scientific article; zbMATH DE number 1463359

From MaRDI portal
Publication:4487437





The paper deals with the stabilization and guidance of a riderless bicycle. It is assumed here that the bicycle is controlled by a pedalling torque, a directional torque and a rotor mounted on the crossbar that generates a tilting torque. Denote, by \((x_C, y_C)\) the coordinates, in an internal coordinate system, of the point of contact between the bicycle's rear wheel and the ground. Denote, for \(t\geq 0\) \[ \eta(t)= (x_C(t),\dot x_C(t), \ddot x_C(t), y_C(t), \dot y_C(t), \ddot y_C(t))^T. \] If for any two points \(Q_1\) and \(Q_2\) in \(\mathbb{R}^6\), and for any finite interval \([0, t_f)\), control laws for the pedalling torque, for the directional torque and for the mounted rotor's input torque can be found such that \(\eta(.)\) will move from \(\eta(0)= Q_1\) to \(\eta(t_f)= Q_2\), then it is said that the bicycle is path controllable. By using a kind of inverse dynamic control, control laws are derived under which the motion of the bicycle is stabilized and the rear wheel rolls from \(P_1(x_{C1}, y_{C1})\) to \(P_2(x_{C2}, y_{C2})\) during a given time interval \([0, t_f]\). The procedure described here for constructing control laws is straightforward and circumvents dealing with the control Lie algebra.




Cited in
(20)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4487437)