As the main consumer of the electrical energy the electromechanical systems attract attention of many researchers. Especially in the field of the controlled systems where the increased efficiency means less consumed power and consequently less environment pollution with improved competitive position of factory. In the frame of the research work the new nonlinear control algorithm for direct drive servosytems with AC motors has been derived where the control function consists of the feedword and feedback part. A new procedure for the determination of the stability of the controlled nonlinear systems based on the Lyapunov function has been performed. The application of the continuous sliding mode control law for the laboratory set up of the direct drive robot ensures the linear behavior of the closed control loop system with the reduce chattering of the control signal. The implementation of the position and force control of the robot based on the external linearization is a demanding task and beside that also very sensitive to the perturbation of the robot dynamic model parameters. A novel robust impedance control algortihm based on the PI estimator where the integral part is built into the inner control loop has been developed in order to overcome above mentioned difficulties. Its stability in the mode of the robot free motion, the mode of the robot contact with the enviroment as weel as in the transition between those modes has been proved by Lyapunov method. The developed position control algorithm with PI disturbance estimator has been tested on the XY table where friction, elasticity and compliance of the belt-drives and the gearboxes significantly impact the motion dynamics of the servodrives. We have added a differentation into the PI disturbance estimator signal and with special attention to signal/noise ratio we could extended position closed-loop frequency bandwidth. Furthermore, another research has involved fuzzy logic algorithms in order to improve PI disturbance estimator. This approach featured even better signal/noise ratio. The sensorless control of the AC motor drive gains more and more attention. The control problems in the frequency range of 0-2 Hz with the precisely formed torque to assure the smooth mechanism motion remain unsolved on the international level. The low speed range including zero speed is one of the most important research area for the last few years. The improved robustness and the lowest mechanism price will increase the use of AC motor drive in mechatronics application. The result of the research work is a speed sensorless control based on a new observer of the induced voltage and rotor flux. A new DSP2 controller based on a signal processor TMS320C32-60 that enables the execution of the floating-point algorithms has been built. It has all the necessary peripherial units for the implementation of motor drive control. The important part of the research work was computer aided design (CAD) of the real time control system using MATLAB/Simulink software. The fusion of the CAD phase with the real-time implementation increases the productivity in the design and making of embeded systems and computer automation products. Based on the signal processor Texas Instruments C31 a new software compiling tool has been developed. It generates C code automatically from Simulink to be executed on the target processor and is used for the control of induction motor.