Sharif Digital Repository

We propose an observer for attitude and position estimation of a rigid body using translational and angular velocity measurements together with a single landmark. Under an observability condition depending on landmark position, global and asymptotic convergence of attitude and position estimation errors to zero is achieved. We extend the observer for the case of nonideal velocity measurements. Simulation examples demonstrate convergence properties of the observer in the face of noise and bias in velocity measurements  

Time delay in attitude sensors is one of the performance limiting factors in controlling the satellite attitude. In this paper, we investigate attitude control of a fully actuated satellite in presence of constant and known delay in attitude measurement. The proposed controller consists of a matrix gain which is computed by solving a time varying matrix differential equation depending on the time delay. We assume that the moment-of-inertia matrix of the satellite is unknown. The controller guarantees asymptotic convergence of the satellite attitude and angular velocity to their desired values in presence of delay. Finally, a simulation example is presented that illustrates performance of the... 

Neglecting actuator dynamics in control of rigid manipulators can degrade performance and stability of the system. However, consideration of actuator dynamics usually requires measurement of joint torque or armature currents. In this paper, we present an adaptive controller for motion tracking of an n-DOF manipulator without using joint torque measurement. Semi-global convergence of motion and torque tracking errors to zero is proven. We show that the use of non-minimal representation of manipulator dynamics significantly simplifies its linear parametrization. Simulation example shows that avoiding joint torque measurement reduces system sensitivity to sensor failure and noise  

This paper studies the leader-follower flocking of multi-agent systems for the linear second-order dynamics, subject to the external disturbance problem. It is assumed that the dynamic of the leader is Lipschitz-type. Also, the velocity is the output of the system, and full-state information is not available for feedback. A distributed full-order observer is employed to estimate every agent's states and external disturbance. A control protocol for each agent is designed based on the measurement of its output/velocity and relative velocity of its neighbors. Under the proposed protocol, the velocity convergence of whole agents to the velocity of the virtual leader is guaranteed as well as the... 

We present a geometric approach for fault detection and isolation (FDI) in robotic manipulators in presence of model uncertainty. A systematic procedure is introduced for representing robotic system model being affine with respect to faults and disturbances. The proposed residual generator has smooth dynamics with freely selectable functions and it does not require high gains or threshold adjustment for the FDI purpose. No assumption on amplitude of faults and their rate is used. The solvability conditions for the FDI problem lead to a quotient observable subspace unaffected by all unknown inputs except the faults. Simulation example demonstrates localization of faults in presence of... 

In this paper, we introduce a new approach to fault detection and isolation for robot manipulators. Our technique is based on using a new simplified Euler-Lagrange (EL) equation that reduces complexity of the proposed fault detection method. The proposed approach isolates the faults and is capable of handling the uncertainty in manipulator gravity vector. It is shown that the effect of uncalibrated torque sensor measurement is asymptotically rejected in the detection process. A simulation example is presented to illustrate the results. © 2009 IEEE  

Ignoring actuator dynamics in control of rigid manipulators can in practice result in performance degradation or loss of system stability. However, consideration of actuator dynamics usually requires measurement of robot joint torques. This paper addresses motion tracking control of an n-DOF rigid robot by taking into account its actuator dynamics. Joint torque measurement is avoided by using an adaptive observer. The backstepping technique is adopted to develop a dynamically smooth adaptive nonlinear controller dealing with uncertainties in manipulator and actuator dynamics. Semi-global convergence of motion tracking errors as well as torque estimation error are proven without any... 

Time-delay in sensor measurements can be a frequent cause of instability and performance degradation in a robotic system. In this paper, motion tracking of rigid manipulators in presence of constant and known delay in sensors is investigated. By using non-minimal model of a manipulator, a dynamically smooth controller based on the Linear Matrix Inequality (LMI) approach is proposed which guarantees asymptotic tracking of desired joint angles and velocities in presence of delayed measurements. For a given controller the maximum amount of delay that preserves system stability is computed by solving an LMI optimization and also by numerical simulations, and the results are compared. Finally, a... 

In this paper, a decentralized model predictive control approach is proposed for discrete linear systems with a high number of inputs and states. The system is decomposed into several interacting subsystems. The interaction among subsystems is modeled as external disturbances. Then, using the concept of robust positively invariant ellipsoids, a robust model predictive control law is obtained for each subsystem solving several linear matrix inequalities. Maintaining the recursive feasibility while considering the attenuation of mutual coupling at each time step and the stability of the overall system are investigated. Moreover, an illustrative simulation example is provided to demonstrate the... 

We present a method for global estimation of joint velocities in rigid manipulators. A class of velocity observers with smooth dynamics are introduced whose structure depend on freely selectable functions and gains giving the flexibility of achieving multiple design objectives at the same time. Unlike most methods, no a priori knowledge of an upper bound for velocity magnitude is used. An adaptive version of the observer is also presented to handle a class of structured uncertainties in manipulator model. Simulation example illustrates low noise sensitivity of the globally convergent observer in comparison with semi-globally convergent observers. © 2009 IEEE