainties, disturbances and changes introduced by the environment.

Best regards,

Khalil

Another one: the integration of control systems and estimators into an integrated framework - allowing simultaneous "optimal" control and estimation.

Thank you everyone for your answers. It really helped me. Furthermore, I would like to ask that which are the advantages of the transfer function over the state space approach? 

Dear Ankur,

      Already you got many useful answers. Hope, my answer may useful to you. I agree with the Mohamed Benbouzid's statement. Nowadays, modern industries combined many systems in a single process for getting maximum productivity. For these complicated systems, classical control theory  doesn't give the guaranteed robustness and performances when uncertainties occurred, also, they  are not generalized for Multi Input Multi Output systems. Thats y modern control theory evolved for these complicated systems to ensure robustness as well to give the generalized methodology for both SISO and MIMO systems.

Best regards

K. Mohaideen abdul kadhar

Dear Ankur,

      You have already  got many useful answers. I hope, my answer will also be in the useful to you. As per my opinion the there are no. of limitations of classical control theory such as it is applicable for only  single Input single  Output systems. Also it is based on transfer function which does not accounts the initial conditions of the system.

Further it is not reliable for complicated systems. In order to overcome these drawbacks modern control theory developed.