We present a time-varying global output-feedback controller that solves both tracking and stabilization for unicycle-type mobile robots simultaneously at the torque level. The controller synthesis is based on a coordinate transformation, Lyapunov's direct method, and backstepping technique. Simulations demonstrate the result.
A constructive method is presented to design controllers that force the output of nonlinear systems in a strict feedback form to track a bounded and sufficiently smooth reference trajectory asymptotically. Under a suitable condition on the initial output tracking error, the proposed controllers guarantee the output tracking error within a symmetric or an asymmetric pre-specified limit range, and boundedness of all signals of the closed-loop system. A transformation is introduced to take care of the output tracking error constraint. Smooth and/or p-times differentiable step functions are proposed and incorporated in the output tracking error transformation to overcome difficulties due to the asymmetric limit range on the output tracking error. As a result, there are no switchings in the proposed controllers despite the asymmetric limit range. The proposed control design is then applied to design a tracking controller for active magnetic bearings as an illustrating application.
1 September 2011 in Journal of Marine Science and Application
In existing flocking and coordination control systems of multiple agents, an agent is considered as a single point or a circular disk. For agents with a long and narrow shape, fitting them to circular disks results in a problem with the large conservative area. This paper presents a design of distributed controllers for flocking of N mobile agents with an elliptical shape and with limited communication ranges. A separation condition for elliptical agents is first derived. Smooth or p-times differentiable step functions are then introduced. These functions and the separation condition between the elliptical agents are embedded in novel potential functions to design control algorithms for flocking. The controllers guarantee 1) no switchings in the controllers despite the agents' limited communication ranges; 2) no collisions between any agents; 3) asymptotic convergence of each agent's generalized velocity to a desired velocity; and 4) boundedness of the flock size, which is defined as the sum of all generalized distances between the agents, by a constant.