Swinging control of nonlinear oscillations. (English) Zbl 0866.93049

The equation of a controlled plant is: \(dx/d=F(x,u,t)\), \(t\geq 0\), where \(u\) denotes the control. An equivalent Hamiltonian system is defined in terms of phase coordinates \(q\) and \(p\). \(\dot p=\partial H(p,q,u)/\partial q\), \(\dot q=-\partial H(p,q,u)/\partial p\). Observe that \(q=x\) is not necessarily true. The Hamiltonian \(H\) is linear in \(u\). Let \(H_0\) denote the Hamiltonian for the uncontrolled system, i.e. \(H_0=H(p,q,0)\). The energy of the uncontrolled system is denoted by \(S=\{(p,q): H=H_0\}\). Because of linear dependence on \(u\), we have: \(H(p,q,u)=H_0+H_1(p,q)u\). The objective demand is: \(H_0(p,q)\rightarrow H*\) as \(t\rightarrow \infty\). Let us regard the functional \(Q={1\over 2}[H_0(p,q)-H*]^2\) as the cost functional. A calculation shows that \(\dot Q=(H_0(p,q)-H*)[H_0,H_1]\), where \([,]\) denotes the Poisson bracket. To design the speed gradient algorithm, the author computes the speed \(\omega\) of change of \(Q\) along a trajectory of the system. Subject to assumptions on smoothness and convexity of \(\omega(x,u,t)\) he also proves boundedness of the cost function along any trajectory. If \(\omega(x,u,t)\) is bounded from above by a positive, continuous function, then \(\lim_{t\to \infty} \nabla_u\omega(x,u,t)=0\) is attained when \(u\) is the optimal control.
As an example of application, the author computes the optimal control of a simple two-dimensional pendulum.
Reviewer: V.Komkov (Roswell)


93C15 Control/observation systems governed by ordinary differential equations
93C10 Nonlinear systems in control theory
70H05 Hamilton’s equations
Full Text: DOI


[1] AKULENKO L., Applied Mathematics and Mechanics 57 pp 82– (1993)
[2] ANDRIEVSKY B. R., Automation and Remote Control 4 pp 4– (1996)
[3] CHERNOUS’KO , F. , AKULENKO , L. , and SOKOLOV , B. , 1980 , The Control of Oscillations ( Moscow : Nauka ) , in Russian .
[4] FOMIN V. N., Adaptive Control of Dynamic Objects (1981)
[5] FRADKOV A. L., Automation and Remote Control (1979)
[6] FRADKOV A. L., 3rd IFAC Symposium on Nonlinear Control Systems 2 pp 655– (1995)
[7] GUCKENHEIMER J., Nonlinear Oscillations, Dynamical Systems and Bifurcations of Vector Fields (1983) · Zbl 0515.34001
[8] MORI S., International Journal of Control 23 pp 673– (1976)
[9] VAN DER SCHAFT A., Three Decades of Mathematical Systems Theory 135 pp 426– (1990) · Zbl 0714.93022
[10] WIKLUND M., Preprints of the 12th IFAC World Congress 9 pp 151– (1993)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.