Browsing by Author "Guvenc,L."

Now showing 1 - 2 of 2

Citation Count: 5
Communication disturbance observer approach to control of integral plant with time delay
(2013) Emirler,M.T.; Guvenc,B.A.; Guvenc,L.
The presence of time delay can cause stability problems in closed loop systems as it adds large negative phase angle to the system frequency response. The Smith predictor is a well-known method of dealing with fixed and known time delays in control systems. Errors in the knowledge of the time delay will cause degradation of the Smith predictor compensation performance. A solution to this problem is to use a communication disturbance observer. Time delayed integral plants are typical examples of open loop unstable systems with time delay. In this paper, the communication disturbance observer based time delay compensation method is applied to high order time delayed integral plants. A robust stability condition is derived for time delayed feedback control systems with the communication disturbance observer. The effect of the communication disturbance observer Q filter cut-off frequency selection on robust stability is investigated. Simulation results are presented for both constant and time varying delays to illustrate the effectiveness of the proposed communication disturbance observer approach. © 2013 IEEE.
Citation Count: 4
A simulation study of GPS/INS integration for use in ACC/CACC and HAD
(2013) Altay,I.; Guvenc,B.A.; Guvenc,L.
Use of communicated GPS position and velocity information in adaptive cruise control and cooperative adaptive cruise control with the purpose of cooperative driving and the use of GPS position for automated path following in highly automated driving are two current research applications requiring fast and accurate GPS updates. Based on previous experience, a GPS/INS integration system is presented in this paper to allow faster updates as compared to the use of GPS only and to provide accurate position/velocity information in the presence of temporary losses of GPS fix. An INS algorithm and GPS/INS integration are presented in this paper in a realistic simulation setting. The aim is to analyze the effects of different sampling rates of sensors, errors, covariances and to use GPS/INS fusion to broadcast position data at a higher rate than GPS. The DCM method is used to estimate rotations in a inertial measurement unit. Drifts and biases are seen to be the main error sources. As a prerequisite before road testing with our experimental vehicle, GPS/INS is fused using the extended Kalman filter in a highly realistic simulation setting. © 2013 IEEE.