Discontinuous feedback stabilization of nonholonomic wheeled mobile robots (Q676058)

Feedback control of a class of mobile robots moving on a two dimensional plane is considered. The robot is modelled by the equations \[ \begin{aligned} & \dot x=\nu \cos \theta,\\ & \dot y = \nu\sin \theta,\\ & \dot \theta=\omega,\end{aligned} \] where \(x\), \(y\) denote the Cartesian coordinates of the robot in the \(x\)-\(y\) plane and \(\theta\) denotes its angular orientation with respect to the \(x\)-axis. The control variables \(\nu\) and \(\omega\) are the tangent and angular velocities, respectively. The control objective of the paper is to design feedback loops so that the mobile robot can be automatically driven from an arbitrary initial configuration \((x_0,y_0,\theta_0)^T\) in \(R^2\times [-\pi,\pi]\) to a given target configuration in such a way that the deviation between \((x(t),y(t),\theta(t))^T\) and the target configuration gets smaller as \((x_0,y_0,\theta_0)^T\) gets closer to the target configuration. For simplicity, it is assumed that the target configuration coincides with the origin of the state space. A new discontinuous feedback control strategy which achieves stabilization of the robot and a smooth trajectory of the path is proposed. The main idea is based on the following observations: the mobile robot can achieve a pure rotation by letting \(\nu=0\) and for any point \((x_0,y_0)\) in the \(x\)-\(y\) plane, there is a circle which passes both \((x_0,y_0)\) and the origin and is centered on the \(y\)-axis. Thus, the feedback stabilization can be solved by first taking a pure rotation until the orientation angle is almost tangent to the circle mentioned above at the initial Cartesian position, and then driving the robot asymptotically to the origin along the circle while controlled to make the tangent of the path approach the tangent of the circle. This may result in high-frequency switching in the control law, which can be easily avoided by appropriately choosing the control parameters.